{
  "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": 1.1260029673576355,
      "epoch": 0.0037313432835820895,
      "grad_norm": 1.6953052282333374,
      "learning_rate": 0.0002,
      "loss": 2.460799217224121,
      "mean_token_accuracy": 0.5385793447494507,
      "num_tokens": 16458.0,
      "step": 1
    },
    {
      "entropy": 1.2484523057937622,
      "epoch": 0.007462686567164179,
      "grad_norm": 1.5863436460494995,
      "learning_rate": 0.0002,
      "loss": 2.1654767990112305,
      "mean_token_accuracy": 0.5669277608394623,
      "num_tokens": 32789.0,
      "step": 2
    },
    {
      "entropy": 1.4055505394935608,
      "epoch": 0.011194029850746268,
      "grad_norm": 1.1718486547470093,
      "learning_rate": 0.0002,
      "loss": 1.714382290840149,
      "mean_token_accuracy": 0.5935665369033813,
      "num_tokens": 49349.0,
      "step": 3
    },
    {
      "entropy": 1.3796880543231964,
      "epoch": 0.014925373134328358,
      "grad_norm": 0.9294341802597046,
      "learning_rate": 0.0002,
      "loss": 1.3914273977279663,
      "mean_token_accuracy": 0.6369634121656418,
      "num_tokens": 65829.0,
      "step": 4
    },
    {
      "entropy": 1.3194991052150726,
      "epoch": 0.018656716417910446,
      "grad_norm": 1.0537753105163574,
      "learning_rate": 0.0002,
      "loss": 1.2675070762634277,
      "mean_token_accuracy": 0.6544732749462128,
      "num_tokens": 82394.0,
      "step": 5
    },
    {
      "entropy": 1.2329450845718384,
      "epoch": 0.022388059701492536,
      "grad_norm": 0.5125759840011597,
      "learning_rate": 0.0002,
      "loss": 1.1630359888076782,
      "mean_token_accuracy": 0.6683834344148636,
      "num_tokens": 98903.0,
      "step": 6
    },
    {
      "entropy": 1.1502311825752258,
      "epoch": 0.026119402985074626,
      "grad_norm": 0.39078545570373535,
      "learning_rate": 0.0002,
      "loss": 1.0693285465240479,
      "mean_token_accuracy": 0.675154522061348,
      "num_tokens": 115247.0,
      "step": 7
    },
    {
      "entropy": 1.0659821331501007,
      "epoch": 0.029850746268656716,
      "grad_norm": 0.4620763957500458,
      "learning_rate": 0.0002,
      "loss": 1.0068438053131104,
      "mean_token_accuracy": 0.6829047352075577,
      "num_tokens": 131731.0,
      "step": 8
    },
    {
      "entropy": 1.0000898838043213,
      "epoch": 0.033582089552238806,
      "grad_norm": 0.5140990614891052,
      "learning_rate": 0.0002,
      "loss": 0.9476600885391235,
      "mean_token_accuracy": 0.6949260085821152,
      "num_tokens": 148350.0,
      "step": 9
    },
    {
      "entropy": 0.9560111612081528,
      "epoch": 0.03731343283582089,
      "grad_norm": 0.497260719537735,
      "learning_rate": 0.0002,
      "loss": 0.8725440502166748,
      "mean_token_accuracy": 0.710997924208641,
      "num_tokens": 164467.0,
      "step": 10
    },
    {
      "entropy": 0.9134543389081955,
      "epoch": 0.041044776119402986,
      "grad_norm": 0.45270034670829773,
      "learning_rate": 0.0002,
      "loss": 0.8241041898727417,
      "mean_token_accuracy": 0.7159350216388702,
      "num_tokens": 180927.0,
      "step": 11
    },
    {
      "entropy": 0.798617348074913,
      "epoch": 0.04477611940298507,
      "grad_norm": 0.48476189374923706,
      "learning_rate": 0.0002,
      "loss": 0.770041286945343,
      "mean_token_accuracy": 0.7293703109025955,
      "num_tokens": 197219.0,
      "step": 12
    },
    {
      "entropy": 0.7840605825185776,
      "epoch": 0.048507462686567165,
      "grad_norm": 0.40834739804267883,
      "learning_rate": 0.0002,
      "loss": 0.7530187368392944,
      "mean_token_accuracy": 0.7250247299671173,
      "num_tokens": 213644.0,
      "step": 13
    },
    {
      "entropy": 0.7022175937891006,
      "epoch": 0.05223880597014925,
      "grad_norm": 0.40251079201698303,
      "learning_rate": 0.0002,
      "loss": 0.7203142046928406,
      "mean_token_accuracy": 0.734594002366066,
      "num_tokens": 229979.0,
      "step": 14
    },
    {
      "entropy": 0.7001681327819824,
      "epoch": 0.055970149253731345,
      "grad_norm": 0.35986149311065674,
      "learning_rate": 0.0002,
      "loss": 0.7040289044380188,
      "mean_token_accuracy": 0.7360049784183502,
      "num_tokens": 246473.0,
      "step": 15
    },
    {
      "entropy": 0.6849737167358398,
      "epoch": 0.05970149253731343,
      "grad_norm": 0.33282366394996643,
      "learning_rate": 0.0002,
      "loss": 0.6554020047187805,
      "mean_token_accuracy": 0.7522932589054108,
      "num_tokens": 262988.0,
      "step": 16
    },
    {
      "entropy": 0.6813573390245438,
      "epoch": 0.06343283582089553,
      "grad_norm": 0.3554367423057556,
      "learning_rate": 0.0002,
      "loss": 0.6626776456832886,
      "mean_token_accuracy": 0.7477623820304871,
      "num_tokens": 279264.0,
      "step": 17
    },
    {
      "entropy": 0.6491282731294632,
      "epoch": 0.06716417910447761,
      "grad_norm": 0.312103271484375,
      "learning_rate": 0.0002,
      "loss": 0.6249456405639648,
      "mean_token_accuracy": 0.7609707713127136,
      "num_tokens": 295556.0,
      "step": 18
    },
    {
      "entropy": 0.6441417187452316,
      "epoch": 0.0708955223880597,
      "grad_norm": 0.33169475197792053,
      "learning_rate": 0.0002,
      "loss": 0.6275812387466431,
      "mean_token_accuracy": 0.755386084318161,
      "num_tokens": 311911.0,
      "step": 19
    },
    {
      "entropy": 0.63529072701931,
      "epoch": 0.07462686567164178,
      "grad_norm": 0.30479201674461365,
      "learning_rate": 0.0002,
      "loss": 0.6244862079620361,
      "mean_token_accuracy": 0.757283627986908,
      "num_tokens": 328411.0,
      "step": 20
    },
    {
      "entropy": 0.5928637981414795,
      "epoch": 0.07835820895522388,
      "grad_norm": 0.339279443025589,
      "learning_rate": 0.0002,
      "loss": 0.6037462949752808,
      "mean_token_accuracy": 0.7652928382158279,
      "num_tokens": 344891.0,
      "step": 21
    },
    {
      "entropy": 0.5829604864120483,
      "epoch": 0.08208955223880597,
      "grad_norm": 0.27806463837623596,
      "learning_rate": 0.0002,
      "loss": 0.5940107107162476,
      "mean_token_accuracy": 0.7702941447496414,
      "num_tokens": 361235.0,
      "step": 22
    },
    {
      "entropy": 0.5957961678504944,
      "epoch": 0.08582089552238806,
      "grad_norm": 0.24885332584381104,
      "learning_rate": 0.0002,
      "loss": 0.5948885679244995,
      "mean_token_accuracy": 0.7657869160175323,
      "num_tokens": 377384.0,
      "step": 23
    },
    {
      "entropy": 0.6124956607818604,
      "epoch": 0.08955223880597014,
      "grad_norm": 0.25812840461730957,
      "learning_rate": 0.0002,
      "loss": 0.6049087047576904,
      "mean_token_accuracy": 0.7600619345903397,
      "num_tokens": 393591.0,
      "step": 24
    },
    {
      "entropy": 0.6115936785936356,
      "epoch": 0.09328358208955224,
      "grad_norm": 0.24209237098693848,
      "learning_rate": 0.0002,
      "loss": 0.5928857326507568,
      "mean_token_accuracy": 0.7660015225410461,
      "num_tokens": 410275.0,
      "step": 25
    },
    {
      "entropy": 0.6117522269487381,
      "epoch": 0.09701492537313433,
      "grad_norm": 0.2623499631881714,
      "learning_rate": 0.0002,
      "loss": 0.6010225415229797,
      "mean_token_accuracy": 0.7596103847026825,
      "num_tokens": 426343.0,
      "step": 26
    },
    {
      "entropy": 0.5846991240978241,
      "epoch": 0.10074626865671642,
      "grad_norm": 0.255848228931427,
      "learning_rate": 0.0002,
      "loss": 0.5731462836265564,
      "mean_token_accuracy": 0.7730763554573059,
      "num_tokens": 442529.0,
      "step": 27
    },
    {
      "entropy": 0.5835026651620865,
      "epoch": 0.1044776119402985,
      "grad_norm": 0.26115483045578003,
      "learning_rate": 0.0002,
      "loss": 0.5819462537765503,
      "mean_token_accuracy": 0.7699420154094696,
      "num_tokens": 458911.0,
      "step": 28
    },
    {
      "entropy": 0.5770604312419891,
      "epoch": 0.10820895522388059,
      "grad_norm": 0.21723432838916779,
      "learning_rate": 0.0002,
      "loss": 0.5792397856712341,
      "mean_token_accuracy": 0.7707104980945587,
      "num_tokens": 475390.0,
      "step": 29
    },
    {
      "entropy": 0.5611739307641983,
      "epoch": 0.11194029850746269,
      "grad_norm": 0.22502686083316803,
      "learning_rate": 0.0002,
      "loss": 0.5769323706626892,
      "mean_token_accuracy": 0.7717459946870804,
      "num_tokens": 491664.0,
      "step": 30
    },
    {
      "entropy": 0.5530466288328171,
      "epoch": 0.11567164179104478,
      "grad_norm": 0.24965344369411469,
      "learning_rate": 0.0002,
      "loss": 0.5644280910491943,
      "mean_token_accuracy": 0.7790123075246811,
      "num_tokens": 508002.0,
      "step": 31
    },
    {
      "entropy": 0.5743643939495087,
      "epoch": 0.11940298507462686,
      "grad_norm": 0.23363712430000305,
      "learning_rate": 0.0002,
      "loss": 0.5805435180664062,
      "mean_token_accuracy": 0.7706304490566254,
      "num_tokens": 524313.0,
      "step": 32
    },
    {
      "entropy": 0.5664798021316528,
      "epoch": 0.12313432835820895,
      "grad_norm": 0.2516216039657593,
      "learning_rate": 0.0002,
      "loss": 0.5730225443840027,
      "mean_token_accuracy": 0.7700050920248032,
      "num_tokens": 540778.0,
      "step": 33
    },
    {
      "entropy": 0.5729327946901321,
      "epoch": 0.12686567164179105,
      "grad_norm": 0.19088800251483917,
      "learning_rate": 0.0002,
      "loss": 0.570796012878418,
      "mean_token_accuracy": 0.7728039622306824,
      "num_tokens": 557177.0,
      "step": 34
    },
    {
      "entropy": 0.5797266364097595,
      "epoch": 0.13059701492537312,
      "grad_norm": 0.23257242143154144,
      "learning_rate": 0.0002,
      "loss": 0.5671622157096863,
      "mean_token_accuracy": 0.7741479426622391,
      "num_tokens": 573749.0,
      "step": 35
    },
    {
      "entropy": 0.571784034371376,
      "epoch": 0.13432835820895522,
      "grad_norm": 0.20902124047279358,
      "learning_rate": 0.0002,
      "loss": 0.5635585784912109,
      "mean_token_accuracy": 0.776234894990921,
      "num_tokens": 590172.0,
      "step": 36
    },
    {
      "entropy": 0.5775258988142014,
      "epoch": 0.13805970149253732,
      "grad_norm": 0.21658626198768616,
      "learning_rate": 0.0002,
      "loss": 0.577404797077179,
      "mean_token_accuracy": 0.7704328000545502,
      "num_tokens": 606865.0,
      "step": 37
    },
    {
      "entropy": 0.5636124461889267,
      "epoch": 0.1417910447761194,
      "grad_norm": 0.181837797164917,
      "learning_rate": 0.0002,
      "loss": 0.5649010539054871,
      "mean_token_accuracy": 0.774622842669487,
      "num_tokens": 623186.0,
      "step": 38
    },
    {
      "entropy": 0.5586818605661392,
      "epoch": 0.1455223880597015,
      "grad_norm": 0.19703401625156403,
      "learning_rate": 0.0002,
      "loss": 0.5703071355819702,
      "mean_token_accuracy": 0.7706339806318283,
      "num_tokens": 639625.0,
      "step": 39
    },
    {
      "entropy": 0.5555090457201004,
      "epoch": 0.14925373134328357,
      "grad_norm": 0.16290432214736938,
      "learning_rate": 0.0002,
      "loss": 0.5560703277587891,
      "mean_token_accuracy": 0.7768503576517105,
      "num_tokens": 656135.0,
      "step": 40
    },
    {
      "entropy": 0.5701133161783218,
      "epoch": 0.15298507462686567,
      "grad_norm": 0.19164100289344788,
      "learning_rate": 0.0002,
      "loss": 0.576887309551239,
      "mean_token_accuracy": 0.7674428522586823,
      "num_tokens": 672233.0,
      "step": 41
    },
    {
      "entropy": 0.5744345933198929,
      "epoch": 0.15671641791044777,
      "grad_norm": 0.18373753130435944,
      "learning_rate": 0.0002,
      "loss": 0.5698904991149902,
      "mean_token_accuracy": 0.7698992490768433,
      "num_tokens": 688517.0,
      "step": 42
    },
    {
      "entropy": 0.5747536867856979,
      "epoch": 0.16044776119402984,
      "grad_norm": 0.17863287031650543,
      "learning_rate": 0.0002,
      "loss": 0.5794167518615723,
      "mean_token_accuracy": 0.7684431076049805,
      "num_tokens": 704885.0,
      "step": 43
    },
    {
      "entropy": 0.5701879560947418,
      "epoch": 0.16417910447761194,
      "grad_norm": 0.16616570949554443,
      "learning_rate": 0.0002,
      "loss": 0.5643965005874634,
      "mean_token_accuracy": 0.7729478478431702,
      "num_tokens": 721192.0,
      "step": 44
    },
    {
      "entropy": 0.559878334403038,
      "epoch": 0.16791044776119404,
      "grad_norm": 0.14239990711212158,
      "learning_rate": 0.0002,
      "loss": 0.5554589033126831,
      "mean_token_accuracy": 0.7787265032529831,
      "num_tokens": 737493.0,
      "step": 45
    },
    {
      "entropy": 0.5644266754388809,
      "epoch": 0.17164179104477612,
      "grad_norm": 0.17384031414985657,
      "learning_rate": 0.0002,
      "loss": 0.5654000639915466,
      "mean_token_accuracy": 0.7739932388067245,
      "num_tokens": 753949.0,
      "step": 46
    },
    {
      "entropy": 0.5645141154527664,
      "epoch": 0.17537313432835822,
      "grad_norm": 0.17348957061767578,
      "learning_rate": 0.0002,
      "loss": 0.5617217421531677,
      "mean_token_accuracy": 0.7774634659290314,
      "num_tokens": 770544.0,
      "step": 47
    },
    {
      "entropy": 0.5625235587358475,
      "epoch": 0.1791044776119403,
      "grad_norm": 0.14307668805122375,
      "learning_rate": 0.0002,
      "loss": 0.560384213924408,
      "mean_token_accuracy": 0.7760685384273529,
      "num_tokens": 787084.0,
      "step": 48
    },
    {
      "entropy": 0.555452972650528,
      "epoch": 0.1828358208955224,
      "grad_norm": 0.16662120819091797,
      "learning_rate": 0.0002,
      "loss": 0.5601282715797424,
      "mean_token_accuracy": 0.7779025435447693,
      "num_tokens": 803757.0,
      "step": 49
    },
    {
      "entropy": 0.557629257440567,
      "epoch": 0.1865671641791045,
      "grad_norm": 0.15931881964206696,
      "learning_rate": 0.0002,
      "loss": 0.5621690154075623,
      "mean_token_accuracy": 0.7754093408584595,
      "num_tokens": 820259.0,
      "step": 50
    },
    {
      "entropy": 0.5400655269622803,
      "epoch": 0.19029850746268656,
      "grad_norm": 0.16219788789749146,
      "learning_rate": 0.0002,
      "loss": 0.5453745722770691,
      "mean_token_accuracy": 0.7794964462518692,
      "num_tokens": 836165.0,
      "step": 51
    },
    {
      "entropy": 0.567253589630127,
      "epoch": 0.19402985074626866,
      "grad_norm": 0.1521536111831665,
      "learning_rate": 0.0002,
      "loss": 0.5732303857803345,
      "mean_token_accuracy": 0.7715951949357986,
      "num_tokens": 852770.0,
      "step": 52
    },
    {
      "entropy": 0.5600331574678421,
      "epoch": 0.19776119402985073,
      "grad_norm": 0.18043731153011322,
      "learning_rate": 0.0002,
      "loss": 0.5502007007598877,
      "mean_token_accuracy": 0.7802667319774628,
      "num_tokens": 869142.0,
      "step": 53
    },
    {
      "entropy": 0.5579064935445786,
      "epoch": 0.20149253731343283,
      "grad_norm": 0.14194370806217194,
      "learning_rate": 0.0002,
      "loss": 0.5532926321029663,
      "mean_token_accuracy": 0.7787930965423584,
      "num_tokens": 885527.0,
      "step": 54
    },
    {
      "entropy": 0.5481145083904266,
      "epoch": 0.20522388059701493,
      "grad_norm": 0.18247617781162262,
      "learning_rate": 0.0002,
      "loss": 0.5540817975997925,
      "mean_token_accuracy": 0.774936631321907,
      "num_tokens": 901951.0,
      "step": 55
    },
    {
      "entropy": 0.5435811579227448,
      "epoch": 0.208955223880597,
      "grad_norm": 0.14611691236495972,
      "learning_rate": 0.0002,
      "loss": 0.5460960865020752,
      "mean_token_accuracy": 0.7804523259401321,
      "num_tokens": 918549.0,
      "step": 56
    },
    {
      "entropy": 0.5274965688586235,
      "epoch": 0.2126865671641791,
      "grad_norm": 0.16689632833003998,
      "learning_rate": 0.0002,
      "loss": 0.5317626595497131,
      "mean_token_accuracy": 0.7861348390579224,
      "num_tokens": 934568.0,
      "step": 57
    },
    {
      "entropy": 0.5469426512718201,
      "epoch": 0.21641791044776118,
      "grad_norm": 0.17422115802764893,
      "learning_rate": 0.0002,
      "loss": 0.5483155846595764,
      "mean_token_accuracy": 0.7792017906904221,
      "num_tokens": 950845.0,
      "step": 58
    },
    {
      "entropy": 0.5521796494722366,
      "epoch": 0.22014925373134328,
      "grad_norm": 0.15713948011398315,
      "learning_rate": 0.0002,
      "loss": 0.5532113313674927,
      "mean_token_accuracy": 0.7776449769735336,
      "num_tokens": 967389.0,
      "step": 59
    },
    {
      "entropy": 0.5459235310554504,
      "epoch": 0.22388059701492538,
      "grad_norm": 0.159347265958786,
      "learning_rate": 0.0002,
      "loss": 0.5439324378967285,
      "mean_token_accuracy": 0.7833367586135864,
      "num_tokens": 983960.0,
      "step": 60
    },
    {
      "entropy": 0.5611889511346817,
      "epoch": 0.22761194029850745,
      "grad_norm": 0.19656279683113098,
      "learning_rate": 0.0002,
      "loss": 0.5588216185569763,
      "mean_token_accuracy": 0.7787718325853348,
      "num_tokens": 1000264.0,
      "step": 61
    },
    {
      "entropy": 0.5443999469280243,
      "epoch": 0.23134328358208955,
      "grad_norm": 0.16265363991260529,
      "learning_rate": 0.0002,
      "loss": 0.5430524945259094,
      "mean_token_accuracy": 0.783473938703537,
      "num_tokens": 1016591.0,
      "step": 62
    },
    {
      "entropy": 0.5739233940839767,
      "epoch": 0.23507462686567165,
      "grad_norm": 0.16212719678878784,
      "learning_rate": 0.0002,
      "loss": 0.5744223594665527,
      "mean_token_accuracy": 0.7719471603631973,
      "num_tokens": 1032986.0,
      "step": 63
    },
    {
      "entropy": 0.5672716349363327,
      "epoch": 0.23880597014925373,
      "grad_norm": 0.16331897675991058,
      "learning_rate": 0.0002,
      "loss": 0.5665188431739807,
      "mean_token_accuracy": 0.7713304907083511,
      "num_tokens": 1049424.0,
      "step": 64
    },
    {
      "entropy": 0.5762098878622055,
      "epoch": 0.24253731343283583,
      "grad_norm": 0.13884824514389038,
      "learning_rate": 0.0002,
      "loss": 0.5767008662223816,
      "mean_token_accuracy": 0.7696898132562637,
      "num_tokens": 1065969.0,
      "step": 65
    },
    {
      "entropy": 0.5569287538528442,
      "epoch": 0.2462686567164179,
      "grad_norm": 0.1740816980600357,
      "learning_rate": 0.0002,
      "loss": 0.5631753206253052,
      "mean_token_accuracy": 0.7727806568145752,
      "num_tokens": 1082229.0,
      "step": 66
    },
    {
      "entropy": 0.5548796504735947,
      "epoch": 0.25,
      "grad_norm": 0.1667051911354065,
      "learning_rate": 0.0002,
      "loss": 0.5509235262870789,
      "mean_token_accuracy": 0.7790889889001846,
      "num_tokens": 1099010.0,
      "step": 67
    },
    {
      "entropy": 0.5462881028652191,
      "epoch": 0.2537313432835821,
      "grad_norm": 0.14191995561122894,
      "learning_rate": 0.0002,
      "loss": 0.5450850129127502,
      "mean_token_accuracy": 0.781976968050003,
      "num_tokens": 1115328.0,
      "step": 68
    },
    {
      "entropy": 0.5640436559915543,
      "epoch": 0.2574626865671642,
      "grad_norm": 0.16489970684051514,
      "learning_rate": 0.0002,
      "loss": 0.5714234113693237,
      "mean_token_accuracy": 0.7743055671453476,
      "num_tokens": 1131527.0,
      "step": 69
    },
    {
      "entropy": 0.5581524074077606,
      "epoch": 0.26119402985074625,
      "grad_norm": 0.17524468898773193,
      "learning_rate": 0.0002,
      "loss": 0.5653409361839294,
      "mean_token_accuracy": 0.7740405648946762,
      "num_tokens": 1147997.0,
      "step": 70
    },
    {
      "entropy": 0.5587889850139618,
      "epoch": 0.26492537313432835,
      "grad_norm": 0.13126389682292938,
      "learning_rate": 0.0002,
      "loss": 0.563335657119751,
      "mean_token_accuracy": 0.7715445905923843,
      "num_tokens": 1164153.0,
      "step": 71
    },
    {
      "entropy": 0.5487140715122223,
      "epoch": 0.26865671641791045,
      "grad_norm": 0.21885372698307037,
      "learning_rate": 0.0002,
      "loss": 0.5488876104354858,
      "mean_token_accuracy": 0.7801245003938675,
      "num_tokens": 1180525.0,
      "step": 72
    },
    {
      "entropy": 0.5341571718454361,
      "epoch": 0.27238805970149255,
      "grad_norm": 0.13752351701259613,
      "learning_rate": 0.0002,
      "loss": 0.5344278812408447,
      "mean_token_accuracy": 0.784472331404686,
      "num_tokens": 1196847.0,
      "step": 73
    },
    {
      "entropy": 0.551477774977684,
      "epoch": 0.27611940298507465,
      "grad_norm": 0.13981129229068756,
      "learning_rate": 0.0002,
      "loss": 0.5494024753570557,
      "mean_token_accuracy": 0.7799880355596542,
      "num_tokens": 1213481.0,
      "step": 74
    },
    {
      "entropy": 0.5447621941566467,
      "epoch": 0.2798507462686567,
      "grad_norm": 0.15920379757881165,
      "learning_rate": 0.0002,
      "loss": 0.5405699610710144,
      "mean_token_accuracy": 0.7826831936836243,
      "num_tokens": 1230084.0,
      "step": 75
    },
    {
      "entropy": 0.5486634969711304,
      "epoch": 0.2835820895522388,
      "grad_norm": 0.13304069638252258,
      "learning_rate": 0.0002,
      "loss": 0.5446336269378662,
      "mean_token_accuracy": 0.7830937057733536,
      "num_tokens": 1246769.0,
      "step": 76
    },
    {
      "entropy": 0.5452822148799896,
      "epoch": 0.2873134328358209,
      "grad_norm": 0.15798641741275787,
      "learning_rate": 0.0002,
      "loss": 0.557198703289032,
      "mean_token_accuracy": 0.7773135006427765,
      "num_tokens": 1263008.0,
      "step": 77
    },
    {
      "entropy": 0.5309847742319107,
      "epoch": 0.291044776119403,
      "grad_norm": 0.1608300358057022,
      "learning_rate": 0.0002,
      "loss": 0.5421797037124634,
      "mean_token_accuracy": 0.7828473746776581,
      "num_tokens": 1279441.0,
      "step": 78
    },
    {
      "entropy": 0.5622646063566208,
      "epoch": 0.2947761194029851,
      "grad_norm": 0.13157972693443298,
      "learning_rate": 0.0002,
      "loss": 0.558945894241333,
      "mean_token_accuracy": 0.7717398554086685,
      "num_tokens": 1295651.0,
      "step": 79
    },
    {
      "entropy": 0.5753958076238632,
      "epoch": 0.29850746268656714,
      "grad_norm": 0.14517387747764587,
      "learning_rate": 0.0002,
      "loss": 0.5727032423019409,
      "mean_token_accuracy": 0.7672933489084244,
      "num_tokens": 1311962.0,
      "step": 80
    },
    {
      "entropy": 0.5534999370574951,
      "epoch": 0.30223880597014924,
      "grad_norm": 0.1413838267326355,
      "learning_rate": 0.0002,
      "loss": 0.5444930195808411,
      "mean_token_accuracy": 0.7826777994632721,
      "num_tokens": 1328248.0,
      "step": 81
    },
    {
      "entropy": 0.5375708788633347,
      "epoch": 0.30597014925373134,
      "grad_norm": 0.13538850843906403,
      "learning_rate": 0.0002,
      "loss": 0.538631796836853,
      "mean_token_accuracy": 0.7803685069084167,
      "num_tokens": 1344467.0,
      "step": 82
    },
    {
      "entropy": 0.5297208279371262,
      "epoch": 0.30970149253731344,
      "grad_norm": 0.14185431599617004,
      "learning_rate": 0.0002,
      "loss": 0.5332384705543518,
      "mean_token_accuracy": 0.7804697453975677,
      "num_tokens": 1360718.0,
      "step": 83
    },
    {
      "entropy": 0.5498329550027847,
      "epoch": 0.31343283582089554,
      "grad_norm": 0.14588527381420135,
      "learning_rate": 0.0002,
      "loss": 0.5577006340026855,
      "mean_token_accuracy": 0.7738551795482635,
      "num_tokens": 1376845.0,
      "step": 84
    },
    {
      "entropy": 0.5527947694063187,
      "epoch": 0.31716417910447764,
      "grad_norm": 0.1357346624135971,
      "learning_rate": 0.0002,
      "loss": 0.5527830719947815,
      "mean_token_accuracy": 0.7788930237293243,
      "num_tokens": 1393076.0,
      "step": 85
    },
    {
      "entropy": 0.5394750982522964,
      "epoch": 0.3208955223880597,
      "grad_norm": 0.13744208216667175,
      "learning_rate": 0.0002,
      "loss": 0.5478135943412781,
      "mean_token_accuracy": 0.780300036072731,
      "num_tokens": 1409217.0,
      "step": 86
    },
    {
      "entropy": 0.5559671819210052,
      "epoch": 0.3246268656716418,
      "grad_norm": 0.13147583603858948,
      "learning_rate": 0.0002,
      "loss": 0.5562646985054016,
      "mean_token_accuracy": 0.7762454152107239,
      "num_tokens": 1425477.0,
      "step": 87
    },
    {
      "entropy": 0.5543066263198853,
      "epoch": 0.3283582089552239,
      "grad_norm": 0.11719482392072678,
      "learning_rate": 0.0002,
      "loss": 0.5529443621635437,
      "mean_token_accuracy": 0.7772444784641266,
      "num_tokens": 1441671.0,
      "step": 88
    },
    {
      "entropy": 0.5419782102108002,
      "epoch": 0.332089552238806,
      "grad_norm": 0.14719344675540924,
      "learning_rate": 0.0002,
      "loss": 0.5404766798019409,
      "mean_token_accuracy": 0.7831247895956039,
      "num_tokens": 1457971.0,
      "step": 89
    },
    {
      "entropy": 0.5399269163608551,
      "epoch": 0.3358208955223881,
      "grad_norm": 0.12220615893602371,
      "learning_rate": 0.0002,
      "loss": 0.5387459397315979,
      "mean_token_accuracy": 0.7855129837989807,
      "num_tokens": 1474274.0,
      "step": 90
    },
    {
      "entropy": 0.5337241441011429,
      "epoch": 0.33955223880597013,
      "grad_norm": 0.13422612845897675,
      "learning_rate": 0.0002,
      "loss": 0.5408236980438232,
      "mean_token_accuracy": 0.7799499779939651,
      "num_tokens": 1490481.0,
      "step": 91
    },
    {
      "entropy": 0.5499130040407181,
      "epoch": 0.34328358208955223,
      "grad_norm": 0.1426658183336258,
      "learning_rate": 0.0002,
      "loss": 0.5537517666816711,
      "mean_token_accuracy": 0.7782512009143829,
      "num_tokens": 1506700.0,
      "step": 92
    },
    {
      "entropy": 0.5318698287010193,
      "epoch": 0.34701492537313433,
      "grad_norm": 0.12993979454040527,
      "learning_rate": 0.0002,
      "loss": 0.5316786766052246,
      "mean_token_accuracy": 0.7850093394517899,
      "num_tokens": 1523002.0,
      "step": 93
    },
    {
      "entropy": 0.5481119602918625,
      "epoch": 0.35074626865671643,
      "grad_norm": 0.13163146376609802,
      "learning_rate": 0.0002,
      "loss": 0.548253059387207,
      "mean_token_accuracy": 0.780367836356163,
      "num_tokens": 1539288.0,
      "step": 94
    },
    {
      "entropy": 0.5703603178262711,
      "epoch": 0.35447761194029853,
      "grad_norm": 0.12978118658065796,
      "learning_rate": 0.0002,
      "loss": 0.5666322708129883,
      "mean_token_accuracy": 0.7723395079374313,
      "num_tokens": 1555506.0,
      "step": 95
    },
    {
      "entropy": 0.5624447613954544,
      "epoch": 0.3582089552238806,
      "grad_norm": 0.12965592741966248,
      "learning_rate": 0.0002,
      "loss": 0.5602748394012451,
      "mean_token_accuracy": 0.7747233361005783,
      "num_tokens": 1572115.0,
      "step": 96
    },
    {
      "entropy": 0.5613084137439728,
      "epoch": 0.3619402985074627,
      "grad_norm": 0.14199502766132355,
      "learning_rate": 0.0002,
      "loss": 0.555010974407196,
      "mean_token_accuracy": 0.7760839462280273,
      "num_tokens": 1588821.0,
      "step": 97
    },
    {
      "entropy": 0.5324004143476486,
      "epoch": 0.3656716417910448,
      "grad_norm": 0.13919012248516083,
      "learning_rate": 0.0002,
      "loss": 0.5334999561309814,
      "mean_token_accuracy": 0.7850101590156555,
      "num_tokens": 1605300.0,
      "step": 98
    },
    {
      "entropy": 0.5360228419303894,
      "epoch": 0.3694029850746269,
      "grad_norm": 0.14360296726226807,
      "learning_rate": 0.0002,
      "loss": 0.5472567081451416,
      "mean_token_accuracy": 0.7786711901426315,
      "num_tokens": 1621431.0,
      "step": 99
    },
    {
      "entropy": 0.5455528795719147,
      "epoch": 0.373134328358209,
      "grad_norm": 0.1277836412191391,
      "learning_rate": 0.0002,
      "loss": 0.5538268089294434,
      "mean_token_accuracy": 0.7776066958904266,
      "num_tokens": 1637964.0,
      "step": 100
    },
    {
      "entropy": 0.5532989650964737,
      "epoch": 0.376865671641791,
      "grad_norm": 0.13713549077510834,
      "learning_rate": 0.0002,
      "loss": 0.5561457872390747,
      "mean_token_accuracy": 0.7771011143922806,
      "num_tokens": 1654089.0,
      "step": 101
    },
    {
      "entropy": 0.5741413533687592,
      "epoch": 0.3805970149253731,
      "grad_norm": 0.13025760650634766,
      "learning_rate": 0.0002,
      "loss": 0.5691399574279785,
      "mean_token_accuracy": 0.772384986281395,
      "num_tokens": 1670665.0,
      "step": 102
    },
    {
      "entropy": 0.5667874813079834,
      "epoch": 0.3843283582089552,
      "grad_norm": 0.12920764088630676,
      "learning_rate": 0.0002,
      "loss": 0.5599198341369629,
      "mean_token_accuracy": 0.7729279398918152,
      "num_tokens": 1687063.0,
      "step": 103
    },
    {
      "entropy": 0.5548859238624573,
      "epoch": 0.3880597014925373,
      "grad_norm": 0.13648352026939392,
      "learning_rate": 0.0002,
      "loss": 0.5587888956069946,
      "mean_token_accuracy": 0.7738962918519974,
      "num_tokens": 1703545.0,
      "step": 104
    },
    {
      "entropy": 0.5481197685003281,
      "epoch": 0.3917910447761194,
      "grad_norm": 0.13556665182113647,
      "learning_rate": 0.0002,
      "loss": 0.551975667476654,
      "mean_token_accuracy": 0.7755906730890274,
      "num_tokens": 1720094.0,
      "step": 105
    },
    {
      "entropy": 0.5406413227319717,
      "epoch": 0.39552238805970147,
      "grad_norm": 0.12847866117954254,
      "learning_rate": 0.0002,
      "loss": 0.5447836518287659,
      "mean_token_accuracy": 0.7771267741918564,
      "num_tokens": 1736466.0,
      "step": 106
    },
    {
      "entropy": 0.5483510047197342,
      "epoch": 0.39925373134328357,
      "grad_norm": 0.14766491949558258,
      "learning_rate": 0.0002,
      "loss": 0.54759281873703,
      "mean_token_accuracy": 0.7777384519577026,
      "num_tokens": 1753014.0,
      "step": 107
    },
    {
      "entropy": 0.5594521313905716,
      "epoch": 0.40298507462686567,
      "grad_norm": 0.1325310319662094,
      "learning_rate": 0.0002,
      "loss": 0.5559228658676147,
      "mean_token_accuracy": 0.7776539623737335,
      "num_tokens": 1769397.0,
      "step": 108
    },
    {
      "entropy": 0.5235730409622192,
      "epoch": 0.40671641791044777,
      "grad_norm": 0.14838212728500366,
      "learning_rate": 0.0002,
      "loss": 0.5363551378250122,
      "mean_token_accuracy": 0.7855585068464279,
      "num_tokens": 1785496.0,
      "step": 109
    },
    {
      "entropy": 0.529867097735405,
      "epoch": 0.41044776119402987,
      "grad_norm": 0.1400843858718872,
      "learning_rate": 0.0002,
      "loss": 0.5262739062309265,
      "mean_token_accuracy": 0.7853756546974182,
      "num_tokens": 1802161.0,
      "step": 110
    },
    {
      "entropy": 0.5641083866357803,
      "epoch": 0.4141791044776119,
      "grad_norm": 0.12946833670139313,
      "learning_rate": 0.0002,
      "loss": 0.5594510436058044,
      "mean_token_accuracy": 0.7754105180501938,
      "num_tokens": 1818597.0,
      "step": 111
    },
    {
      "entropy": 0.5582448691129684,
      "epoch": 0.417910447761194,
      "grad_norm": 0.141814723610878,
      "learning_rate": 0.0002,
      "loss": 0.5595210194587708,
      "mean_token_accuracy": 0.7736571133136749,
      "num_tokens": 1834990.0,
      "step": 112
    },
    {
      "entropy": 0.5417420417070389,
      "epoch": 0.4216417910447761,
      "grad_norm": 0.12932373583316803,
      "learning_rate": 0.0002,
      "loss": 0.547935962677002,
      "mean_token_accuracy": 0.7809092104434967,
      "num_tokens": 1851431.0,
      "step": 113
    },
    {
      "entropy": 0.5411195009946823,
      "epoch": 0.4253731343283582,
      "grad_norm": 0.13595683872699738,
      "learning_rate": 0.0002,
      "loss": 0.5450493097305298,
      "mean_token_accuracy": 0.7794583141803741,
      "num_tokens": 1867771.0,
      "step": 114
    },
    {
      "entropy": 0.5582379996776581,
      "epoch": 0.4291044776119403,
      "grad_norm": 0.13590583205223083,
      "learning_rate": 0.0002,
      "loss": 0.5637767910957336,
      "mean_token_accuracy": 0.7740200757980347,
      "num_tokens": 1884142.0,
      "step": 115
    },
    {
      "entropy": 0.5411937683820724,
      "epoch": 0.43283582089552236,
      "grad_norm": 0.14205913245677948,
      "learning_rate": 0.0002,
      "loss": 0.542431116104126,
      "mean_token_accuracy": 0.7799777090549469,
      "num_tokens": 1900516.0,
      "step": 116
    },
    {
      "entropy": 0.5482146292924881,
      "epoch": 0.43656716417910446,
      "grad_norm": 0.11477427929639816,
      "learning_rate": 0.0002,
      "loss": 0.5456505417823792,
      "mean_token_accuracy": 0.7757869809865952,
      "num_tokens": 1916885.0,
      "step": 117
    },
    {
      "entropy": 0.5370079576969147,
      "epoch": 0.44029850746268656,
      "grad_norm": 0.13816410303115845,
      "learning_rate": 0.0002,
      "loss": 0.5429069399833679,
      "mean_token_accuracy": 0.7780599594116211,
      "num_tokens": 1933169.0,
      "step": 118
    },
    {
      "entropy": 0.5455102026462555,
      "epoch": 0.44402985074626866,
      "grad_norm": 0.13368608057498932,
      "learning_rate": 0.0002,
      "loss": 0.5398865938186646,
      "mean_token_accuracy": 0.783423513174057,
      "num_tokens": 1949594.0,
      "step": 119
    },
    {
      "entropy": 0.5368959903717041,
      "epoch": 0.44776119402985076,
      "grad_norm": 0.13672706484794617,
      "learning_rate": 0.0002,
      "loss": 0.5356795787811279,
      "mean_token_accuracy": 0.7833909094333649,
      "num_tokens": 1965772.0,
      "step": 120
    },
    {
      "entropy": 0.5283652544021606,
      "epoch": 0.45149253731343286,
      "grad_norm": 0.16315960884094238,
      "learning_rate": 0.0002,
      "loss": 0.5378228425979614,
      "mean_token_accuracy": 0.7841643989086151,
      "num_tokens": 1981885.0,
      "step": 121
    },
    {
      "entropy": 0.5414575785398483,
      "epoch": 0.4552238805970149,
      "grad_norm": 0.13416600227355957,
      "learning_rate": 0.0002,
      "loss": 0.5479776859283447,
      "mean_token_accuracy": 0.7784606367349625,
      "num_tokens": 1998430.0,
      "step": 122
    },
    {
      "entropy": 0.5641719400882721,
      "epoch": 0.458955223880597,
      "grad_norm": 0.13336408138275146,
      "learning_rate": 0.0002,
      "loss": 0.5604909658432007,
      "mean_token_accuracy": 0.772265836596489,
      "num_tokens": 2015003.0,
      "step": 123
    },
    {
      "entropy": 0.5521317422389984,
      "epoch": 0.4626865671641791,
      "grad_norm": 0.13578884303569794,
      "learning_rate": 0.0002,
      "loss": 0.5430817604064941,
      "mean_token_accuracy": 0.780389592051506,
      "num_tokens": 2031375.0,
      "step": 124
    },
    {
      "entropy": 0.5561365038156509,
      "epoch": 0.4664179104477612,
      "grad_norm": 0.13006186485290527,
      "learning_rate": 0.0002,
      "loss": 0.5517395734786987,
      "mean_token_accuracy": 0.7766251713037491,
      "num_tokens": 2047573.0,
      "step": 125
    },
    {
      "entropy": 0.5579850822687149,
      "epoch": 0.4701492537313433,
      "grad_norm": 0.12219003587961197,
      "learning_rate": 0.0002,
      "loss": 0.557759702205658,
      "mean_token_accuracy": 0.7729216068983078,
      "num_tokens": 2064084.0,
      "step": 126
    },
    {
      "entropy": 0.5411650538444519,
      "epoch": 0.47388059701492535,
      "grad_norm": 0.15279285609722137,
      "learning_rate": 0.0002,
      "loss": 0.5496426224708557,
      "mean_token_accuracy": 0.7773544490337372,
      "num_tokens": 2080693.0,
      "step": 127
    },
    {
      "entropy": 0.5413447618484497,
      "epoch": 0.47761194029850745,
      "grad_norm": 0.15482240915298462,
      "learning_rate": 0.0002,
      "loss": 0.5425417423248291,
      "mean_token_accuracy": 0.780493900179863,
      "num_tokens": 2097090.0,
      "step": 128
    },
    {
      "entropy": 0.5401688069105148,
      "epoch": 0.48134328358208955,
      "grad_norm": 0.13282036781311035,
      "learning_rate": 0.0002,
      "loss": 0.543215274810791,
      "mean_token_accuracy": 0.7801757156848907,
      "num_tokens": 2113369.0,
      "step": 129
    },
    {
      "entropy": 0.5527068078517914,
      "epoch": 0.48507462686567165,
      "grad_norm": 0.13907410204410553,
      "learning_rate": 0.0002,
      "loss": 0.5536910891532898,
      "mean_token_accuracy": 0.7762171775102615,
      "num_tokens": 2129711.0,
      "step": 130
    },
    {
      "entropy": 0.5255335569381714,
      "epoch": 0.48880597014925375,
      "grad_norm": 0.12128844857215881,
      "learning_rate": 0.0002,
      "loss": 0.5272364020347595,
      "mean_token_accuracy": 0.7870801836252213,
      "num_tokens": 2145726.0,
      "step": 131
    },
    {
      "entropy": 0.5413011759519577,
      "epoch": 0.4925373134328358,
      "grad_norm": 0.11150290817022324,
      "learning_rate": 0.0002,
      "loss": 0.5447565317153931,
      "mean_token_accuracy": 0.7812841534614563,
      "num_tokens": 2162226.0,
      "step": 132
    },
    {
      "entropy": 0.5405918657779694,
      "epoch": 0.4962686567164179,
      "grad_norm": 0.180214062333107,
      "learning_rate": 0.0002,
      "loss": 0.5419008731842041,
      "mean_token_accuracy": 0.779103696346283,
      "num_tokens": 2178734.0,
      "step": 133
    },
    {
      "entropy": 0.5386396795511246,
      "epoch": 0.5,
      "grad_norm": 0.11724670231342316,
      "learning_rate": 0.0002,
      "loss": 0.5388697981834412,
      "mean_token_accuracy": 0.7833260148763657,
      "num_tokens": 2195006.0,
      "step": 134
    },
    {
      "entropy": 0.5358224213123322,
      "epoch": 0.503731343283582,
      "grad_norm": 0.16803957521915436,
      "learning_rate": 0.0002,
      "loss": 0.5442015528678894,
      "mean_token_accuracy": 0.7798204123973846,
      "num_tokens": 2211091.0,
      "step": 135
    },
    {
      "entropy": 0.5409800261259079,
      "epoch": 0.5074626865671642,
      "grad_norm": 0.11882071197032928,
      "learning_rate": 0.0002,
      "loss": 0.5377349257469177,
      "mean_token_accuracy": 0.7844288647174835,
      "num_tokens": 2227557.0,
      "step": 136
    },
    {
      "entropy": 0.5705572813749313,
      "epoch": 0.5111940298507462,
      "grad_norm": 0.14624297618865967,
      "learning_rate": 0.0002,
      "loss": 0.5614476203918457,
      "mean_token_accuracy": 0.7717961966991425,
      "num_tokens": 2243947.0,
      "step": 137
    },
    {
      "entropy": 0.5261306613683701,
      "epoch": 0.5149253731343284,
      "grad_norm": 0.12288782745599747,
      "learning_rate": 0.0002,
      "loss": 0.5277607440948486,
      "mean_token_accuracy": 0.7837703675031662,
      "num_tokens": 2260248.0,
      "step": 138
    },
    {
      "entropy": 0.5489343255758286,
      "epoch": 0.5186567164179104,
      "grad_norm": 0.12894777953624725,
      "learning_rate": 0.0002,
      "loss": 0.5560361742973328,
      "mean_token_accuracy": 0.7759750783443451,
      "num_tokens": 2276747.0,
      "step": 139
    },
    {
      "entropy": 0.5430862605571747,
      "epoch": 0.5223880597014925,
      "grad_norm": 0.14144180715084076,
      "learning_rate": 0.0002,
      "loss": 0.547459602355957,
      "mean_token_accuracy": 0.7805726677179337,
      "num_tokens": 2293077.0,
      "step": 140
    },
    {
      "entropy": 0.5440323650836945,
      "epoch": 0.5261194029850746,
      "grad_norm": 0.11917843669652939,
      "learning_rate": 0.0002,
      "loss": 0.5477656722068787,
      "mean_token_accuracy": 0.7797518819570541,
      "num_tokens": 2309337.0,
      "step": 141
    },
    {
      "entropy": 0.5420307070016861,
      "epoch": 0.5298507462686567,
      "grad_norm": 0.1192653626203537,
      "learning_rate": 0.0002,
      "loss": 0.5394734144210815,
      "mean_token_accuracy": 0.7826057225465775,
      "num_tokens": 2325670.0,
      "step": 142
    },
    {
      "entropy": 0.5393340289592743,
      "epoch": 0.5335820895522388,
      "grad_norm": 0.13303329050540924,
      "learning_rate": 0.0002,
      "loss": 0.5385794639587402,
      "mean_token_accuracy": 0.7815165519714355,
      "num_tokens": 2341755.0,
      "step": 143
    },
    {
      "entropy": 0.5444102138280869,
      "epoch": 0.5373134328358209,
      "grad_norm": 0.14008203148841858,
      "learning_rate": 0.0002,
      "loss": 0.5488988757133484,
      "mean_token_accuracy": 0.7799032479524612,
      "num_tokens": 2358153.0,
      "step": 144
    },
    {
      "entropy": 0.5365025699138641,
      "epoch": 0.5410447761194029,
      "grad_norm": 0.14566808938980103,
      "learning_rate": 0.0002,
      "loss": 0.5419079661369324,
      "mean_token_accuracy": 0.7805670350790024,
      "num_tokens": 2374467.0,
      "step": 145
    },
    {
      "entropy": 0.5426971465349197,
      "epoch": 0.5447761194029851,
      "grad_norm": 0.1357433944940567,
      "learning_rate": 0.0002,
      "loss": 0.5379341244697571,
      "mean_token_accuracy": 0.7826648503541946,
      "num_tokens": 2390600.0,
      "step": 146
    },
    {
      "entropy": 0.5661343783140182,
      "epoch": 0.5485074626865671,
      "grad_norm": 0.13934989273548126,
      "learning_rate": 0.0002,
      "loss": 0.5629899501800537,
      "mean_token_accuracy": 0.7758099287748337,
      "num_tokens": 2407080.0,
      "step": 147
    },
    {
      "entropy": 0.541400671005249,
      "epoch": 0.5522388059701493,
      "grad_norm": 0.14024044573307037,
      "learning_rate": 0.0002,
      "loss": 0.541731595993042,
      "mean_token_accuracy": 0.7802949547767639,
      "num_tokens": 2423568.0,
      "step": 148
    },
    {
      "entropy": 0.5476364940404892,
      "epoch": 0.5559701492537313,
      "grad_norm": 0.1358904242515564,
      "learning_rate": 0.0002,
      "loss": 0.5512872934341431,
      "mean_token_accuracy": 0.7763868570327759,
      "num_tokens": 2439947.0,
      "step": 149
    },
    {
      "entropy": 0.546050176024437,
      "epoch": 0.5597014925373134,
      "grad_norm": 0.14050431549549103,
      "learning_rate": 0.0002,
      "loss": 0.5520421266555786,
      "mean_token_accuracy": 0.7763620764017105,
      "num_tokens": 2456394.0,
      "step": 150
    },
    {
      "entropy": 0.5617920160293579,
      "epoch": 0.5634328358208955,
      "grad_norm": 0.12195290625095367,
      "learning_rate": 0.0002,
      "loss": 0.5633358955383301,
      "mean_token_accuracy": 0.7721698135137558,
      "num_tokens": 2472801.0,
      "step": 151
    },
    {
      "entropy": 0.5335766077041626,
      "epoch": 0.5671641791044776,
      "grad_norm": 0.1224474310874939,
      "learning_rate": 0.0002,
      "loss": 0.5250234007835388,
      "mean_token_accuracy": 0.7903417944908142,
      "num_tokens": 2489048.0,
      "step": 152
    },
    {
      "entropy": 0.5577341765165329,
      "epoch": 0.5708955223880597,
      "grad_norm": 0.12528087198734283,
      "learning_rate": 0.0002,
      "loss": 0.561303436756134,
      "mean_token_accuracy": 0.7743656784296036,
      "num_tokens": 2505442.0,
      "step": 153
    },
    {
      "entropy": 0.5553326904773712,
      "epoch": 0.5746268656716418,
      "grad_norm": 0.12870363891124725,
      "learning_rate": 0.0002,
      "loss": 0.5615543127059937,
      "mean_token_accuracy": 0.7726895660161972,
      "num_tokens": 2521815.0,
      "step": 154
    },
    {
      "entropy": 0.5630124509334564,
      "epoch": 0.5783582089552238,
      "grad_norm": 0.12597967684268951,
      "learning_rate": 0.0002,
      "loss": 0.5687354803085327,
      "mean_token_accuracy": 0.7691132724285126,
      "num_tokens": 2538286.0,
      "step": 155
    },
    {
      "entropy": 0.5581531077623367,
      "epoch": 0.582089552238806,
      "grad_norm": 0.1168755516409874,
      "learning_rate": 0.0002,
      "loss": 0.5532963275909424,
      "mean_token_accuracy": 0.775028184056282,
      "num_tokens": 2554519.0,
      "step": 156
    },
    {
      "entropy": 0.5546217858791351,
      "epoch": 0.585820895522388,
      "grad_norm": 0.131391242146492,
      "learning_rate": 0.0002,
      "loss": 0.5544172525405884,
      "mean_token_accuracy": 0.772693082690239,
      "num_tokens": 2570785.0,
      "step": 157
    },
    {
      "entropy": 0.5427669435739517,
      "epoch": 0.5895522388059702,
      "grad_norm": 0.14516305923461914,
      "learning_rate": 0.0002,
      "loss": 0.5510904788970947,
      "mean_token_accuracy": 0.7808411121368408,
      "num_tokens": 2587443.0,
      "step": 158
    },
    {
      "entropy": 0.5409528166055679,
      "epoch": 0.5932835820895522,
      "grad_norm": 0.17116999626159668,
      "learning_rate": 0.0002,
      "loss": 0.5488826036453247,
      "mean_token_accuracy": 0.777285099029541,
      "num_tokens": 2603592.0,
      "step": 159
    },
    {
      "entropy": 0.5733516365289688,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.13355422019958496,
      "learning_rate": 0.0002,
      "loss": 0.5614065527915955,
      "mean_token_accuracy": 0.7743426263332367,
      "num_tokens": 2619909.0,
      "step": 160
    },
    {
      "entropy": 0.5415901988744736,
      "epoch": 0.6007462686567164,
      "grad_norm": 0.13737861812114716,
      "learning_rate": 0.0002,
      "loss": 0.5314146280288696,
      "mean_token_accuracy": 0.785938948392868,
      "num_tokens": 2636085.0,
      "step": 161
    },
    {
      "entropy": 0.5612253397703171,
      "epoch": 0.6044776119402985,
      "grad_norm": 0.14785514771938324,
      "learning_rate": 0.0002,
      "loss": 0.5628178715705872,
      "mean_token_accuracy": 0.7715339511632919,
      "num_tokens": 2652634.0,
      "step": 162
    },
    {
      "entropy": 0.5437108278274536,
      "epoch": 0.6082089552238806,
      "grad_norm": 0.10948356240987778,
      "learning_rate": 0.0002,
      "loss": 0.547478437423706,
      "mean_token_accuracy": 0.7772163152694702,
      "num_tokens": 2668983.0,
      "step": 163
    },
    {
      "entropy": 0.5426544100046158,
      "epoch": 0.6119402985074627,
      "grad_norm": 0.18059010803699493,
      "learning_rate": 0.0002,
      "loss": 0.5537011027336121,
      "mean_token_accuracy": 0.7747623771429062,
      "num_tokens": 2685249.0,
      "step": 164
    },
    {
      "entropy": 0.529934898018837,
      "epoch": 0.6156716417910447,
      "grad_norm": 0.11337770521640778,
      "learning_rate": 0.0002,
      "loss": 0.5303418636322021,
      "mean_token_accuracy": 0.7853435575962067,
      "num_tokens": 2701660.0,
      "step": 165
    },
    {
      "entropy": 0.5464923083782196,
      "epoch": 0.6194029850746269,
      "grad_norm": 0.14328055083751678,
      "learning_rate": 0.0002,
      "loss": 0.5412318706512451,
      "mean_token_accuracy": 0.7794769406318665,
      "num_tokens": 2718009.0,
      "step": 166
    },
    {
      "entropy": 0.5332994908094406,
      "epoch": 0.6231343283582089,
      "grad_norm": 0.124599389731884,
      "learning_rate": 0.0002,
      "loss": 0.5295602679252625,
      "mean_token_accuracy": 0.7858390063047409,
      "num_tokens": 2733940.0,
      "step": 167
    },
    {
      "entropy": 0.5405105203390121,
      "epoch": 0.6268656716417911,
      "grad_norm": 0.1258486807346344,
      "learning_rate": 0.0002,
      "loss": 0.5447245240211487,
      "mean_token_accuracy": 0.7793877273797989,
      "num_tokens": 2750120.0,
      "step": 168
    },
    {
      "entropy": 0.5483063012361526,
      "epoch": 0.6305970149253731,
      "grad_norm": 0.12760283052921295,
      "learning_rate": 0.0002,
      "loss": 0.5542790293693542,
      "mean_token_accuracy": 0.777314305305481,
      "num_tokens": 2766400.0,
      "step": 169
    },
    {
      "entropy": 0.5282255113124847,
      "epoch": 0.6343283582089553,
      "grad_norm": 0.11275236308574677,
      "learning_rate": 0.0002,
      "loss": 0.532568633556366,
      "mean_token_accuracy": 0.7843974977731705,
      "num_tokens": 2782639.0,
      "step": 170
    },
    {
      "entropy": 0.5487422794103622,
      "epoch": 0.6380597014925373,
      "grad_norm": 0.12859784066677094,
      "learning_rate": 0.0002,
      "loss": 0.5562790632247925,
      "mean_token_accuracy": 0.7746353149414062,
      "num_tokens": 2798847.0,
      "step": 171
    },
    {
      "entropy": 0.5516662746667862,
      "epoch": 0.6417910447761194,
      "grad_norm": 0.1135534718632698,
      "learning_rate": 0.0002,
      "loss": 0.5541255474090576,
      "mean_token_accuracy": 0.7761951684951782,
      "num_tokens": 2815484.0,
      "step": 172
    },
    {
      "entropy": 0.5466583147644997,
      "epoch": 0.6455223880597015,
      "grad_norm": 0.12440499663352966,
      "learning_rate": 0.0002,
      "loss": 0.5376479625701904,
      "mean_token_accuracy": 0.7826560735702515,
      "num_tokens": 2831953.0,
      "step": 173
    },
    {
      "entropy": 0.5650184452533722,
      "epoch": 0.6492537313432836,
      "grad_norm": 0.12500424683094025,
      "learning_rate": 0.0002,
      "loss": 0.5570256114006042,
      "mean_token_accuracy": 0.775683656334877,
      "num_tokens": 2848576.0,
      "step": 174
    },
    {
      "entropy": 0.5299219936132431,
      "epoch": 0.6529850746268657,
      "grad_norm": 0.12209604680538177,
      "learning_rate": 0.0002,
      "loss": 0.5341060757637024,
      "mean_token_accuracy": 0.7854541093111038,
      "num_tokens": 2865028.0,
      "step": 175
    },
    {
      "entropy": 0.527473658323288,
      "epoch": 0.6567164179104478,
      "grad_norm": 0.13374948501586914,
      "learning_rate": 0.0002,
      "loss": 0.5358298420906067,
      "mean_token_accuracy": 0.7821341753005981,
      "num_tokens": 2881235.0,
      "step": 176
    },
    {
      "entropy": 0.530792161822319,
      "epoch": 0.6604477611940298,
      "grad_norm": 0.1418503224849701,
      "learning_rate": 0.0002,
      "loss": 0.5484696626663208,
      "mean_token_accuracy": 0.7771261185407639,
      "num_tokens": 2897352.0,
      "step": 177
    },
    {
      "entropy": 0.529504582285881,
      "epoch": 0.664179104477612,
      "grad_norm": 0.1355515867471695,
      "learning_rate": 0.0002,
      "loss": 0.5337001085281372,
      "mean_token_accuracy": 0.7831253409385681,
      "num_tokens": 2913466.0,
      "step": 178
    },
    {
      "entropy": 0.5478560030460358,
      "epoch": 0.667910447761194,
      "grad_norm": 0.12079652398824692,
      "learning_rate": 0.0002,
      "loss": 0.5434887409210205,
      "mean_token_accuracy": 0.7814661264419556,
      "num_tokens": 2929627.0,
      "step": 179
    },
    {
      "entropy": 0.5615387558937073,
      "epoch": 0.6716417910447762,
      "grad_norm": 0.12244576960802078,
      "learning_rate": 0.0002,
      "loss": 0.5516577959060669,
      "mean_token_accuracy": 0.7753744870424271,
      "num_tokens": 2946165.0,
      "step": 180
    },
    {
      "entropy": 0.5508608520030975,
      "epoch": 0.6753731343283582,
      "grad_norm": 0.14292050898075104,
      "learning_rate": 0.0002,
      "loss": 0.5470070242881775,
      "mean_token_accuracy": 0.7745987325906754,
      "num_tokens": 2962296.0,
      "step": 181
    },
    {
      "entropy": 0.5406470000743866,
      "epoch": 0.6791044776119403,
      "grad_norm": 0.12849874794483185,
      "learning_rate": 0.0002,
      "loss": 0.5443695187568665,
      "mean_token_accuracy": 0.7796088755130768,
      "num_tokens": 2978771.0,
      "step": 182
    },
    {
      "entropy": 0.5391182154417038,
      "epoch": 0.6828358208955224,
      "grad_norm": 0.14676682651042938,
      "learning_rate": 0.0002,
      "loss": 0.5457857251167297,
      "mean_token_accuracy": 0.7784734219312668,
      "num_tokens": 2995061.0,
      "step": 183
    },
    {
      "entropy": 0.560190424323082,
      "epoch": 0.6865671641791045,
      "grad_norm": 0.12630562484264374,
      "learning_rate": 0.0002,
      "loss": 0.5650532841682434,
      "mean_token_accuracy": 0.7698299288749695,
      "num_tokens": 3011564.0,
      "step": 184
    },
    {
      "entropy": 0.56383316218853,
      "epoch": 0.6902985074626866,
      "grad_norm": 0.11451014131307602,
      "learning_rate": 0.0002,
      "loss": 0.5619198083877563,
      "mean_token_accuracy": 0.7752022296190262,
      "num_tokens": 3028041.0,
      "step": 185
    },
    {
      "entropy": 0.5337186753749847,
      "epoch": 0.6940298507462687,
      "grad_norm": 0.11182348430156708,
      "learning_rate": 0.0002,
      "loss": 0.5291807055473328,
      "mean_token_accuracy": 0.7860076576471329,
      "num_tokens": 3044427.0,
      "step": 186
    },
    {
      "entropy": 0.5426793694496155,
      "epoch": 0.6977611940298507,
      "grad_norm": 0.1289970427751541,
      "learning_rate": 0.0002,
      "loss": 0.5458519458770752,
      "mean_token_accuracy": 0.7804786115884781,
      "num_tokens": 3060550.0,
      "step": 187
    },
    {
      "entropy": 0.5481364578008652,
      "epoch": 0.7014925373134329,
      "grad_norm": 0.11624104529619217,
      "learning_rate": 0.0002,
      "loss": 0.5494486689567566,
      "mean_token_accuracy": 0.7777003347873688,
      "num_tokens": 3076985.0,
      "step": 188
    },
    {
      "entropy": 0.5405264496803284,
      "epoch": 0.7052238805970149,
      "grad_norm": 0.12466499209403992,
      "learning_rate": 0.0002,
      "loss": 0.5457755923271179,
      "mean_token_accuracy": 0.7773817032575607,
      "num_tokens": 3093305.0,
      "step": 189
    },
    {
      "entropy": 0.5467436462640762,
      "epoch": 0.7089552238805971,
      "grad_norm": 0.16258826851844788,
      "learning_rate": 0.0002,
      "loss": 0.5626595616340637,
      "mean_token_accuracy": 0.7722655385732651,
      "num_tokens": 3109841.0,
      "step": 190
    },
    {
      "entropy": 0.5435369312763214,
      "epoch": 0.7126865671641791,
      "grad_norm": 0.1250208616256714,
      "learning_rate": 0.0002,
      "loss": 0.5456626415252686,
      "mean_token_accuracy": 0.7792533487081528,
      "num_tokens": 3126390.0,
      "step": 191
    },
    {
      "entropy": 0.5665273517370224,
      "epoch": 0.7164179104477612,
      "grad_norm": 0.13022343814373016,
      "learning_rate": 0.0002,
      "loss": 0.5511953234672546,
      "mean_token_accuracy": 0.7755104452371597,
      "num_tokens": 3142789.0,
      "step": 192
    },
    {
      "entropy": 0.5572501718997955,
      "epoch": 0.7201492537313433,
      "grad_norm": 0.13365279138088226,
      "learning_rate": 0.0002,
      "loss": 0.5513952374458313,
      "mean_token_accuracy": 0.7761309891939163,
      "num_tokens": 3159053.0,
      "step": 193
    },
    {
      "entropy": 0.5266987681388855,
      "epoch": 0.7238805970149254,
      "grad_norm": 0.1201450526714325,
      "learning_rate": 0.0002,
      "loss": 0.5288004875183105,
      "mean_token_accuracy": 0.7855914831161499,
      "num_tokens": 3175525.0,
      "step": 194
    },
    {
      "entropy": 0.5352121591567993,
      "epoch": 0.7276119402985075,
      "grad_norm": 0.1398552507162094,
      "learning_rate": 0.0002,
      "loss": 0.5395978689193726,
      "mean_token_accuracy": 0.7819965928792953,
      "num_tokens": 3192099.0,
      "step": 195
    },
    {
      "entropy": 0.533184140920639,
      "epoch": 0.7313432835820896,
      "grad_norm": 0.1487068384885788,
      "learning_rate": 0.0002,
      "loss": 0.5468820929527283,
      "mean_token_accuracy": 0.7791697829961777,
      "num_tokens": 3208505.0,
      "step": 196
    },
    {
      "entropy": 0.5317975580692291,
      "epoch": 0.7350746268656716,
      "grad_norm": 0.12477895617485046,
      "learning_rate": 0.0002,
      "loss": 0.5355808734893799,
      "mean_token_accuracy": 0.7824379354715347,
      "num_tokens": 3225010.0,
      "step": 197
    },
    {
      "entropy": 0.5748307108879089,
      "epoch": 0.7388059701492538,
      "grad_norm": 0.12914563715457916,
      "learning_rate": 0.0002,
      "loss": 0.5788255929946899,
      "mean_token_accuracy": 0.7650759667158127,
      "num_tokens": 3241658.0,
      "step": 198
    },
    {
      "entropy": 0.5368754714727402,
      "epoch": 0.7425373134328358,
      "grad_norm": 0.11934071779251099,
      "learning_rate": 0.0002,
      "loss": 0.5274195075035095,
      "mean_token_accuracy": 0.7854534685611725,
      "num_tokens": 3258064.0,
      "step": 199
    },
    {
      "entropy": 0.5409929752349854,
      "epoch": 0.746268656716418,
      "grad_norm": 0.1189829558134079,
      "learning_rate": 0.0002,
      "loss": 0.5376108288764954,
      "mean_token_accuracy": 0.7814257591962814,
      "num_tokens": 3274512.0,
      "step": 200
    },
    {
      "entropy": 0.5288984030485153,
      "epoch": 0.75,
      "grad_norm": 0.18263517320156097,
      "learning_rate": 0.0002,
      "loss": 0.5427844524383545,
      "mean_token_accuracy": 0.7802873849868774,
      "num_tokens": 3290639.0,
      "step": 201
    },
    {
      "entropy": 0.5691705942153931,
      "epoch": 0.753731343283582,
      "grad_norm": 0.1280129849910736,
      "learning_rate": 0.0002,
      "loss": 0.5745768547058105,
      "mean_token_accuracy": 0.7654204666614532,
      "num_tokens": 3307020.0,
      "step": 202
    },
    {
      "entropy": 0.5719635188579559,
      "epoch": 0.7574626865671642,
      "grad_norm": 0.11226256936788559,
      "learning_rate": 0.0002,
      "loss": 0.5669863224029541,
      "mean_token_accuracy": 0.7705253958702087,
      "num_tokens": 3323426.0,
      "step": 203
    },
    {
      "entropy": 0.5740833729505539,
      "epoch": 0.7611940298507462,
      "grad_norm": 0.16702376306056976,
      "learning_rate": 0.0002,
      "loss": 0.5620068907737732,
      "mean_token_accuracy": 0.769936203956604,
      "num_tokens": 3339801.0,
      "step": 204
    },
    {
      "entropy": 0.5441322475671768,
      "epoch": 0.7649253731343284,
      "grad_norm": 0.13137376308441162,
      "learning_rate": 0.0002,
      "loss": 0.5305750370025635,
      "mean_token_accuracy": 0.7868027091026306,
      "num_tokens": 3356118.0,
      "step": 205
    },
    {
      "entropy": 0.5498729795217514,
      "epoch": 0.7686567164179104,
      "grad_norm": 0.14498212933540344,
      "learning_rate": 0.0002,
      "loss": 0.5578135251998901,
      "mean_token_accuracy": 0.7715203762054443,
      "num_tokens": 3372313.0,
      "step": 206
    },
    {
      "entropy": 0.518561378121376,
      "epoch": 0.7723880597014925,
      "grad_norm": 0.17109735310077667,
      "learning_rate": 0.0002,
      "loss": 0.5355443954467773,
      "mean_token_accuracy": 0.7833957225084305,
      "num_tokens": 3388823.0,
      "step": 207
    },
    {
      "entropy": 0.5251610577106476,
      "epoch": 0.7761194029850746,
      "grad_norm": 0.1357729136943817,
      "learning_rate": 0.0002,
      "loss": 0.5393021106719971,
      "mean_token_accuracy": 0.7840308547019958,
      "num_tokens": 3405230.0,
      "step": 208
    },
    {
      "entropy": 0.5503880083560944,
      "epoch": 0.7798507462686567,
      "grad_norm": 0.1302410066127777,
      "learning_rate": 0.0002,
      "loss": 0.5439841151237488,
      "mean_token_accuracy": 0.7808199971914291,
      "num_tokens": 3421538.0,
      "step": 209
    },
    {
      "entropy": 0.5504791587591171,
      "epoch": 0.7835820895522388,
      "grad_norm": 0.17588983476161957,
      "learning_rate": 0.0002,
      "loss": 0.5450472831726074,
      "mean_token_accuracy": 0.7778802812099457,
      "num_tokens": 3437751.0,
      "step": 210
    },
    {
      "entropy": 0.5515212118625641,
      "epoch": 0.7873134328358209,
      "grad_norm": 0.12591375410556793,
      "learning_rate": 0.0002,
      "loss": 0.5451067686080933,
      "mean_token_accuracy": 0.7791963666677475,
      "num_tokens": 3453958.0,
      "step": 211
    },
    {
      "entropy": 0.5373163223266602,
      "epoch": 0.7910447761194029,
      "grad_norm": 0.11395501345396042,
      "learning_rate": 0.0002,
      "loss": 0.5358302593231201,
      "mean_token_accuracy": 0.7810690551996231,
      "num_tokens": 3470360.0,
      "step": 212
    },
    {
      "entropy": 0.5450779050588608,
      "epoch": 0.7947761194029851,
      "grad_norm": 0.13792671263217926,
      "learning_rate": 0.0002,
      "loss": 0.544927179813385,
      "mean_token_accuracy": 0.7764876931905746,
      "num_tokens": 3486976.0,
      "step": 213
    },
    {
      "entropy": 0.542982280254364,
      "epoch": 0.7985074626865671,
      "grad_norm": 0.14203859865665436,
      "learning_rate": 0.0002,
      "loss": 0.5473698377609253,
      "mean_token_accuracy": 0.7803648114204407,
      "num_tokens": 3503108.0,
      "step": 214
    },
    {
      "entropy": 0.5468839257955551,
      "epoch": 0.8022388059701493,
      "grad_norm": 0.14546480774879456,
      "learning_rate": 0.0002,
      "loss": 0.5536583065986633,
      "mean_token_accuracy": 0.7759220898151398,
      "num_tokens": 3519438.0,
      "step": 215
    },
    {
      "entropy": 0.5323249697685242,
      "epoch": 0.8059701492537313,
      "grad_norm": 0.12205041199922562,
      "learning_rate": 0.0002,
      "loss": 0.5366013050079346,
      "mean_token_accuracy": 0.7834390252828598,
      "num_tokens": 3535744.0,
      "step": 216
    },
    {
      "entropy": 0.5549503713846207,
      "epoch": 0.8097014925373134,
      "grad_norm": 0.12942132353782654,
      "learning_rate": 0.0002,
      "loss": 0.5488545298576355,
      "mean_token_accuracy": 0.7754624336957932,
      "num_tokens": 3552229.0,
      "step": 217
    },
    {
      "entropy": 0.5418603718280792,
      "epoch": 0.8134328358208955,
      "grad_norm": 0.11578696221113205,
      "learning_rate": 0.0002,
      "loss": 0.5355917811393738,
      "mean_token_accuracy": 0.7838476449251175,
      "num_tokens": 3568579.0,
      "step": 218
    },
    {
      "entropy": 0.5710959583520889,
      "epoch": 0.8171641791044776,
      "grad_norm": 0.1490243822336197,
      "learning_rate": 0.0002,
      "loss": 0.5620180368423462,
      "mean_token_accuracy": 0.774310976266861,
      "num_tokens": 3585243.0,
      "step": 219
    },
    {
      "entropy": 0.5469939708709717,
      "epoch": 0.8208955223880597,
      "grad_norm": 0.12220045179128647,
      "learning_rate": 0.0002,
      "loss": 0.5480121374130249,
      "mean_token_accuracy": 0.7782059013843536,
      "num_tokens": 3601764.0,
      "step": 220
    },
    {
      "entropy": 0.5289393365383148,
      "epoch": 0.8246268656716418,
      "grad_norm": 0.1687396615743637,
      "learning_rate": 0.0002,
      "loss": 0.5451465249061584,
      "mean_token_accuracy": 0.7791816592216492,
      "num_tokens": 3618085.0,
      "step": 221
    },
    {
      "entropy": 0.5347074121236801,
      "epoch": 0.8283582089552238,
      "grad_norm": 0.13434253633022308,
      "learning_rate": 0.0002,
      "loss": 0.542661190032959,
      "mean_token_accuracy": 0.7783772945404053,
      "num_tokens": 3634378.0,
      "step": 222
    },
    {
      "entropy": 0.5412849485874176,
      "epoch": 0.832089552238806,
      "grad_norm": 0.11510826647281647,
      "learning_rate": 0.0002,
      "loss": 0.5439392328262329,
      "mean_token_accuracy": 0.7777635306119919,
      "num_tokens": 3650552.0,
      "step": 223
    },
    {
      "entropy": 0.5579446852207184,
      "epoch": 0.835820895522388,
      "grad_norm": 0.14014939963817596,
      "learning_rate": 0.0002,
      "loss": 0.5538932681083679,
      "mean_token_accuracy": 0.7754973918199539,
      "num_tokens": 3666791.0,
      "step": 224
    },
    {
      "entropy": 0.5415721386671066,
      "epoch": 0.8395522388059702,
      "grad_norm": 0.13600188493728638,
      "learning_rate": 0.0002,
      "loss": 0.5392742156982422,
      "mean_token_accuracy": 0.7821011990308762,
      "num_tokens": 3682947.0,
      "step": 225
    },
    {
      "entropy": 0.5572647303342819,
      "epoch": 0.8432835820895522,
      "grad_norm": 0.10934413224458694,
      "learning_rate": 0.0002,
      "loss": 0.5546595454216003,
      "mean_token_accuracy": 0.7696610689163208,
      "num_tokens": 3699234.0,
      "step": 226
    },
    {
      "entropy": 0.5482315272092819,
      "epoch": 0.8470149253731343,
      "grad_norm": 0.12785379588603973,
      "learning_rate": 0.0002,
      "loss": 0.5524696707725525,
      "mean_token_accuracy": 0.7756107896566391,
      "num_tokens": 3715642.0,
      "step": 227
    },
    {
      "entropy": 0.5469722747802734,
      "epoch": 0.8507462686567164,
      "grad_norm": 0.17460216581821442,
      "learning_rate": 0.0002,
      "loss": 0.5504593849182129,
      "mean_token_accuracy": 0.7762730270624161,
      "num_tokens": 3732150.0,
      "step": 228
    },
    {
      "entropy": 0.5381356775760651,
      "epoch": 0.8544776119402985,
      "grad_norm": 0.13738028705120087,
      "learning_rate": 0.0002,
      "loss": 0.5410696268081665,
      "mean_token_accuracy": 0.7804866135120392,
      "num_tokens": 3748356.0,
      "step": 229
    },
    {
      "entropy": 0.5419336557388306,
      "epoch": 0.8582089552238806,
      "grad_norm": 0.1366986632347107,
      "learning_rate": 0.0002,
      "loss": 0.54702228307724,
      "mean_token_accuracy": 0.7779697477817535,
      "num_tokens": 3764864.0,
      "step": 230
    },
    {
      "entropy": 0.5650088787078857,
      "epoch": 0.8619402985074627,
      "grad_norm": 0.1245863139629364,
      "learning_rate": 0.0002,
      "loss": 0.5644104480743408,
      "mean_token_accuracy": 0.7698388695716858,
      "num_tokens": 3781158.0,
      "step": 231
    },
    {
      "entropy": 0.5491328537464142,
      "epoch": 0.8656716417910447,
      "grad_norm": 0.12666523456573486,
      "learning_rate": 0.0002,
      "loss": 0.547432005405426,
      "mean_token_accuracy": 0.7800118178129196,
      "num_tokens": 3797547.0,
      "step": 232
    },
    {
      "entropy": 0.5397607833147049,
      "epoch": 0.8694029850746269,
      "grad_norm": 0.13737380504608154,
      "learning_rate": 0.0002,
      "loss": 0.5392321348190308,
      "mean_token_accuracy": 0.7841152101755142,
      "num_tokens": 3813843.0,
      "step": 233
    },
    {
      "entropy": 0.544382706284523,
      "epoch": 0.8731343283582089,
      "grad_norm": 0.11425275355577469,
      "learning_rate": 0.0002,
      "loss": 0.5442847609519958,
      "mean_token_accuracy": 0.7817698568105698,
      "num_tokens": 3830087.0,
      "step": 234
    },
    {
      "entropy": 0.5501336306333542,
      "epoch": 0.8768656716417911,
      "grad_norm": 0.11622001230716705,
      "learning_rate": 0.0002,
      "loss": 0.5528742074966431,
      "mean_token_accuracy": 0.7744111120700836,
      "num_tokens": 3846206.0,
      "step": 235
    },
    {
      "entropy": 0.5614665448665619,
      "epoch": 0.8805970149253731,
      "grad_norm": 0.13408295810222626,
      "learning_rate": 0.0002,
      "loss": 0.5573604106903076,
      "mean_token_accuracy": 0.7725346237421036,
      "num_tokens": 3862800.0,
      "step": 236
    },
    {
      "entropy": 0.5298355966806412,
      "epoch": 0.8843283582089553,
      "grad_norm": 0.12251684814691544,
      "learning_rate": 0.0002,
      "loss": 0.5347483158111572,
      "mean_token_accuracy": 0.7849987000226974,
      "num_tokens": 3879053.0,
      "step": 237
    },
    {
      "entropy": 0.527460902929306,
      "epoch": 0.8880597014925373,
      "grad_norm": 0.1435721516609192,
      "learning_rate": 0.0002,
      "loss": 0.5373647809028625,
      "mean_token_accuracy": 0.7806102335453033,
      "num_tokens": 3895482.0,
      "step": 238
    },
    {
      "entropy": 0.5427086651325226,
      "epoch": 0.8917910447761194,
      "grad_norm": 0.15594077110290527,
      "learning_rate": 0.0002,
      "loss": 0.5516656637191772,
      "mean_token_accuracy": 0.77671217918396,
      "num_tokens": 3911639.0,
      "step": 239
    },
    {
      "entropy": 0.5452292412519455,
      "epoch": 0.8955223880597015,
      "grad_norm": 0.13117194175720215,
      "learning_rate": 0.0002,
      "loss": 0.547135055065155,
      "mean_token_accuracy": 0.7797868549823761,
      "num_tokens": 3927678.0,
      "step": 240
    },
    {
      "entropy": 0.5607290267944336,
      "epoch": 0.8992537313432836,
      "grad_norm": 0.1290806084871292,
      "learning_rate": 0.0002,
      "loss": 0.5505275726318359,
      "mean_token_accuracy": 0.7754765301942825,
      "num_tokens": 3944068.0,
      "step": 241
    },
    {
      "entropy": 0.5452372431755066,
      "epoch": 0.9029850746268657,
      "grad_norm": 0.1041363850235939,
      "learning_rate": 0.0002,
      "loss": 0.5423585772514343,
      "mean_token_accuracy": 0.7772356569766998,
      "num_tokens": 3960455.0,
      "step": 242
    },
    {
      "entropy": 0.5599673539400101,
      "epoch": 0.9067164179104478,
      "grad_norm": 0.12637630105018616,
      "learning_rate": 0.0002,
      "loss": 0.5576352477073669,
      "mean_token_accuracy": 0.7743173092603683,
      "num_tokens": 3976930.0,
      "step": 243
    },
    {
      "entropy": 0.5532275587320328,
      "epoch": 0.9104477611940298,
      "grad_norm": 0.1246461272239685,
      "learning_rate": 0.0002,
      "loss": 0.5524167418479919,
      "mean_token_accuracy": 0.7746891975402832,
      "num_tokens": 3993200.0,
      "step": 244
    },
    {
      "entropy": 0.5353778898715973,
      "epoch": 0.914179104477612,
      "grad_norm": 0.1204264834523201,
      "learning_rate": 0.0002,
      "loss": 0.5398914217948914,
      "mean_token_accuracy": 0.780600368976593,
      "num_tokens": 4009579.0,
      "step": 245
    },
    {
      "entropy": 0.5454006344079971,
      "epoch": 0.917910447761194,
      "grad_norm": 0.14190614223480225,
      "learning_rate": 0.0002,
      "loss": 0.5499029159545898,
      "mean_token_accuracy": 0.7774828523397446,
      "num_tokens": 4025835.0,
      "step": 246
    },
    {
      "entropy": 0.5335248857736588,
      "epoch": 0.9216417910447762,
      "grad_norm": 0.12253561615943909,
      "learning_rate": 0.0002,
      "loss": 0.531761884689331,
      "mean_token_accuracy": 0.782997652888298,
      "num_tokens": 4041832.0,
      "step": 247
    },
    {
      "entropy": 0.5199212953448296,
      "epoch": 0.9253731343283582,
      "grad_norm": 0.12186253070831299,
      "learning_rate": 0.0002,
      "loss": 0.5216535329818726,
      "mean_token_accuracy": 0.788480818271637,
      "num_tokens": 4057892.0,
      "step": 248
    },
    {
      "entropy": 0.5205824747681618,
      "epoch": 0.9291044776119403,
      "grad_norm": 0.1254664659500122,
      "learning_rate": 0.0002,
      "loss": 0.5266774296760559,
      "mean_token_accuracy": 0.7851553857326508,
      "num_tokens": 4074337.0,
      "step": 249
    },
    {
      "entropy": 0.5461059212684631,
      "epoch": 0.9328358208955224,
      "grad_norm": 0.12566933035850525,
      "learning_rate": 0.0002,
      "loss": 0.5446832776069641,
      "mean_token_accuracy": 0.7797155976295471,
      "num_tokens": 4090894.0,
      "step": 250
    },
    {
      "entropy": 0.5468244701623917,
      "epoch": 0.9365671641791045,
      "grad_norm": 0.1111629381775856,
      "learning_rate": 0.0002,
      "loss": 0.5508494973182678,
      "mean_token_accuracy": 0.7770780026912689,
      "num_tokens": 4107223.0,
      "step": 251
    },
    {
      "entropy": 0.5277251750230789,
      "epoch": 0.9402985074626866,
      "grad_norm": 0.12095754593610764,
      "learning_rate": 0.0002,
      "loss": 0.5359346866607666,
      "mean_token_accuracy": 0.7836727201938629,
      "num_tokens": 4123421.0,
      "step": 252
    },
    {
      "entropy": 0.5456234216690063,
      "epoch": 0.9440298507462687,
      "grad_norm": 0.15432782471179962,
      "learning_rate": 0.0002,
      "loss": 0.5424333810806274,
      "mean_token_accuracy": 0.7781681269407272,
      "num_tokens": 4139814.0,
      "step": 253
    },
    {
      "entropy": 0.561904102563858,
      "epoch": 0.9477611940298507,
      "grad_norm": 0.11046113818883896,
      "learning_rate": 0.0002,
      "loss": 0.555654764175415,
      "mean_token_accuracy": 0.77543506026268,
      "num_tokens": 4156145.0,
      "step": 254
    },
    {
      "entropy": 0.550317719578743,
      "epoch": 0.9514925373134329,
      "grad_norm": 0.11810830980539322,
      "learning_rate": 0.0002,
      "loss": 0.551179051399231,
      "mean_token_accuracy": 0.7750379741191864,
      "num_tokens": 4172486.0,
      "step": 255
    },
    {
      "entropy": 0.5419327318668365,
      "epoch": 0.9552238805970149,
      "grad_norm": 0.12468105554580688,
      "learning_rate": 0.0002,
      "loss": 0.5484491586685181,
      "mean_token_accuracy": 0.7767989039421082,
      "num_tokens": 4188789.0,
      "step": 256
    },
    {
      "entropy": 0.5456331819295883,
      "epoch": 0.9589552238805971,
      "grad_norm": 0.10254428535699844,
      "learning_rate": 0.0002,
      "loss": 0.5463854074478149,
      "mean_token_accuracy": 0.7775218188762665,
      "num_tokens": 4205039.0,
      "step": 257
    },
    {
      "entropy": 0.5631006211042404,
      "epoch": 0.9626865671641791,
      "grad_norm": 0.11667738854885101,
      "learning_rate": 0.0002,
      "loss": 0.5608720183372498,
      "mean_token_accuracy": 0.770975336432457,
      "num_tokens": 4221737.0,
      "step": 258
    },
    {
      "entropy": 0.5438289195299149,
      "epoch": 0.9664179104477612,
      "grad_norm": 0.10653994977474213,
      "learning_rate": 0.0002,
      "loss": 0.5415982007980347,
      "mean_token_accuracy": 0.7795556634664536,
      "num_tokens": 4238154.0,
      "step": 259
    },
    {
      "entropy": 0.54365074634552,
      "epoch": 0.9701492537313433,
      "grad_norm": 0.10474390536546707,
      "learning_rate": 0.0002,
      "loss": 0.5424013137817383,
      "mean_token_accuracy": 0.7802460342645645,
      "num_tokens": 4254544.0,
      "step": 260
    },
    {
      "entropy": 0.5336029231548309,
      "epoch": 0.9738805970149254,
      "grad_norm": 0.10472694039344788,
      "learning_rate": 0.0002,
      "loss": 0.5306434631347656,
      "mean_token_accuracy": 0.786760687828064,
      "num_tokens": 4270769.0,
      "step": 261
    },
    {
      "entropy": 0.5328535884618759,
      "epoch": 0.9776119402985075,
      "grad_norm": 0.12627209722995758,
      "learning_rate": 0.0002,
      "loss": 0.5387364625930786,
      "mean_token_accuracy": 0.7820819616317749,
      "num_tokens": 4287087.0,
      "step": 262
    },
    {
      "entropy": 0.5569794625043869,
      "epoch": 0.9813432835820896,
      "grad_norm": 0.12556159496307373,
      "learning_rate": 0.0002,
      "loss": 0.5633010268211365,
      "mean_token_accuracy": 0.773500457406044,
      "num_tokens": 4303275.0,
      "step": 263
    },
    {
      "entropy": 0.5476076602935791,
      "epoch": 0.9850746268656716,
      "grad_norm": 0.11449912190437317,
      "learning_rate": 0.0002,
      "loss": 0.5478817224502563,
      "mean_token_accuracy": 0.7760190218687057,
      "num_tokens": 4319447.0,
      "step": 264
    },
    {
      "entropy": 0.5338123887777328,
      "epoch": 0.9888059701492538,
      "grad_norm": 0.11093062907457352,
      "learning_rate": 0.0002,
      "loss": 0.5374420881271362,
      "mean_token_accuracy": 0.7827984094619751,
      "num_tokens": 4335610.0,
      "step": 265
    },
    {
      "entropy": 0.5231072157621384,
      "epoch": 0.9925373134328358,
      "grad_norm": 0.11385290324687958,
      "learning_rate": 0.0002,
      "loss": 0.518035352230072,
      "mean_token_accuracy": 0.7871184349060059,
      "num_tokens": 4352006.0,
      "step": 266
    },
    {
      "entropy": 0.5597157925367355,
      "epoch": 0.996268656716418,
      "grad_norm": 0.12694767117500305,
      "learning_rate": 0.0002,
      "loss": 0.5584501624107361,
      "mean_token_accuracy": 0.7767712473869324,
      "num_tokens": 4368643.0,
      "step": 267
    },
    {
      "entropy": 0.5244968682527542,
      "epoch": 1.0,
      "grad_norm": 0.11100418865680695,
      "learning_rate": 0.0002,
      "loss": 0.5219011306762695,
      "mean_token_accuracy": 0.7878229916095734,
      "num_tokens": 4385170.0,
      "step": 268
    },
    {
      "entropy": 0.5507489442825317,
      "epoch": 1.0037313432835822,
      "grad_norm": 0.11917725205421448,
      "learning_rate": 0.0002,
      "loss": 0.5477783679962158,
      "mean_token_accuracy": 0.7768960893154144,
      "num_tokens": 4401649.0,
      "step": 269
    },
    {
      "entropy": 0.5179261267185211,
      "epoch": 1.007462686567164,
      "grad_norm": 0.114885114133358,
      "learning_rate": 0.0002,
      "loss": 0.5203835964202881,
      "mean_token_accuracy": 0.7879552990198135,
      "num_tokens": 4417657.0,
      "step": 270
    },
    {
      "entropy": 0.5413845926523209,
      "epoch": 1.0111940298507462,
      "grad_norm": 0.11034537851810455,
      "learning_rate": 0.0002,
      "loss": 0.5424505472183228,
      "mean_token_accuracy": 0.780542716383934,
      "num_tokens": 4433808.0,
      "step": 271
    },
    {
      "entropy": 0.5339591354131699,
      "epoch": 1.0149253731343284,
      "grad_norm": 0.10123753547668457,
      "learning_rate": 0.0002,
      "loss": 0.5324500203132629,
      "mean_token_accuracy": 0.7834791243076324,
      "num_tokens": 4450074.0,
      "step": 272
    },
    {
      "entropy": 0.5246681123971939,
      "epoch": 1.0186567164179103,
      "grad_norm": 0.11839871853590012,
      "learning_rate": 0.0002,
      "loss": 0.530142605304718,
      "mean_token_accuracy": 0.7845751047134399,
      "num_tokens": 4466501.0,
      "step": 273
    },
    {
      "entropy": 0.5286023616790771,
      "epoch": 1.0223880597014925,
      "grad_norm": 0.12706530094146729,
      "learning_rate": 0.0002,
      "loss": 0.5256507396697998,
      "mean_token_accuracy": 0.7853004187345505,
      "num_tokens": 4482854.0,
      "step": 274
    },
    {
      "entropy": 0.5193610340356827,
      "epoch": 1.0261194029850746,
      "grad_norm": 0.1515689492225647,
      "learning_rate": 0.0002,
      "loss": 0.5248588919639587,
      "mean_token_accuracy": 0.7882117629051208,
      "num_tokens": 4499325.0,
      "step": 275
    },
    {
      "entropy": 0.5320159494876862,
      "epoch": 1.0298507462686568,
      "grad_norm": 0.13122239708900452,
      "learning_rate": 0.0002,
      "loss": 0.534252941608429,
      "mean_token_accuracy": 0.7805082648992538,
      "num_tokens": 4515563.0,
      "step": 276
    },
    {
      "entropy": 0.5442035794258118,
      "epoch": 1.0335820895522387,
      "grad_norm": 0.12928421795368195,
      "learning_rate": 0.0002,
      "loss": 0.5358187556266785,
      "mean_token_accuracy": 0.7840018421411514,
      "num_tokens": 4531822.0,
      "step": 277
    },
    {
      "entropy": 0.5246908813714981,
      "epoch": 1.037313432835821,
      "grad_norm": 0.14177928864955902,
      "learning_rate": 0.0002,
      "loss": 0.522698163986206,
      "mean_token_accuracy": 0.7914385795593262,
      "num_tokens": 4548179.0,
      "step": 278
    },
    {
      "entropy": 0.5337726771831512,
      "epoch": 1.041044776119403,
      "grad_norm": 0.12917082011699677,
      "learning_rate": 0.0002,
      "loss": 0.5394143462181091,
      "mean_token_accuracy": 0.7827410697937012,
      "num_tokens": 4564544.0,
      "step": 279
    },
    {
      "entropy": 0.5249736607074738,
      "epoch": 1.044776119402985,
      "grad_norm": 0.15507692098617554,
      "learning_rate": 0.0002,
      "loss": 0.530266523361206,
      "mean_token_accuracy": 0.784500926733017,
      "num_tokens": 4580899.0,
      "step": 280
    },
    {
      "entropy": 0.5224004536867142,
      "epoch": 1.0485074626865671,
      "grad_norm": 0.13328292965888977,
      "learning_rate": 0.0002,
      "loss": 0.5272551774978638,
      "mean_token_accuracy": 0.7841868251562119,
      "num_tokens": 4597148.0,
      "step": 281
    },
    {
      "entropy": 0.5343264937400818,
      "epoch": 1.0522388059701493,
      "grad_norm": 0.12672054767608643,
      "learning_rate": 0.0002,
      "loss": 0.5322040319442749,
      "mean_token_accuracy": 0.7872680425643921,
      "num_tokens": 4613527.0,
      "step": 282
    },
    {
      "entropy": 0.5167456269264221,
      "epoch": 1.0559701492537314,
      "grad_norm": 0.1300593614578247,
      "learning_rate": 0.0002,
      "loss": 0.5081961750984192,
      "mean_token_accuracy": 0.796296164393425,
      "num_tokens": 4629820.0,
      "step": 283
    },
    {
      "entropy": 0.5502501279115677,
      "epoch": 1.0597014925373134,
      "grad_norm": 0.11848345398902893,
      "learning_rate": 0.0002,
      "loss": 0.5440031886100769,
      "mean_token_accuracy": 0.7831816971302032,
      "num_tokens": 4646409.0,
      "step": 284
    },
    {
      "entropy": 0.537413626909256,
      "epoch": 1.0634328358208955,
      "grad_norm": 0.1294218748807907,
      "learning_rate": 0.0002,
      "loss": 0.5396757125854492,
      "mean_token_accuracy": 0.7809827625751495,
      "num_tokens": 4662669.0,
      "step": 285
    },
    {
      "entropy": 0.5442153364419937,
      "epoch": 1.0671641791044777,
      "grad_norm": 0.1345270872116089,
      "learning_rate": 0.0002,
      "loss": 0.5449458956718445,
      "mean_token_accuracy": 0.7765765637159348,
      "num_tokens": 4679238.0,
      "step": 286
    },
    {
      "entropy": 0.5223373919725418,
      "epoch": 1.0708955223880596,
      "grad_norm": 0.13911806046962738,
      "learning_rate": 0.0002,
      "loss": 0.527599036693573,
      "mean_token_accuracy": 0.7872483432292938,
      "num_tokens": 4695791.0,
      "step": 287
    },
    {
      "entropy": 0.5344910472631454,
      "epoch": 1.0746268656716418,
      "grad_norm": 0.12189274281263351,
      "learning_rate": 0.0002,
      "loss": 0.5306644439697266,
      "mean_token_accuracy": 0.7878366857767105,
      "num_tokens": 4711915.0,
      "step": 288
    },
    {
      "entropy": 0.5503940731287003,
      "epoch": 1.078358208955224,
      "grad_norm": 0.12994466722011566,
      "learning_rate": 0.0002,
      "loss": 0.5426635146141052,
      "mean_token_accuracy": 0.7800120115280151,
      "num_tokens": 4728355.0,
      "step": 289
    },
    {
      "entropy": 0.5371982753276825,
      "epoch": 1.0820895522388059,
      "grad_norm": 0.11566822230815887,
      "learning_rate": 0.0002,
      "loss": 0.5357840061187744,
      "mean_token_accuracy": 0.7830845862627029,
      "num_tokens": 4744985.0,
      "step": 290
    },
    {
      "entropy": 0.5363182723522186,
      "epoch": 1.085820895522388,
      "grad_norm": 0.14267492294311523,
      "learning_rate": 0.0002,
      "loss": 0.547325849533081,
      "mean_token_accuracy": 0.778549462556839,
      "num_tokens": 4761335.0,
      "step": 291
    },
    {
      "entropy": 0.5316378027200699,
      "epoch": 1.0895522388059702,
      "grad_norm": 0.14865896105766296,
      "learning_rate": 0.0002,
      "loss": 0.5413302183151245,
      "mean_token_accuracy": 0.7811136245727539,
      "num_tokens": 4777695.0,
      "step": 292
    },
    {
      "entropy": 0.5230907201766968,
      "epoch": 1.0932835820895523,
      "grad_norm": 0.11756227165460587,
      "learning_rate": 0.0002,
      "loss": 0.5194687843322754,
      "mean_token_accuracy": 0.7905423790216446,
      "num_tokens": 4794345.0,
      "step": 293
    },
    {
      "entropy": 0.5252106562256813,
      "epoch": 1.0970149253731343,
      "grad_norm": 0.12259907275438309,
      "learning_rate": 0.0002,
      "loss": 0.5216252207756042,
      "mean_token_accuracy": 0.7896531373262405,
      "num_tokens": 4810734.0,
      "step": 294
    },
    {
      "entropy": 0.5349573045969009,
      "epoch": 1.1007462686567164,
      "grad_norm": 0.14738821983337402,
      "learning_rate": 0.0002,
      "loss": 0.5411865711212158,
      "mean_token_accuracy": 0.7810265272855759,
      "num_tokens": 4827208.0,
      "step": 295
    },
    {
      "entropy": 0.5216178447008133,
      "epoch": 1.1044776119402986,
      "grad_norm": 0.1261540949344635,
      "learning_rate": 0.0002,
      "loss": 0.530312180519104,
      "mean_token_accuracy": 0.786735400557518,
      "num_tokens": 4843467.0,
      "step": 296
    },
    {
      "entropy": 0.5475742220878601,
      "epoch": 1.1082089552238805,
      "grad_norm": 0.17317992448806763,
      "learning_rate": 0.0002,
      "loss": 0.5447714328765869,
      "mean_token_accuracy": 0.7785773426294327,
      "num_tokens": 4859845.0,
      "step": 297
    },
    {
      "entropy": 0.5403448045253754,
      "epoch": 1.1119402985074627,
      "grad_norm": 0.11798793077468872,
      "learning_rate": 0.0002,
      "loss": 0.5318464040756226,
      "mean_token_accuracy": 0.7843561172485352,
      "num_tokens": 4876407.0,
      "step": 298
    },
    {
      "entropy": 0.5377328842878342,
      "epoch": 1.1156716417910448,
      "grad_norm": 0.13185109198093414,
      "learning_rate": 0.0002,
      "loss": 0.5353187918663025,
      "mean_token_accuracy": 0.7843449413776398,
      "num_tokens": 4892742.0,
      "step": 299
    },
    {
      "entropy": 0.5279147699475288,
      "epoch": 1.1194029850746268,
      "grad_norm": 0.16606053709983826,
      "learning_rate": 0.0002,
      "loss": 0.5365189909934998,
      "mean_token_accuracy": 0.783987894654274,
      "num_tokens": 4909223.0,
      "step": 300
    },
    {
      "entropy": 0.5365699529647827,
      "epoch": 1.123134328358209,
      "grad_norm": 0.160949245095253,
      "learning_rate": 0.0002,
      "loss": 0.5369877815246582,
      "mean_token_accuracy": 0.7800540775060654,
      "num_tokens": 4925547.0,
      "step": 301
    },
    {
      "entropy": 0.5412300229072571,
      "epoch": 1.126865671641791,
      "grad_norm": 0.13560807704925537,
      "learning_rate": 0.0002,
      "loss": 0.5425742864608765,
      "mean_token_accuracy": 0.7801165878772736,
      "num_tokens": 4941811.0,
      "step": 302
    },
    {
      "entropy": 0.5323121100664139,
      "epoch": 1.1305970149253732,
      "grad_norm": 0.1241421177983284,
      "learning_rate": 0.0002,
      "loss": 0.5286750793457031,
      "mean_token_accuracy": 0.7866266220808029,
      "num_tokens": 4958292.0,
      "step": 303
    },
    {
      "entropy": 0.533287987112999,
      "epoch": 1.1343283582089552,
      "grad_norm": 0.12874163687229156,
      "learning_rate": 0.0002,
      "loss": 0.5292162895202637,
      "mean_token_accuracy": 0.7868117541074753,
      "num_tokens": 4974588.0,
      "step": 304
    },
    {
      "entropy": 0.5256194174289703,
      "epoch": 1.1380597014925373,
      "grad_norm": 0.1257098764181137,
      "learning_rate": 0.0002,
      "loss": 0.5303645730018616,
      "mean_token_accuracy": 0.7867012023925781,
      "num_tokens": 4990937.0,
      "step": 305
    },
    {
      "entropy": 0.5340888798236847,
      "epoch": 1.1417910447761195,
      "grad_norm": 0.1419389545917511,
      "learning_rate": 0.0002,
      "loss": 0.5442762970924377,
      "mean_token_accuracy": 0.778327003121376,
      "num_tokens": 5007117.0,
      "step": 306
    },
    {
      "entropy": 0.5182383954524994,
      "epoch": 1.1455223880597014,
      "grad_norm": 0.13628648221492767,
      "learning_rate": 0.0002,
      "loss": 0.5177187323570251,
      "mean_token_accuracy": 0.7890002727508545,
      "num_tokens": 5023154.0,
      "step": 307
    },
    {
      "entropy": 0.5281811505556107,
      "epoch": 1.1492537313432836,
      "grad_norm": 0.13477671146392822,
      "learning_rate": 0.0002,
      "loss": 0.5306061506271362,
      "mean_token_accuracy": 0.7834227383136749,
      "num_tokens": 5039244.0,
      "step": 308
    },
    {
      "entropy": 0.526213601231575,
      "epoch": 1.1529850746268657,
      "grad_norm": 0.12357262521982193,
      "learning_rate": 0.0002,
      "loss": 0.5250381231307983,
      "mean_token_accuracy": 0.7864978313446045,
      "num_tokens": 5055638.0,
      "step": 309
    },
    {
      "entropy": 0.5203096121549606,
      "epoch": 1.1567164179104479,
      "grad_norm": 0.14019498229026794,
      "learning_rate": 0.0002,
      "loss": 0.5250931978225708,
      "mean_token_accuracy": 0.7855905443429947,
      "num_tokens": 5072051.0,
      "step": 310
    },
    {
      "entropy": 0.5365833044052124,
      "epoch": 1.1604477611940298,
      "grad_norm": 0.11511947959661484,
      "learning_rate": 0.0002,
      "loss": 0.5366431474685669,
      "mean_token_accuracy": 0.7812765091657639,
      "num_tokens": 5088455.0,
      "step": 311
    },
    {
      "entropy": 0.549386665225029,
      "epoch": 1.164179104477612,
      "grad_norm": 0.22190974652767181,
      "learning_rate": 0.0002,
      "loss": 0.5439768433570862,
      "mean_token_accuracy": 0.7783290147781372,
      "num_tokens": 5104840.0,
      "step": 312
    },
    {
      "entropy": 0.5462167263031006,
      "epoch": 1.1679104477611941,
      "grad_norm": 0.12778188288211823,
      "learning_rate": 0.0002,
      "loss": 0.543470025062561,
      "mean_token_accuracy": 0.7785560786724091,
      "num_tokens": 5121470.0,
      "step": 313
    },
    {
      "entropy": 0.5416186302900314,
      "epoch": 1.171641791044776,
      "grad_norm": 0.14187116920948029,
      "learning_rate": 0.0002,
      "loss": 0.5464950799942017,
      "mean_token_accuracy": 0.7785798162221909,
      "num_tokens": 5137875.0,
      "step": 314
    },
    {
      "entropy": 0.5286829024553299,
      "epoch": 1.1753731343283582,
      "grad_norm": 0.13838821649551392,
      "learning_rate": 0.0002,
      "loss": 0.5358678102493286,
      "mean_token_accuracy": 0.7815852910280228,
      "num_tokens": 5154241.0,
      "step": 315
    },
    {
      "entropy": 0.5266743376851082,
      "epoch": 1.1791044776119404,
      "grad_norm": 0.147015780210495,
      "learning_rate": 0.0002,
      "loss": 0.5362037420272827,
      "mean_token_accuracy": 0.7841357439756393,
      "num_tokens": 5170521.0,
      "step": 316
    },
    {
      "entropy": 0.5310831069946289,
      "epoch": 1.1828358208955223,
      "grad_norm": 0.14254894852638245,
      "learning_rate": 0.0002,
      "loss": 0.5336026549339294,
      "mean_token_accuracy": 0.7839198410511017,
      "num_tokens": 5186829.0,
      "step": 317
    },
    {
      "entropy": 0.5473271459341049,
      "epoch": 1.1865671641791045,
      "grad_norm": 0.1311594694852829,
      "learning_rate": 0.0002,
      "loss": 0.544696569442749,
      "mean_token_accuracy": 0.7774120271205902,
      "num_tokens": 5203232.0,
      "step": 318
    },
    {
      "entropy": 0.5363661348819733,
      "epoch": 1.1902985074626866,
      "grad_norm": 0.1488485336303711,
      "learning_rate": 0.0002,
      "loss": 0.528728187084198,
      "mean_token_accuracy": 0.7841808795928955,
      "num_tokens": 5219571.0,
      "step": 319
    },
    {
      "entropy": 0.5424627363681793,
      "epoch": 1.1940298507462686,
      "grad_norm": 0.1739611029624939,
      "learning_rate": 0.0002,
      "loss": 0.5434719920158386,
      "mean_token_accuracy": 0.7807931751012802,
      "num_tokens": 5235639.0,
      "step": 320
    },
    {
      "entropy": 0.5289300680160522,
      "epoch": 1.1977611940298507,
      "grad_norm": 0.16747383773326874,
      "learning_rate": 0.0002,
      "loss": 0.5332000851631165,
      "mean_token_accuracy": 0.7818503081798553,
      "num_tokens": 5252171.0,
      "step": 321
    },
    {
      "entropy": 0.501720704138279,
      "epoch": 1.2014925373134329,
      "grad_norm": 0.14918096363544464,
      "learning_rate": 0.0002,
      "loss": 0.5059674978256226,
      "mean_token_accuracy": 0.7958301901817322,
      "num_tokens": 5268541.0,
      "step": 322
    },
    {
      "entropy": 0.5355239808559418,
      "epoch": 1.205223880597015,
      "grad_norm": 0.13879424333572388,
      "learning_rate": 0.0002,
      "loss": 0.5362867116928101,
      "mean_token_accuracy": 0.7825663089752197,
      "num_tokens": 5285009.0,
      "step": 323
    },
    {
      "entropy": 0.5333442091941833,
      "epoch": 1.208955223880597,
      "grad_norm": 0.14377108216285706,
      "learning_rate": 0.0002,
      "loss": 0.5346744060516357,
      "mean_token_accuracy": 0.7837191075086594,
      "num_tokens": 5301110.0,
      "step": 324
    },
    {
      "entropy": 0.5217743068933487,
      "epoch": 1.212686567164179,
      "grad_norm": 0.14959798753261566,
      "learning_rate": 0.0002,
      "loss": 0.5172246098518372,
      "mean_token_accuracy": 0.7899685055017471,
      "num_tokens": 5317468.0,
      "step": 325
    },
    {
      "entropy": 0.5328510701656342,
      "epoch": 1.2164179104477613,
      "grad_norm": 0.12470122426748276,
      "learning_rate": 0.0002,
      "loss": 0.5278521776199341,
      "mean_token_accuracy": 0.7836970090866089,
      "num_tokens": 5333876.0,
      "step": 326
    },
    {
      "entropy": 0.5309373140335083,
      "epoch": 1.2201492537313432,
      "grad_norm": 0.1433599442243576,
      "learning_rate": 0.0002,
      "loss": 0.5376647710800171,
      "mean_token_accuracy": 0.7800339162349701,
      "num_tokens": 5350199.0,
      "step": 327
    },
    {
      "entropy": 0.5192165076732635,
      "epoch": 1.2238805970149254,
      "grad_norm": 0.15373900532722473,
      "learning_rate": 0.0002,
      "loss": 0.5219945311546326,
      "mean_token_accuracy": 0.7878232598304749,
      "num_tokens": 5366574.0,
      "step": 328
    },
    {
      "entropy": 0.5340954586863518,
      "epoch": 1.2276119402985075,
      "grad_norm": 0.16795483231544495,
      "learning_rate": 0.0002,
      "loss": 0.5392008423805237,
      "mean_token_accuracy": 0.7791661024093628,
      "num_tokens": 5382990.0,
      "step": 329
    },
    {
      "entropy": 0.5285765826702118,
      "epoch": 1.2313432835820897,
      "grad_norm": 0.1866690218448639,
      "learning_rate": 0.0002,
      "loss": 0.5312389135360718,
      "mean_token_accuracy": 0.7833608090877533,
      "num_tokens": 5399472.0,
      "step": 330
    },
    {
      "entropy": 0.5517836213111877,
      "epoch": 1.2350746268656716,
      "grad_norm": 0.1322752982378006,
      "learning_rate": 0.0002,
      "loss": 0.5441842079162598,
      "mean_token_accuracy": 0.7805340141057968,
      "num_tokens": 5415825.0,
      "step": 331
    },
    {
      "entropy": 0.5583052486181259,
      "epoch": 1.2388059701492538,
      "grad_norm": 0.15239977836608887,
      "learning_rate": 0.0002,
      "loss": 0.546027421951294,
      "mean_token_accuracy": 0.7804329693317413,
      "num_tokens": 5432413.0,
      "step": 332
    },
    {
      "entropy": 0.5434032082557678,
      "epoch": 1.242537313432836,
      "grad_norm": 0.18549253046512604,
      "learning_rate": 0.0002,
      "loss": 0.5535240769386292,
      "mean_token_accuracy": 0.7762450277805328,
      "num_tokens": 5448590.0,
      "step": 333
    },
    {
      "entropy": 0.5200468450784683,
      "epoch": 1.2462686567164178,
      "grad_norm": 0.1469220519065857,
      "learning_rate": 0.0002,
      "loss": 0.5291376113891602,
      "mean_token_accuracy": 0.783469021320343,
      "num_tokens": 5464965.0,
      "step": 334
    },
    {
      "entropy": 0.5271460115909576,
      "epoch": 1.25,
      "grad_norm": 0.18688258528709412,
      "learning_rate": 0.0002,
      "loss": 0.5423641204833984,
      "mean_token_accuracy": 0.7798116505146027,
      "num_tokens": 5481288.0,
      "step": 335
    },
    {
      "entropy": 0.5354730933904648,
      "epoch": 1.2537313432835822,
      "grad_norm": 0.14062704145908356,
      "learning_rate": 0.0002,
      "loss": 0.5326344966888428,
      "mean_token_accuracy": 0.7858006954193115,
      "num_tokens": 5497484.0,
      "step": 336
    },
    {
      "entropy": 0.5392645597457886,
      "epoch": 1.2574626865671643,
      "grad_norm": 0.1739262342453003,
      "learning_rate": 0.0002,
      "loss": 0.5398157835006714,
      "mean_token_accuracy": 0.7812083959579468,
      "num_tokens": 5513800.0,
      "step": 337
    },
    {
      "entropy": 0.5331858694553375,
      "epoch": 1.2611940298507462,
      "grad_norm": 0.1530216783285141,
      "learning_rate": 0.0002,
      "loss": 0.5339893698692322,
      "mean_token_accuracy": 0.7844341546297073,
      "num_tokens": 5529910.0,
      "step": 338
    },
    {
      "entropy": 0.5276523530483246,
      "epoch": 1.2649253731343284,
      "grad_norm": 0.15110304951667786,
      "learning_rate": 0.0002,
      "loss": 0.5268523097038269,
      "mean_token_accuracy": 0.7853612899780273,
      "num_tokens": 5546422.0,
      "step": 339
    },
    {
      "entropy": 0.5267068892717361,
      "epoch": 1.2686567164179103,
      "grad_norm": 0.16843296587467194,
      "learning_rate": 0.0002,
      "loss": 0.5251078009605408,
      "mean_token_accuracy": 0.7890652269124985,
      "num_tokens": 5562574.0,
      "step": 340
    },
    {
      "entropy": 0.5488177984952927,
      "epoch": 1.2723880597014925,
      "grad_norm": 0.1422610580921173,
      "learning_rate": 0.0002,
      "loss": 0.5420413017272949,
      "mean_token_accuracy": 0.7802799195051193,
      "num_tokens": 5579060.0,
      "step": 341
    },
    {
      "entropy": 0.5321039855480194,
      "epoch": 1.2761194029850746,
      "grad_norm": 0.27034249901771545,
      "learning_rate": 0.0002,
      "loss": 0.534512996673584,
      "mean_token_accuracy": 0.7857411950826645,
      "num_tokens": 5595396.0,
      "step": 342
    },
    {
      "entropy": 0.5352365374565125,
      "epoch": 1.2798507462686568,
      "grad_norm": 0.15001201629638672,
      "learning_rate": 0.0002,
      "loss": 0.5391510725021362,
      "mean_token_accuracy": 0.7823953479528427,
      "num_tokens": 5611886.0,
      "step": 343
    },
    {
      "entropy": 0.5359922051429749,
      "epoch": 1.2835820895522387,
      "grad_norm": 0.18623347580432892,
      "learning_rate": 0.0002,
      "loss": 0.5446541905403137,
      "mean_token_accuracy": 0.780242919921875,
      "num_tokens": 5628417.0,
      "step": 344
    },
    {
      "entropy": 0.5499957650899887,
      "epoch": 1.287313432835821,
      "grad_norm": 0.13785430788993835,
      "learning_rate": 0.0002,
      "loss": 0.5447918772697449,
      "mean_token_accuracy": 0.7773040980100632,
      "num_tokens": 5644807.0,
      "step": 345
    },
    {
      "entropy": 0.5422873198986053,
      "epoch": 1.291044776119403,
      "grad_norm": 0.19493389129638672,
      "learning_rate": 0.0002,
      "loss": 0.545580267906189,
      "mean_token_accuracy": 0.7785766869783401,
      "num_tokens": 5660968.0,
      "step": 346
    },
    {
      "entropy": 0.5600688457489014,
      "epoch": 1.294776119402985,
      "grad_norm": 0.13508763909339905,
      "learning_rate": 0.0002,
      "loss": 0.5586625337600708,
      "mean_token_accuracy": 0.7736444920301437,
      "num_tokens": 5677583.0,
      "step": 347
    },
    {
      "entropy": 0.5331953912973404,
      "epoch": 1.2985074626865671,
      "grad_norm": 0.13123084604740143,
      "learning_rate": 0.0002,
      "loss": 0.5309450030326843,
      "mean_token_accuracy": 0.7863212525844574,
      "num_tokens": 5694104.0,
      "step": 348
    },
    {
      "entropy": 0.5230339020490646,
      "epoch": 1.3022388059701493,
      "grad_norm": 0.15445858240127563,
      "learning_rate": 0.0002,
      "loss": 0.5290713310241699,
      "mean_token_accuracy": 0.7851019501686096,
      "num_tokens": 5710331.0,
      "step": 349
    },
    {
      "entropy": 0.5597670078277588,
      "epoch": 1.3059701492537314,
      "grad_norm": 0.12102854996919632,
      "learning_rate": 0.0002,
      "loss": 0.5601732730865479,
      "mean_token_accuracy": 0.7745916843414307,
      "num_tokens": 5726639.0,
      "step": 350
    },
    {
      "entropy": 0.5379252731800079,
      "epoch": 1.3097014925373134,
      "grad_norm": 0.15875613689422607,
      "learning_rate": 0.0002,
      "loss": 0.5450211763381958,
      "mean_token_accuracy": 0.7802536338567734,
      "num_tokens": 5743022.0,
      "step": 351
    },
    {
      "entropy": 0.5242641121149063,
      "epoch": 1.3134328358208955,
      "grad_norm": 0.1778116673231125,
      "learning_rate": 0.0002,
      "loss": 0.5308142304420471,
      "mean_token_accuracy": 0.7841234058141708,
      "num_tokens": 5759431.0,
      "step": 352
    },
    {
      "entropy": 0.5329146087169647,
      "epoch": 1.3171641791044777,
      "grad_norm": 0.13093073666095734,
      "learning_rate": 0.0002,
      "loss": 0.5333167314529419,
      "mean_token_accuracy": 0.7824307978153229,
      "num_tokens": 5775870.0,
      "step": 353
    },
    {
      "entropy": 0.5363575518131256,
      "epoch": 1.3208955223880596,
      "grad_norm": 0.15412437915802002,
      "learning_rate": 0.0002,
      "loss": 0.5358852744102478,
      "mean_token_accuracy": 0.7839270085096359,
      "num_tokens": 5792217.0,
      "step": 354
    },
    {
      "entropy": 0.538811706006527,
      "epoch": 1.3246268656716418,
      "grad_norm": 0.12728764116764069,
      "learning_rate": 0.0002,
      "loss": 0.5357580780982971,
      "mean_token_accuracy": 0.7833316326141357,
      "num_tokens": 5808781.0,
      "step": 355
    },
    {
      "entropy": 0.5331800431013107,
      "epoch": 1.328358208955224,
      "grad_norm": 0.14081059396266937,
      "learning_rate": 0.0002,
      "loss": 0.5337068438529968,
      "mean_token_accuracy": 0.7839716076850891,
      "num_tokens": 5824951.0,
      "step": 356
    },
    {
      "entropy": 0.5460695028305054,
      "epoch": 1.332089552238806,
      "grad_norm": 0.16031600534915924,
      "learning_rate": 0.0002,
      "loss": 0.5452313423156738,
      "mean_token_accuracy": 0.7771268039941788,
      "num_tokens": 5841261.0,
      "step": 357
    },
    {
      "entropy": 0.540604442358017,
      "epoch": 1.335820895522388,
      "grad_norm": 0.12970028817653656,
      "learning_rate": 0.0002,
      "loss": 0.5406724810600281,
      "mean_token_accuracy": 0.7837041467428207,
      "num_tokens": 5857446.0,
      "step": 358
    },
    {
      "entropy": 0.5167972147464752,
      "epoch": 1.3395522388059702,
      "grad_norm": 0.15045662224292755,
      "learning_rate": 0.0002,
      "loss": 0.5290297865867615,
      "mean_token_accuracy": 0.785188764333725,
      "num_tokens": 5873984.0,
      "step": 359
    },
    {
      "entropy": 0.5397705882787704,
      "epoch": 1.3432835820895521,
      "grad_norm": 0.12482494115829468,
      "learning_rate": 0.0002,
      "loss": 0.5406928062438965,
      "mean_token_accuracy": 0.7810540497303009,
      "num_tokens": 5890369.0,
      "step": 360
    },
    {
      "entropy": 0.5350628793239594,
      "epoch": 1.3470149253731343,
      "grad_norm": 0.12060931324958801,
      "learning_rate": 0.0002,
      "loss": 0.5288930535316467,
      "mean_token_accuracy": 0.7861282080411911,
      "num_tokens": 5906651.0,
      "step": 361
    },
    {
      "entropy": 0.5198515579104424,
      "epoch": 1.3507462686567164,
      "grad_norm": 0.12899695336818695,
      "learning_rate": 0.0002,
      "loss": 0.5199739336967468,
      "mean_token_accuracy": 0.7908322215080261,
      "num_tokens": 5922992.0,
      "step": 362
    },
    {
      "entropy": 0.5311819463968277,
      "epoch": 1.3544776119402986,
      "grad_norm": 0.12395589053630829,
      "learning_rate": 0.0002,
      "loss": 0.5295090675354004,
      "mean_token_accuracy": 0.7864852249622345,
      "num_tokens": 5939475.0,
      "step": 363
    },
    {
      "entropy": 0.5441964268684387,
      "epoch": 1.3582089552238805,
      "grad_norm": 0.15559257566928864,
      "learning_rate": 0.0002,
      "loss": 0.5557464957237244,
      "mean_token_accuracy": 0.7727831602096558,
      "num_tokens": 5955917.0,
      "step": 364
    },
    {
      "entropy": 0.5318419188261032,
      "epoch": 1.3619402985074627,
      "grad_norm": 0.13309583067893982,
      "learning_rate": 0.0002,
      "loss": 0.531787633895874,
      "mean_token_accuracy": 0.7829223573207855,
      "num_tokens": 5972429.0,
      "step": 365
    },
    {
      "entropy": 0.5325569957494736,
      "epoch": 1.3656716417910448,
      "grad_norm": 0.15286150574684143,
      "learning_rate": 0.0002,
      "loss": 0.5281562805175781,
      "mean_token_accuracy": 0.7858050912618637,
      "num_tokens": 5988860.0,
      "step": 366
    },
    {
      "entropy": 0.5343448221683502,
      "epoch": 1.3694029850746268,
      "grad_norm": 0.15089887380599976,
      "learning_rate": 0.0002,
      "loss": 0.5317696332931519,
      "mean_token_accuracy": 0.7849821001291275,
      "num_tokens": 6005244.0,
      "step": 367
    },
    {
      "entropy": 0.5514230877161026,
      "epoch": 1.373134328358209,
      "grad_norm": 0.13945485651493073,
      "learning_rate": 0.0002,
      "loss": 0.5445812940597534,
      "mean_token_accuracy": 0.7807283848524094,
      "num_tokens": 6021687.0,
      "step": 368
    },
    {
      "entropy": 0.5257244408130646,
      "epoch": 1.376865671641791,
      "grad_norm": 0.15246176719665527,
      "learning_rate": 0.0002,
      "loss": 0.5266861319541931,
      "mean_token_accuracy": 0.7850677073001862,
      "num_tokens": 6038259.0,
      "step": 369
    },
    {
      "entropy": 0.5162273794412613,
      "epoch": 1.3805970149253732,
      "grad_norm": 0.12324219942092896,
      "learning_rate": 0.0002,
      "loss": 0.5183256268501282,
      "mean_token_accuracy": 0.7884418070316315,
      "num_tokens": 6054637.0,
      "step": 370
    },
    {
      "entropy": 0.5300852656364441,
      "epoch": 1.3843283582089552,
      "grad_norm": 0.1871775984764099,
      "learning_rate": 0.0002,
      "loss": 0.5378446578979492,
      "mean_token_accuracy": 0.7807964831590652,
      "num_tokens": 6071175.0,
      "step": 371
    },
    {
      "entropy": 0.5412422120571136,
      "epoch": 1.3880597014925373,
      "grad_norm": 0.14725005626678467,
      "learning_rate": 0.0002,
      "loss": 0.5465957522392273,
      "mean_token_accuracy": 0.7783457934856415,
      "num_tokens": 6087459.0,
      "step": 372
    },
    {
      "entropy": 0.5440727770328522,
      "epoch": 1.3917910447761195,
      "grad_norm": 0.1507187932729721,
      "learning_rate": 0.0002,
      "loss": 0.5401308536529541,
      "mean_token_accuracy": 0.7813372761011124,
      "num_tokens": 6103779.0,
      "step": 373
    },
    {
      "entropy": 0.5449319109320641,
      "epoch": 1.3955223880597014,
      "grad_norm": 0.1400064378976822,
      "learning_rate": 0.0002,
      "loss": 0.5477454662322998,
      "mean_token_accuracy": 0.7775075435638428,
      "num_tokens": 6120211.0,
      "step": 374
    },
    {
      "entropy": 0.5422904789447784,
      "epoch": 1.3992537313432836,
      "grad_norm": 0.14892356097698212,
      "learning_rate": 0.0002,
      "loss": 0.5438749194145203,
      "mean_token_accuracy": 0.7779098898172379,
      "num_tokens": 6136613.0,
      "step": 375
    },
    {
      "entropy": 0.5359798967838287,
      "epoch": 1.4029850746268657,
      "grad_norm": 0.1638849824666977,
      "learning_rate": 0.0002,
      "loss": 0.5464264154434204,
      "mean_token_accuracy": 0.7796951085329056,
      "num_tokens": 6153118.0,
      "step": 376
    },
    {
      "entropy": 0.5272762179374695,
      "epoch": 1.4067164179104479,
      "grad_norm": 0.1572565734386444,
      "learning_rate": 0.0002,
      "loss": 0.5275948643684387,
      "mean_token_accuracy": 0.7868656516075134,
      "num_tokens": 6169351.0,
      "step": 377
    },
    {
      "entropy": 0.5318129509687424,
      "epoch": 1.4104477611940298,
      "grad_norm": 0.1217207983136177,
      "learning_rate": 0.0002,
      "loss": 0.5282657742500305,
      "mean_token_accuracy": 0.7874875515699387,
      "num_tokens": 6185947.0,
      "step": 378
    },
    {
      "entropy": 0.5453623086214066,
      "epoch": 1.414179104477612,
      "grad_norm": 0.3408029079437256,
      "learning_rate": 0.0002,
      "loss": 0.5384775400161743,
      "mean_token_accuracy": 0.7859825044870377,
      "num_tokens": 6202395.0,
      "step": 379
    },
    {
      "entropy": 0.5526222735643387,
      "epoch": 1.417910447761194,
      "grad_norm": 0.23189330101013184,
      "learning_rate": 0.0002,
      "loss": 0.5524420738220215,
      "mean_token_accuracy": 0.7748881280422211,
      "num_tokens": 6218653.0,
      "step": 380
    },
    {
      "entropy": 0.5264930576086044,
      "epoch": 1.421641791044776,
      "grad_norm": 0.19049955904483795,
      "learning_rate": 0.0002,
      "loss": 0.5374311208724976,
      "mean_token_accuracy": 0.7826297730207443,
      "num_tokens": 6234993.0,
      "step": 381
    },
    {
      "entropy": 0.5191953629255295,
      "epoch": 1.4253731343283582,
      "grad_norm": 0.1691526621580124,
      "learning_rate": 0.0002,
      "loss": 0.5358802080154419,
      "mean_token_accuracy": 0.7822402864694595,
      "num_tokens": 6251114.0,
      "step": 382
    },
    {
      "entropy": 0.5427180528640747,
      "epoch": 1.4291044776119404,
      "grad_norm": 0.13084714114665985,
      "learning_rate": 0.0002,
      "loss": 0.537938117980957,
      "mean_token_accuracy": 0.783057376742363,
      "num_tokens": 6267469.0,
      "step": 383
    },
    {
      "entropy": 0.5605024993419647,
      "epoch": 1.4328358208955223,
      "grad_norm": 0.15293866395950317,
      "learning_rate": 0.0002,
      "loss": 0.5501874685287476,
      "mean_token_accuracy": 0.7784303724765778,
      "num_tokens": 6283955.0,
      "step": 384
    },
    {
      "entropy": 0.5423373132944107,
      "epoch": 1.4365671641791045,
      "grad_norm": 0.12526267766952515,
      "learning_rate": 0.0002,
      "loss": 0.5338963866233826,
      "mean_token_accuracy": 0.7840876579284668,
      "num_tokens": 6300241.0,
      "step": 385
    },
    {
      "entropy": 0.5322032272815704,
      "epoch": 1.4402985074626866,
      "grad_norm": 0.14180147647857666,
      "learning_rate": 0.0002,
      "loss": 0.5326216220855713,
      "mean_token_accuracy": 0.7845493853092194,
      "num_tokens": 6316506.0,
      "step": 386
    },
    {
      "entropy": 0.5429232567548752,
      "epoch": 1.4440298507462686,
      "grad_norm": 0.16007299721240997,
      "learning_rate": 0.0002,
      "loss": 0.5557198524475098,
      "mean_token_accuracy": 0.7756443470716476,
      "num_tokens": 6333160.0,
      "step": 387
    },
    {
      "entropy": 0.5302825719118118,
      "epoch": 1.4477611940298507,
      "grad_norm": 0.12839363515377045,
      "learning_rate": 0.0002,
      "loss": 0.5346431136131287,
      "mean_token_accuracy": 0.7858215421438217,
      "num_tokens": 6349692.0,
      "step": 388
    },
    {
      "entropy": 0.5415708720684052,
      "epoch": 1.4514925373134329,
      "grad_norm": 0.1452096849679947,
      "learning_rate": 0.0002,
      "loss": 0.5441015362739563,
      "mean_token_accuracy": 0.7760088890790939,
      "num_tokens": 6365979.0,
      "step": 389
    },
    {
      "entropy": 0.5408393442630768,
      "epoch": 1.455223880597015,
      "grad_norm": 0.1318868100643158,
      "learning_rate": 0.0002,
      "loss": 0.5366842746734619,
      "mean_token_accuracy": 0.7860211282968521,
      "num_tokens": 6382253.0,
      "step": 390
    },
    {
      "entropy": 0.5344153642654419,
      "epoch": 1.458955223880597,
      "grad_norm": 0.1528550684452057,
      "learning_rate": 0.0002,
      "loss": 0.5343711972236633,
      "mean_token_accuracy": 0.7834310978651047,
      "num_tokens": 6398277.0,
      "step": 391
    },
    {
      "entropy": 0.5448261797428131,
      "epoch": 1.462686567164179,
      "grad_norm": 0.13751660287380219,
      "learning_rate": 0.0002,
      "loss": 0.5372248888015747,
      "mean_token_accuracy": 0.781394749879837,
      "num_tokens": 6414655.0,
      "step": 392
    },
    {
      "entropy": 0.5271979123353958,
      "epoch": 1.4664179104477613,
      "grad_norm": 0.14335128664970398,
      "learning_rate": 0.0002,
      "loss": 0.5272061824798584,
      "mean_token_accuracy": 0.7860262989997864,
      "num_tokens": 6430931.0,
      "step": 393
    },
    {
      "entropy": 0.5503392070531845,
      "epoch": 1.4701492537313432,
      "grad_norm": 0.1436121165752411,
      "learning_rate": 0.0002,
      "loss": 0.5583488345146179,
      "mean_token_accuracy": 0.7723452150821686,
      "num_tokens": 6447385.0,
      "step": 394
    },
    {
      "entropy": 0.5310524106025696,
      "epoch": 1.4738805970149254,
      "grad_norm": 0.16840456426143646,
      "learning_rate": 0.0002,
      "loss": 0.5381408929824829,
      "mean_token_accuracy": 0.7820066809654236,
      "num_tokens": 6463801.0,
      "step": 395
    },
    {
      "entropy": 0.544038251042366,
      "epoch": 1.4776119402985075,
      "grad_norm": 0.12804877758026123,
      "learning_rate": 0.0002,
      "loss": 0.5466898083686829,
      "mean_token_accuracy": 0.779111385345459,
      "num_tokens": 6480225.0,
      "step": 396
    },
    {
      "entropy": 0.5286617875099182,
      "epoch": 1.4813432835820897,
      "grad_norm": 0.1583615094423294,
      "learning_rate": 0.0002,
      "loss": 0.5371429324150085,
      "mean_token_accuracy": 0.781273365020752,
      "num_tokens": 6496250.0,
      "step": 397
    },
    {
      "entropy": 0.5369281619787216,
      "epoch": 1.4850746268656716,
      "grad_norm": 0.13522376120090485,
      "learning_rate": 0.0002,
      "loss": 0.5328642129898071,
      "mean_token_accuracy": 0.7812697738409042,
      "num_tokens": 6512604.0,
      "step": 398
    },
    {
      "entropy": 0.535246454179287,
      "epoch": 1.4888059701492538,
      "grad_norm": 0.12748968601226807,
      "learning_rate": 0.0002,
      "loss": 0.529126763343811,
      "mean_token_accuracy": 0.7845304161310196,
      "num_tokens": 6528888.0,
      "step": 399
    },
    {
      "entropy": 0.5552913099527359,
      "epoch": 1.4925373134328357,
      "grad_norm": 0.1282271295785904,
      "learning_rate": 0.0002,
      "loss": 0.5470327138900757,
      "mean_token_accuracy": 0.7776638716459274,
      "num_tokens": 6545412.0,
      "step": 400
    },
    {
      "entropy": 0.5173196047544479,
      "epoch": 1.4962686567164178,
      "grad_norm": 0.14874818921089172,
      "learning_rate": 0.0002,
      "loss": 0.524068295955658,
      "mean_token_accuracy": 0.7867140471935272,
      "num_tokens": 6561535.0,
      "step": 401
    },
    {
      "entropy": 0.5213709771633148,
      "epoch": 1.5,
      "grad_norm": 0.13309182226657867,
      "learning_rate": 0.0002,
      "loss": 0.5244147777557373,
      "mean_token_accuracy": 0.7871381193399429,
      "num_tokens": 6577842.0,
      "step": 402
    },
    {
      "entropy": 0.5317850112915039,
      "epoch": 1.5037313432835822,
      "grad_norm": 0.1433606594800949,
      "learning_rate": 0.0002,
      "loss": 0.5365285277366638,
      "mean_token_accuracy": 0.7824452221393585,
      "num_tokens": 6594006.0,
      "step": 403
    },
    {
      "entropy": 0.544880673289299,
      "epoch": 1.5074626865671643,
      "grad_norm": 0.12236304581165314,
      "learning_rate": 0.0002,
      "loss": 0.5444092750549316,
      "mean_token_accuracy": 0.778602346777916,
      "num_tokens": 6610288.0,
      "step": 404
    },
    {
      "entropy": 0.532042846083641,
      "epoch": 1.5111940298507462,
      "grad_norm": 0.12565302848815918,
      "learning_rate": 0.0002,
      "loss": 0.5315436124801636,
      "mean_token_accuracy": 0.7820066958665848,
      "num_tokens": 6626761.0,
      "step": 405
    },
    {
      "entropy": 0.5506146252155304,
      "epoch": 1.5149253731343284,
      "grad_norm": 0.12706807255744934,
      "learning_rate": 0.0002,
      "loss": 0.5524188876152039,
      "mean_token_accuracy": 0.7748337537050247,
      "num_tokens": 6643344.0,
      "step": 406
    },
    {
      "entropy": 0.5542557537555695,
      "epoch": 1.5186567164179103,
      "grad_norm": 0.12174059450626373,
      "learning_rate": 0.0002,
      "loss": 0.5521225929260254,
      "mean_token_accuracy": 0.778037965297699,
      "num_tokens": 6660150.0,
      "step": 407
    },
    {
      "entropy": 0.5313189476728439,
      "epoch": 1.5223880597014925,
      "grad_norm": 0.16201771795749664,
      "learning_rate": 0.0002,
      "loss": 0.5316515564918518,
      "mean_token_accuracy": 0.7825613915920258,
      "num_tokens": 6676257.0,
      "step": 408
    },
    {
      "entropy": 0.5481334328651428,
      "epoch": 1.5261194029850746,
      "grad_norm": 0.14420105516910553,
      "learning_rate": 0.0002,
      "loss": 0.547515869140625,
      "mean_token_accuracy": 0.7776167094707489,
      "num_tokens": 6692716.0,
      "step": 409
    },
    {
      "entropy": 0.5290578305721283,
      "epoch": 1.5298507462686568,
      "grad_norm": 0.14699825644493103,
      "learning_rate": 0.0002,
      "loss": 0.5254528522491455,
      "mean_token_accuracy": 0.788268119096756,
      "num_tokens": 6708994.0,
      "step": 410
    },
    {
      "entropy": 0.5288655012845993,
      "epoch": 1.533582089552239,
      "grad_norm": 0.15162502229213715,
      "learning_rate": 0.0002,
      "loss": 0.5358376502990723,
      "mean_token_accuracy": 0.782762810587883,
      "num_tokens": 6725371.0,
      "step": 411
    },
    {
      "entropy": 0.5213432013988495,
      "epoch": 1.537313432835821,
      "grad_norm": 0.1827428936958313,
      "learning_rate": 0.0002,
      "loss": 0.5308473110198975,
      "mean_token_accuracy": 0.7839875668287277,
      "num_tokens": 6741339.0,
      "step": 412
    },
    {
      "entropy": 0.5378746837377548,
      "epoch": 1.5410447761194028,
      "grad_norm": 0.13657329976558685,
      "learning_rate": 0.0002,
      "loss": 0.5338229537010193,
      "mean_token_accuracy": 0.7829029411077499,
      "num_tokens": 6757780.0,
      "step": 413
    },
    {
      "entropy": 0.5342704057693481,
      "epoch": 1.544776119402985,
      "grad_norm": 0.1464962214231491,
      "learning_rate": 0.0002,
      "loss": 0.529054582118988,
      "mean_token_accuracy": 0.785891056060791,
      "num_tokens": 6774311.0,
      "step": 414
    },
    {
      "entropy": 0.5312474966049194,
      "epoch": 1.5485074626865671,
      "grad_norm": 0.15172705054283142,
      "learning_rate": 0.0002,
      "loss": 0.5394980311393738,
      "mean_token_accuracy": 0.7824061512947083,
      "num_tokens": 6790408.0,
      "step": 415
    },
    {
      "entropy": 0.5170525014400482,
      "epoch": 1.5522388059701493,
      "grad_norm": 0.15336130559444427,
      "learning_rate": 0.0002,
      "loss": 0.5251516699790955,
      "mean_token_accuracy": 0.7875834852457047,
      "num_tokens": 6806555.0,
      "step": 416
    },
    {
      "entropy": 0.5356197506189346,
      "epoch": 1.5559701492537314,
      "grad_norm": 0.14176255464553833,
      "learning_rate": 0.0002,
      "loss": 0.5379368662834167,
      "mean_token_accuracy": 0.781431034207344,
      "num_tokens": 6823014.0,
      "step": 417
    },
    {
      "entropy": 0.5277817100286484,
      "epoch": 1.5597014925373134,
      "grad_norm": 0.14984361827373505,
      "learning_rate": 0.0002,
      "loss": 0.524969756603241,
      "mean_token_accuracy": 0.7859384715557098,
      "num_tokens": 6839244.0,
      "step": 418
    },
    {
      "entropy": 0.535449355840683,
      "epoch": 1.5634328358208955,
      "grad_norm": 0.13689614832401276,
      "learning_rate": 0.0002,
      "loss": 0.5337757468223572,
      "mean_token_accuracy": 0.7802022695541382,
      "num_tokens": 6855421.0,
      "step": 419
    },
    {
      "entropy": 0.5216768980026245,
      "epoch": 1.5671641791044775,
      "grad_norm": 0.13293899595737457,
      "learning_rate": 0.0002,
      "loss": 0.5152947902679443,
      "mean_token_accuracy": 0.7913374304771423,
      "num_tokens": 6871759.0,
      "step": 420
    },
    {
      "entropy": 0.5257552266120911,
      "epoch": 1.5708955223880596,
      "grad_norm": 0.12973323464393616,
      "learning_rate": 0.0002,
      "loss": 0.5320898294448853,
      "mean_token_accuracy": 0.7863506823778152,
      "num_tokens": 6888261.0,
      "step": 421
    },
    {
      "entropy": 0.5438365638256073,
      "epoch": 1.5746268656716418,
      "grad_norm": 0.15889972448349,
      "learning_rate": 0.0002,
      "loss": 0.5526459217071533,
      "mean_token_accuracy": 0.7783207595348358,
      "num_tokens": 6904962.0,
      "step": 422
    },
    {
      "entropy": 0.5427816063165665,
      "epoch": 1.578358208955224,
      "grad_norm": 0.141456738114357,
      "learning_rate": 0.0002,
      "loss": 0.5505415797233582,
      "mean_token_accuracy": 0.7792007029056549,
      "num_tokens": 6921400.0,
      "step": 423
    },
    {
      "entropy": 0.5111957639455795,
      "epoch": 1.582089552238806,
      "grad_norm": 0.1351761519908905,
      "learning_rate": 0.0002,
      "loss": 0.511957585811615,
      "mean_token_accuracy": 0.7941061407327652,
      "num_tokens": 6937506.0,
      "step": 424
    },
    {
      "entropy": 0.5456200540065765,
      "epoch": 1.585820895522388,
      "grad_norm": 0.14772872626781464,
      "learning_rate": 0.0002,
      "loss": 0.5425821542739868,
      "mean_token_accuracy": 0.777664914727211,
      "num_tokens": 6953968.0,
      "step": 425
    },
    {
      "entropy": 0.5384055227041245,
      "epoch": 1.5895522388059702,
      "grad_norm": 0.12057865411043167,
      "learning_rate": 0.0002,
      "loss": 0.5403178930282593,
      "mean_token_accuracy": 0.7802070528268814,
      "num_tokens": 6970101.0,
      "step": 426
    },
    {
      "entropy": 0.5330870598554611,
      "epoch": 1.5932835820895521,
      "grad_norm": 0.14061623811721802,
      "learning_rate": 0.0002,
      "loss": 0.5319790244102478,
      "mean_token_accuracy": 0.7867544293403625,
      "num_tokens": 6986564.0,
      "step": 427
    },
    {
      "entropy": 0.5127943903207779,
      "epoch": 1.5970149253731343,
      "grad_norm": 0.1368899643421173,
      "learning_rate": 0.0002,
      "loss": 0.5163131952285767,
      "mean_token_accuracy": 0.7912420630455017,
      "num_tokens": 7002876.0,
      "step": 428
    },
    {
      "entropy": 0.5254192352294922,
      "epoch": 1.6007462686567164,
      "grad_norm": 0.16491283476352692,
      "learning_rate": 0.0002,
      "loss": 0.5343653559684753,
      "mean_token_accuracy": 0.7813780009746552,
      "num_tokens": 7019170.0,
      "step": 429
    },
    {
      "entropy": 0.5337297320365906,
      "epoch": 1.6044776119402986,
      "grad_norm": 0.13911473751068115,
      "learning_rate": 0.0002,
      "loss": 0.5348519086837769,
      "mean_token_accuracy": 0.7835835814476013,
      "num_tokens": 7035680.0,
      "step": 430
    },
    {
      "entropy": 0.5483486354351044,
      "epoch": 1.6082089552238807,
      "grad_norm": 0.15066570043563843,
      "learning_rate": 0.0002,
      "loss": 0.543323814868927,
      "mean_token_accuracy": 0.7779914885759354,
      "num_tokens": 7051951.0,
      "step": 431
    },
    {
      "entropy": 0.558162733912468,
      "epoch": 1.6119402985074627,
      "grad_norm": 0.14888809621334076,
      "learning_rate": 0.0002,
      "loss": 0.5558294057846069,
      "mean_token_accuracy": 0.7741727977991104,
      "num_tokens": 7068414.0,
      "step": 432
    },
    {
      "entropy": 0.5639995038509369,
      "epoch": 1.6156716417910446,
      "grad_norm": 0.14019054174423218,
      "learning_rate": 0.0002,
      "loss": 0.5553058385848999,
      "mean_token_accuracy": 0.7748651653528214,
      "num_tokens": 7084803.0,
      "step": 433
    },
    {
      "entropy": 0.52562515437603,
      "epoch": 1.6194029850746268,
      "grad_norm": 0.1775631606578827,
      "learning_rate": 0.0002,
      "loss": 0.5248590111732483,
      "mean_token_accuracy": 0.7880850434303284,
      "num_tokens": 7101193.0,
      "step": 434
    },
    {
      "entropy": 0.5416489392518997,
      "epoch": 1.623134328358209,
      "grad_norm": 0.12513571977615356,
      "learning_rate": 0.0002,
      "loss": 0.5416374206542969,
      "mean_token_accuracy": 0.7787177264690399,
      "num_tokens": 7117779.0,
      "step": 435
    },
    {
      "entropy": 0.5430044084787369,
      "epoch": 1.626865671641791,
      "grad_norm": 0.19869297742843628,
      "learning_rate": 0.0002,
      "loss": 0.5577962398529053,
      "mean_token_accuracy": 0.7749052196741104,
      "num_tokens": 7134065.0,
      "step": 436
    },
    {
      "entropy": 0.526473268866539,
      "epoch": 1.6305970149253732,
      "grad_norm": 0.13564684987068176,
      "learning_rate": 0.0002,
      "loss": 0.5336202383041382,
      "mean_token_accuracy": 0.7842326164245605,
      "num_tokens": 7150250.0,
      "step": 437
    },
    {
      "entropy": 0.5367172211408615,
      "epoch": 1.6343283582089554,
      "grad_norm": 0.13337789475917816,
      "learning_rate": 0.0002,
      "loss": 0.5405253171920776,
      "mean_token_accuracy": 0.7821898907423019,
      "num_tokens": 7166822.0,
      "step": 438
    },
    {
      "entropy": 0.5436544269323349,
      "epoch": 1.6380597014925373,
      "grad_norm": 0.13991288840770721,
      "learning_rate": 0.0002,
      "loss": 0.5384564399719238,
      "mean_token_accuracy": 0.7812158763408661,
      "num_tokens": 7183330.0,
      "step": 439
    },
    {
      "entropy": 0.5360020697116852,
      "epoch": 1.6417910447761193,
      "grad_norm": 0.1291709989309311,
      "learning_rate": 0.0002,
      "loss": 0.524196982383728,
      "mean_token_accuracy": 0.7875594347715378,
      "num_tokens": 7199815.0,
      "step": 440
    },
    {
      "entropy": 0.5343951657414436,
      "epoch": 1.6455223880597014,
      "grad_norm": 0.13694606721401215,
      "learning_rate": 0.0002,
      "loss": 0.5314103960990906,
      "mean_token_accuracy": 0.7855215966701508,
      "num_tokens": 7216312.0,
      "step": 441
    },
    {
      "entropy": 0.5499645173549652,
      "epoch": 1.6492537313432836,
      "grad_norm": 0.1579257845878601,
      "learning_rate": 0.0002,
      "loss": 0.5512703061103821,
      "mean_token_accuracy": 0.7742991000413895,
      "num_tokens": 7232728.0,
      "step": 442
    },
    {
      "entropy": 0.5159503519535065,
      "epoch": 1.6529850746268657,
      "grad_norm": 0.16577711701393127,
      "learning_rate": 0.0002,
      "loss": 0.5299012660980225,
      "mean_token_accuracy": 0.7818514108657837,
      "num_tokens": 7248844.0,
      "step": 443
    },
    {
      "entropy": 0.5321104824542999,
      "epoch": 1.6567164179104479,
      "grad_norm": 0.19517181813716888,
      "learning_rate": 0.0002,
      "loss": 0.5426245927810669,
      "mean_token_accuracy": 0.7829670608043671,
      "num_tokens": 7265297.0,
      "step": 444
    },
    {
      "entropy": 0.5357498228549957,
      "epoch": 1.6604477611940298,
      "grad_norm": 0.1414795070886612,
      "learning_rate": 0.0002,
      "loss": 0.5346581935882568,
      "mean_token_accuracy": 0.784277081489563,
      "num_tokens": 7281737.0,
      "step": 445
    },
    {
      "entropy": 0.5516169220209122,
      "epoch": 1.664179104477612,
      "grad_norm": 0.16740073263645172,
      "learning_rate": 0.0002,
      "loss": 0.5406741499900818,
      "mean_token_accuracy": 0.7807598114013672,
      "num_tokens": 7298065.0,
      "step": 446
    },
    {
      "entropy": 0.5355339050292969,
      "epoch": 1.667910447761194,
      "grad_norm": 0.11836501210927963,
      "learning_rate": 0.0002,
      "loss": 0.527145266532898,
      "mean_token_accuracy": 0.7858247607946396,
      "num_tokens": 7314370.0,
      "step": 447
    },
    {
      "entropy": 0.5455258637666702,
      "epoch": 1.671641791044776,
      "grad_norm": 0.12945692241191864,
      "learning_rate": 0.0002,
      "loss": 0.5454037189483643,
      "mean_token_accuracy": 0.7800126224756241,
      "num_tokens": 7330673.0,
      "step": 448
    },
    {
      "entropy": 0.5411469787359238,
      "epoch": 1.6753731343283582,
      "grad_norm": 0.16638043522834778,
      "learning_rate": 0.0002,
      "loss": 0.5542144775390625,
      "mean_token_accuracy": 0.772837832570076,
      "num_tokens": 7346907.0,
      "step": 449
    },
    {
      "entropy": 0.5359169989824295,
      "epoch": 1.6791044776119404,
      "grad_norm": 0.14383791387081146,
      "learning_rate": 0.0002,
      "loss": 0.5425304770469666,
      "mean_token_accuracy": 0.7819065749645233,
      "num_tokens": 7363351.0,
      "step": 450
    },
    {
      "entropy": 0.5467946827411652,
      "epoch": 1.6828358208955225,
      "grad_norm": 0.12469735741615295,
      "learning_rate": 0.0002,
      "loss": 0.5454594492912292,
      "mean_token_accuracy": 0.7800441980361938,
      "num_tokens": 7379994.0,
      "step": 451
    },
    {
      "entropy": 0.5265442132949829,
      "epoch": 1.6865671641791045,
      "grad_norm": 0.123937226831913,
      "learning_rate": 0.0002,
      "loss": 0.5216733813285828,
      "mean_token_accuracy": 0.786560595035553,
      "num_tokens": 7396349.0,
      "step": 452
    },
    {
      "entropy": 0.5408566147089005,
      "epoch": 1.6902985074626866,
      "grad_norm": 0.14807483553886414,
      "learning_rate": 0.0002,
      "loss": 0.5339705944061279,
      "mean_token_accuracy": 0.7806796431541443,
      "num_tokens": 7412716.0,
      "step": 453
    },
    {
      "entropy": 0.542450025677681,
      "epoch": 1.6940298507462686,
      "grad_norm": 0.12422959506511688,
      "learning_rate": 0.0002,
      "loss": 0.538216769695282,
      "mean_token_accuracy": 0.7807506322860718,
      "num_tokens": 7429301.0,
      "step": 454
    },
    {
      "entropy": 0.5282323509454727,
      "epoch": 1.6977611940298507,
      "grad_norm": 0.1373056024312973,
      "learning_rate": 0.0002,
      "loss": 0.5334514379501343,
      "mean_token_accuracy": 0.7819238603115082,
      "num_tokens": 7445698.0,
      "step": 455
    },
    {
      "entropy": 0.5225353688001633,
      "epoch": 1.7014925373134329,
      "grad_norm": 0.1703997701406479,
      "learning_rate": 0.0002,
      "loss": 0.5377897024154663,
      "mean_token_accuracy": 0.7809168249368668,
      "num_tokens": 7462002.0,
      "step": 456
    },
    {
      "entropy": 0.5348514318466187,
      "epoch": 1.705223880597015,
      "grad_norm": 0.15133416652679443,
      "learning_rate": 0.0002,
      "loss": 0.543133020401001,
      "mean_token_accuracy": 0.7799534946680069,
      "num_tokens": 7478153.0,
      "step": 457
    },
    {
      "entropy": 0.54405577480793,
      "epoch": 1.7089552238805972,
      "grad_norm": 0.13994158804416656,
      "learning_rate": 0.0002,
      "loss": 0.5500596165657043,
      "mean_token_accuracy": 0.7780357599258423,
      "num_tokens": 7494573.0,
      "step": 458
    },
    {
      "entropy": 0.5588826686143875,
      "epoch": 1.712686567164179,
      "grad_norm": 0.15153922140598297,
      "learning_rate": 0.0002,
      "loss": 0.5551016330718994,
      "mean_token_accuracy": 0.7748077511787415,
      "num_tokens": 7510790.0,
      "step": 459
    },
    {
      "entropy": 0.5657205730676651,
      "epoch": 1.716417910447761,
      "grad_norm": 0.13676036894321442,
      "learning_rate": 0.0002,
      "loss": 0.5570163726806641,
      "mean_token_accuracy": 0.773433119058609,
      "num_tokens": 7527466.0,
      "step": 460
    },
    {
      "entropy": 0.5257614329457283,
      "epoch": 1.7201492537313432,
      "grad_norm": 0.13067854940891266,
      "learning_rate": 0.0002,
      "loss": 0.5170261263847351,
      "mean_token_accuracy": 0.7894982546567917,
      "num_tokens": 7543803.0,
      "step": 461
    },
    {
      "entropy": 0.5358800739049911,
      "epoch": 1.7238805970149254,
      "grad_norm": 0.12191294878721237,
      "learning_rate": 0.0002,
      "loss": 0.5280119776725769,
      "mean_token_accuracy": 0.7868626117706299,
      "num_tokens": 7560274.0,
      "step": 462
    },
    {
      "entropy": 0.5240240395069122,
      "epoch": 1.7276119402985075,
      "grad_norm": 0.14306314289569855,
      "learning_rate": 0.0002,
      "loss": 0.5353462100028992,
      "mean_token_accuracy": 0.782327190041542,
      "num_tokens": 7576773.0,
      "step": 463
    },
    {
      "entropy": 0.5037208497524261,
      "epoch": 1.7313432835820897,
      "grad_norm": 0.14881490170955658,
      "learning_rate": 0.0002,
      "loss": 0.5086286067962646,
      "mean_token_accuracy": 0.7924383580684662,
      "num_tokens": 7593022.0,
      "step": 464
    },
    {
      "entropy": 0.5487196296453476,
      "epoch": 1.7350746268656716,
      "grad_norm": 0.14098134636878967,
      "learning_rate": 0.0002,
      "loss": 0.5548460483551025,
      "mean_token_accuracy": 0.7751922011375427,
      "num_tokens": 7609497.0,
      "step": 465
    },
    {
      "entropy": 0.5405637472867966,
      "epoch": 1.7388059701492538,
      "grad_norm": 0.13519442081451416,
      "learning_rate": 0.0002,
      "loss": 0.5397500395774841,
      "mean_token_accuracy": 0.7797744125127792,
      "num_tokens": 7625829.0,
      "step": 466
    },
    {
      "entropy": 0.542302668094635,
      "epoch": 1.7425373134328357,
      "grad_norm": 0.12859952449798584,
      "learning_rate": 0.0002,
      "loss": 0.534672200679779,
      "mean_token_accuracy": 0.7826961874961853,
      "num_tokens": 7642312.0,
      "step": 467
    },
    {
      "entropy": 0.533673532307148,
      "epoch": 1.7462686567164178,
      "grad_norm": 0.14180679619312286,
      "learning_rate": 0.0002,
      "loss": 0.5322732925415039,
      "mean_token_accuracy": 0.7828265875577927,
      "num_tokens": 7658803.0,
      "step": 468
    },
    {
      "entropy": 0.5318716764450073,
      "epoch": 1.75,
      "grad_norm": 0.16254329681396484,
      "learning_rate": 0.0002,
      "loss": 0.5418020486831665,
      "mean_token_accuracy": 0.7817637473344803,
      "num_tokens": 7674957.0,
      "step": 469
    },
    {
      "entropy": 0.5334404781460762,
      "epoch": 1.7537313432835822,
      "grad_norm": 0.14448624849319458,
      "learning_rate": 0.0002,
      "loss": 0.5339848399162292,
      "mean_token_accuracy": 0.7814598977565765,
      "num_tokens": 7691620.0,
      "step": 470
    },
    {
      "entropy": 0.551649421453476,
      "epoch": 1.7574626865671643,
      "grad_norm": 0.16072067618370056,
      "learning_rate": 0.0002,
      "loss": 0.5504499673843384,
      "mean_token_accuracy": 0.7762804180383682,
      "num_tokens": 7708049.0,
      "step": 471
    },
    {
      "entropy": 0.5477486550807953,
      "epoch": 1.7611940298507462,
      "grad_norm": 0.1135358139872551,
      "learning_rate": 0.0002,
      "loss": 0.5485340356826782,
      "mean_token_accuracy": 0.7758127301931381,
      "num_tokens": 7724545.0,
      "step": 472
    },
    {
      "entropy": 0.5408584326505661,
      "epoch": 1.7649253731343284,
      "grad_norm": 0.132512167096138,
      "learning_rate": 0.0002,
      "loss": 0.5433669686317444,
      "mean_token_accuracy": 0.7828538417816162,
      "num_tokens": 7741050.0,
      "step": 473
    },
    {
      "entropy": 0.5271462053060532,
      "epoch": 1.7686567164179103,
      "grad_norm": 0.1477174609899521,
      "learning_rate": 0.0002,
      "loss": 0.5306966304779053,
      "mean_token_accuracy": 0.7825304567813873,
      "num_tokens": 7757294.0,
      "step": 474
    },
    {
      "entropy": 0.5110566020011902,
      "epoch": 1.7723880597014925,
      "grad_norm": 0.1395856738090515,
      "learning_rate": 0.0002,
      "loss": 0.5155825018882751,
      "mean_token_accuracy": 0.7897263616323471,
      "num_tokens": 7773651.0,
      "step": 475
    },
    {
      "entropy": 0.5346289277076721,
      "epoch": 1.7761194029850746,
      "grad_norm": 0.12764999270439148,
      "learning_rate": 0.0002,
      "loss": 0.5333414673805237,
      "mean_token_accuracy": 0.7830168902873993,
      "num_tokens": 7789823.0,
      "step": 476
    },
    {
      "entropy": 0.5467153936624527,
      "epoch": 1.7798507462686568,
      "grad_norm": 0.15203405916690826,
      "learning_rate": 0.0002,
      "loss": 0.5423698425292969,
      "mean_token_accuracy": 0.7790202498435974,
      "num_tokens": 7806470.0,
      "step": 477
    },
    {
      "entropy": 0.532525897026062,
      "epoch": 1.783582089552239,
      "grad_norm": 0.1433921456336975,
      "learning_rate": 0.0002,
      "loss": 0.529548168182373,
      "mean_token_accuracy": 0.786345586180687,
      "num_tokens": 7822732.0,
      "step": 478
    },
    {
      "entropy": 0.5197082981467247,
      "epoch": 1.787313432835821,
      "grad_norm": 0.14001280069351196,
      "learning_rate": 0.0002,
      "loss": 0.5263832807540894,
      "mean_token_accuracy": 0.7867995053529739,
      "num_tokens": 7838797.0,
      "step": 479
    },
    {
      "entropy": 0.5292367935180664,
      "epoch": 1.7910447761194028,
      "grad_norm": 0.14335371553897858,
      "learning_rate": 0.0002,
      "loss": 0.5325329303741455,
      "mean_token_accuracy": 0.7837976664304733,
      "num_tokens": 7855144.0,
      "step": 480
    },
    {
      "entropy": 0.5310244560241699,
      "epoch": 1.794776119402985,
      "grad_norm": 0.15651096403598785,
      "learning_rate": 0.0002,
      "loss": 0.5364205837249756,
      "mean_token_accuracy": 0.781629890203476,
      "num_tokens": 7871292.0,
      "step": 481
    },
    {
      "entropy": 0.5531992614269257,
      "epoch": 1.7985074626865671,
      "grad_norm": 0.1264420747756958,
      "learning_rate": 0.0002,
      "loss": 0.5501738786697388,
      "mean_token_accuracy": 0.7759624570608139,
      "num_tokens": 7887723.0,
      "step": 482
    },
    {
      "entropy": 0.5409988686442375,
      "epoch": 1.8022388059701493,
      "grad_norm": 0.12659387290477753,
      "learning_rate": 0.0002,
      "loss": 0.5382291674613953,
      "mean_token_accuracy": 0.7843269854784012,
      "num_tokens": 7904018.0,
      "step": 483
    },
    {
      "entropy": 0.5395593494176865,
      "epoch": 1.8059701492537314,
      "grad_norm": 0.1302158683538437,
      "learning_rate": 0.0002,
      "loss": 0.5366786122322083,
      "mean_token_accuracy": 0.7830972671508789,
      "num_tokens": 7920488.0,
      "step": 484
    },
    {
      "entropy": 0.5380191504955292,
      "epoch": 1.8097014925373134,
      "grad_norm": 0.1527208387851715,
      "learning_rate": 0.0002,
      "loss": 0.5442609786987305,
      "mean_token_accuracy": 0.7774769812822342,
      "num_tokens": 7936844.0,
      "step": 485
    },
    {
      "entropy": 0.5356980487704277,
      "epoch": 1.8134328358208955,
      "grad_norm": 0.1346481293439865,
      "learning_rate": 0.0002,
      "loss": 0.533320426940918,
      "mean_token_accuracy": 0.7837155610322952,
      "num_tokens": 7953142.0,
      "step": 486
    },
    {
      "entropy": 0.528105616569519,
      "epoch": 1.8171641791044775,
      "grad_norm": 0.1391320377588272,
      "learning_rate": 0.0002,
      "loss": 0.5345504879951477,
      "mean_token_accuracy": 0.784235417842865,
      "num_tokens": 7969458.0,
      "step": 487
    },
    {
      "entropy": 0.5304473042488098,
      "epoch": 1.8208955223880596,
      "grad_norm": 0.1381186991930008,
      "learning_rate": 0.0002,
      "loss": 0.5353981852531433,
      "mean_token_accuracy": 0.7815601527690887,
      "num_tokens": 7985948.0,
      "step": 488
    },
    {
      "entropy": 0.5339603275060654,
      "epoch": 1.8246268656716418,
      "grad_norm": 0.12126309424638748,
      "learning_rate": 0.0002,
      "loss": 0.53676837682724,
      "mean_token_accuracy": 0.7817868292331696,
      "num_tokens": 8002545.0,
      "step": 489
    },
    {
      "entropy": 0.542829766869545,
      "epoch": 1.828358208955224,
      "grad_norm": 0.15814469754695892,
      "learning_rate": 0.0002,
      "loss": 0.5454267263412476,
      "mean_token_accuracy": 0.7802938371896744,
      "num_tokens": 8018823.0,
      "step": 490
    },
    {
      "entropy": 0.5319127887487411,
      "epoch": 1.832089552238806,
      "grad_norm": 0.17742769420146942,
      "learning_rate": 0.0002,
      "loss": 0.5341488122940063,
      "mean_token_accuracy": 0.7826338112354279,
      "num_tokens": 8035341.0,
      "step": 491
    },
    {
      "entropy": 0.5450225174427032,
      "epoch": 1.835820895522388,
      "grad_norm": 0.16771593689918518,
      "learning_rate": 0.0002,
      "loss": 0.5443899035453796,
      "mean_token_accuracy": 0.7811568826436996,
      "num_tokens": 8051642.0,
      "step": 492
    },
    {
      "entropy": 0.5478601604700089,
      "epoch": 1.8395522388059702,
      "grad_norm": 0.16015928983688354,
      "learning_rate": 0.0002,
      "loss": 0.5466909408569336,
      "mean_token_accuracy": 0.7799689322710037,
      "num_tokens": 8068174.0,
      "step": 493
    },
    {
      "entropy": 0.5199587792158127,
      "epoch": 1.8432835820895521,
      "grad_norm": 0.18532446026802063,
      "learning_rate": 0.0002,
      "loss": 0.51955246925354,
      "mean_token_accuracy": 0.7891423404216766,
      "num_tokens": 8084364.0,
      "step": 494
    },
    {
      "entropy": 0.5241135209798813,
      "epoch": 1.8470149253731343,
      "grad_norm": 0.1427345871925354,
      "learning_rate": 0.0002,
      "loss": 0.5248921513557434,
      "mean_token_accuracy": 0.787857785820961,
      "num_tokens": 8100628.0,
      "step": 495
    },
    {
      "entropy": 0.5340007096529007,
      "epoch": 1.8507462686567164,
      "grad_norm": 0.17814995348453522,
      "learning_rate": 0.0002,
      "loss": 0.540264368057251,
      "mean_token_accuracy": 0.7790678143501282,
      "num_tokens": 8116959.0,
      "step": 496
    },
    {
      "entropy": 0.527913436293602,
      "epoch": 1.8544776119402986,
      "grad_norm": 0.13427263498306274,
      "learning_rate": 0.0002,
      "loss": 0.5335310697555542,
      "mean_token_accuracy": 0.7840446978807449,
      "num_tokens": 8133469.0,
      "step": 497
    },
    {
      "entropy": 0.5294159948825836,
      "epoch": 1.8582089552238807,
      "grad_norm": 0.15380533039569855,
      "learning_rate": 0.0002,
      "loss": 0.5384203791618347,
      "mean_token_accuracy": 0.7821631729602814,
      "num_tokens": 8149696.0,
      "step": 498
    },
    {
      "entropy": 0.551758736371994,
      "epoch": 1.8619402985074627,
      "grad_norm": 0.14327947795391083,
      "learning_rate": 0.0002,
      "loss": 0.5482276678085327,
      "mean_token_accuracy": 0.7767462432384491,
      "num_tokens": 8165879.0,
      "step": 499
    },
    {
      "entropy": 0.5242659151554108,
      "epoch": 1.8656716417910446,
      "grad_norm": 0.12858544290065765,
      "learning_rate": 0.0002,
      "loss": 0.5157872438430786,
      "mean_token_accuracy": 0.7902811467647552,
      "num_tokens": 8182190.0,
      "step": 500
    },
    {
      "entropy": 0.5306264460086823,
      "epoch": 1.8694029850746268,
      "grad_norm": 0.15147298574447632,
      "learning_rate": 0.0002,
      "loss": 0.5306845903396606,
      "mean_token_accuracy": 0.7868326008319855,
      "num_tokens": 8198532.0,
      "step": 501
    },
    {
      "entropy": 0.5289736092090607,
      "epoch": 1.873134328358209,
      "grad_norm": 0.1485019028186798,
      "learning_rate": 0.0002,
      "loss": 0.5295736789703369,
      "mean_token_accuracy": 0.7867529690265656,
      "num_tokens": 8214876.0,
      "step": 502
    },
    {
      "entropy": 0.5189197212457657,
      "epoch": 1.876865671641791,
      "grad_norm": 0.16673879325389862,
      "learning_rate": 0.0002,
      "loss": 0.530517578125,
      "mean_token_accuracy": 0.7832315862178802,
      "num_tokens": 8231372.0,
      "step": 503
    },
    {
      "entropy": 0.5216372013092041,
      "epoch": 1.8805970149253732,
      "grad_norm": 0.14986100792884827,
      "learning_rate": 0.0002,
      "loss": 0.5253111720085144,
      "mean_token_accuracy": 0.7875290215015411,
      "num_tokens": 8247818.0,
      "step": 504
    },
    {
      "entropy": 0.5231712907552719,
      "epoch": 1.8843283582089554,
      "grad_norm": 0.1456281542778015,
      "learning_rate": 0.0002,
      "loss": 0.5197166204452515,
      "mean_token_accuracy": 0.7897316515445709,
      "num_tokens": 8264150.0,
      "step": 505
    },
    {
      "entropy": 0.5563454926013947,
      "epoch": 1.8880597014925373,
      "grad_norm": 0.16623522341251373,
      "learning_rate": 0.0002,
      "loss": 0.5501665472984314,
      "mean_token_accuracy": 0.7771624773740768,
      "num_tokens": 8280747.0,
      "step": 506
    },
    {
      "entropy": 0.5328278690576553,
      "epoch": 1.8917910447761193,
      "grad_norm": 0.14126084744930267,
      "learning_rate": 0.0002,
      "loss": 0.5289823412895203,
      "mean_token_accuracy": 0.7830969095230103,
      "num_tokens": 8296683.0,
      "step": 507
    },
    {
      "entropy": 0.530360110104084,
      "epoch": 1.8955223880597014,
      "grad_norm": 0.1487768143415451,
      "learning_rate": 0.0002,
      "loss": 0.5299562215805054,
      "mean_token_accuracy": 0.7846151441335678,
      "num_tokens": 8313098.0,
      "step": 508
    },
    {
      "entropy": 0.5228631570935249,
      "epoch": 1.8992537313432836,
      "grad_norm": 0.1883818805217743,
      "learning_rate": 0.0002,
      "loss": 0.5329415798187256,
      "mean_token_accuracy": 0.7835862636566162,
      "num_tokens": 8329292.0,
      "step": 509
    },
    {
      "entropy": 0.5307874977588654,
      "epoch": 1.9029850746268657,
      "grad_norm": 0.17275580763816833,
      "learning_rate": 0.0002,
      "loss": 0.5390638113021851,
      "mean_token_accuracy": 0.7791903018951416,
      "num_tokens": 8345581.0,
      "step": 510
    },
    {
      "entropy": 0.5359426289796829,
      "epoch": 1.9067164179104479,
      "grad_norm": 0.14612574875354767,
      "learning_rate": 0.0002,
      "loss": 0.5378193259239197,
      "mean_token_accuracy": 0.7792247533798218,
      "num_tokens": 8361871.0,
      "step": 511
    },
    {
      "entropy": 0.539593517780304,
      "epoch": 1.9104477611940298,
      "grad_norm": 0.16161257028579712,
      "learning_rate": 0.0002,
      "loss": 0.5343602299690247,
      "mean_token_accuracy": 0.7851338088512421,
      "num_tokens": 8378103.0,
      "step": 512
    },
    {
      "entropy": 0.5392905324697495,
      "epoch": 1.914179104477612,
      "grad_norm": 0.12597458064556122,
      "learning_rate": 0.0002,
      "loss": 0.530954122543335,
      "mean_token_accuracy": 0.7846523523330688,
      "num_tokens": 8394348.0,
      "step": 513
    },
    {
      "entropy": 0.547580674290657,
      "epoch": 1.917910447761194,
      "grad_norm": 0.17028233408927917,
      "learning_rate": 0.0002,
      "loss": 0.5528735518455505,
      "mean_token_accuracy": 0.7771532535552979,
      "num_tokens": 8410989.0,
      "step": 514
    },
    {
      "entropy": 0.5490703731775284,
      "epoch": 1.921641791044776,
      "grad_norm": 0.1378035694360733,
      "learning_rate": 0.0002,
      "loss": 0.5464003086090088,
      "mean_token_accuracy": 0.7778326570987701,
      "num_tokens": 8427380.0,
      "step": 515
    },
    {
      "entropy": 0.5565768778324127,
      "epoch": 1.9253731343283582,
      "grad_norm": 0.1366737335920334,
      "learning_rate": 0.0002,
      "loss": 0.5584871172904968,
      "mean_token_accuracy": 0.7713783830404282,
      "num_tokens": 8443919.0,
      "step": 516
    },
    {
      "entropy": 0.5345391035079956,
      "epoch": 1.9291044776119404,
      "grad_norm": 0.1527157723903656,
      "learning_rate": 0.0002,
      "loss": 0.5405983328819275,
      "mean_token_accuracy": 0.779501810669899,
      "num_tokens": 8460174.0,
      "step": 517
    },
    {
      "entropy": 0.544240266084671,
      "epoch": 1.9328358208955225,
      "grad_norm": 0.14191998541355133,
      "learning_rate": 0.0002,
      "loss": 0.547295331954956,
      "mean_token_accuracy": 0.7777374982833862,
      "num_tokens": 8476737.0,
      "step": 518
    },
    {
      "entropy": 0.5293041467666626,
      "epoch": 1.9365671641791045,
      "grad_norm": 0.15670642256736755,
      "learning_rate": 0.0002,
      "loss": 0.5309778451919556,
      "mean_token_accuracy": 0.7865030914545059,
      "num_tokens": 8492947.0,
      "step": 519
    },
    {
      "entropy": 0.5330537855625153,
      "epoch": 1.9402985074626866,
      "grad_norm": 0.15017330646514893,
      "learning_rate": 0.0002,
      "loss": 0.5395172238349915,
      "mean_token_accuracy": 0.7815082669258118,
      "num_tokens": 8509337.0,
      "step": 520
    },
    {
      "entropy": 0.5353135764598846,
      "epoch": 1.9440298507462686,
      "grad_norm": 0.14300917088985443,
      "learning_rate": 0.0002,
      "loss": 0.5328524112701416,
      "mean_token_accuracy": 0.7861177921295166,
      "num_tokens": 8525672.0,
      "step": 521
    },
    {
      "entropy": 0.530888095498085,
      "epoch": 1.9477611940298507,
      "grad_norm": 0.1593046933412552,
      "learning_rate": 0.0002,
      "loss": 0.53388512134552,
      "mean_token_accuracy": 0.7821628004312515,
      "num_tokens": 8541976.0,
      "step": 522
    },
    {
      "entropy": 0.5539785474538803,
      "epoch": 1.9514925373134329,
      "grad_norm": 0.12615355849266052,
      "learning_rate": 0.0002,
      "loss": 0.5512434840202332,
      "mean_token_accuracy": 0.7770510613918304,
      "num_tokens": 8558299.0,
      "step": 523
    },
    {
      "entropy": 0.518076553940773,
      "epoch": 1.955223880597015,
      "grad_norm": 0.15514200925827026,
      "learning_rate": 0.0002,
      "loss": 0.5218104124069214,
      "mean_token_accuracy": 0.7879758924245834,
      "num_tokens": 8574716.0,
      "step": 524
    },
    {
      "entropy": 0.5353099256753922,
      "epoch": 1.9589552238805972,
      "grad_norm": 0.13487055897712708,
      "learning_rate": 0.0002,
      "loss": 0.538606584072113,
      "mean_token_accuracy": 0.781571164727211,
      "num_tokens": 8591050.0,
      "step": 525
    },
    {
      "entropy": 0.5266296789050102,
      "epoch": 1.962686567164179,
      "grad_norm": 0.1971937119960785,
      "learning_rate": 0.0002,
      "loss": 0.5315322279930115,
      "mean_token_accuracy": 0.7832879722118378,
      "num_tokens": 8607133.0,
      "step": 526
    },
    {
      "entropy": 0.5294190347194672,
      "epoch": 1.966417910447761,
      "grad_norm": 0.14176510274410248,
      "learning_rate": 0.0002,
      "loss": 0.5304378867149353,
      "mean_token_accuracy": 0.7834701687097549,
      "num_tokens": 8623512.0,
      "step": 527
    },
    {
      "entropy": 0.5403010845184326,
      "epoch": 1.9701492537313432,
      "grad_norm": 0.13602930307388306,
      "learning_rate": 0.0002,
      "loss": 0.5389670133590698,
      "mean_token_accuracy": 0.780187577009201,
      "num_tokens": 8639953.0,
      "step": 528
    },
    {
      "entropy": 0.5418667942285538,
      "epoch": 1.9738805970149254,
      "grad_norm": 0.15750795602798462,
      "learning_rate": 0.0002,
      "loss": 0.5364986658096313,
      "mean_token_accuracy": 0.7812219262123108,
      "num_tokens": 8656318.0,
      "step": 529
    },
    {
      "entropy": 0.5283338874578476,
      "epoch": 1.9776119402985075,
      "grad_norm": 0.1357993483543396,
      "learning_rate": 0.0002,
      "loss": 0.5258264541625977,
      "mean_token_accuracy": 0.7875591516494751,
      "num_tokens": 8672429.0,
      "step": 530
    },
    {
      "entropy": 0.508994422852993,
      "epoch": 1.9813432835820897,
      "grad_norm": 0.14409013092517853,
      "learning_rate": 0.0002,
      "loss": 0.5108156800270081,
      "mean_token_accuracy": 0.7937235236167908,
      "num_tokens": 8688776.0,
      "step": 531
    },
    {
      "entropy": 0.5443366467952728,
      "epoch": 1.9850746268656716,
      "grad_norm": 0.1645997315645218,
      "learning_rate": 0.0002,
      "loss": 0.5518733859062195,
      "mean_token_accuracy": 0.7764965444803238,
      "num_tokens": 8705134.0,
      "step": 532
    },
    {
      "entropy": 0.5353538542985916,
      "epoch": 1.9888059701492538,
      "grad_norm": 0.13118219375610352,
      "learning_rate": 0.0002,
      "loss": 0.5298490524291992,
      "mean_token_accuracy": 0.7852030396461487,
      "num_tokens": 8721422.0,
      "step": 533
    },
    {
      "entropy": 0.5499696731567383,
      "epoch": 1.9925373134328357,
      "grad_norm": 0.14602512121200562,
      "learning_rate": 0.0002,
      "loss": 0.5541141629219055,
      "mean_token_accuracy": 0.7733804285526276,
      "num_tokens": 8737587.0,
      "step": 534
    },
    {
      "entropy": 0.5241136103868484,
      "epoch": 1.9962686567164178,
      "grad_norm": 0.13069047033786774,
      "learning_rate": 0.0002,
      "loss": 0.5247742533683777,
      "mean_token_accuracy": 0.7877077162265778,
      "num_tokens": 8753886.0,
      "step": 535
    },
    {
      "entropy": 0.5522785931825638,
      "epoch": 2.0,
      "grad_norm": 0.13797929883003235,
      "learning_rate": 0.0002,
      "loss": 0.5517363548278809,
      "mean_token_accuracy": 0.7764420509338379,
      "num_tokens": 8770409.0,
      "step": 536
    },
    {
      "entropy": 0.5294849127531052,
      "epoch": 2.003731343283582,
      "grad_norm": 0.125420480966568,
      "learning_rate": 0.0002,
      "loss": 0.5204938054084778,
      "mean_token_accuracy": 0.7905426770448685,
      "num_tokens": 8786951.0,
      "step": 537
    },
    {
      "entropy": 0.5258107706904411,
      "epoch": 2.0074626865671643,
      "grad_norm": 0.1486080288887024,
      "learning_rate": 0.0002,
      "loss": 0.5231454372406006,
      "mean_token_accuracy": 0.7883965820074081,
      "num_tokens": 8803315.0,
      "step": 538
    },
    {
      "entropy": 0.5175661221146584,
      "epoch": 2.0111940298507465,
      "grad_norm": 0.13154980540275574,
      "learning_rate": 0.0002,
      "loss": 0.5184586644172668,
      "mean_token_accuracy": 0.7910201251506805,
      "num_tokens": 8819911.0,
      "step": 539
    },
    {
      "entropy": 0.4994604140520096,
      "epoch": 2.014925373134328,
      "grad_norm": 0.16595442593097687,
      "learning_rate": 0.0002,
      "loss": 0.5148791074752808,
      "mean_token_accuracy": 0.7942434102296829,
      "num_tokens": 8836154.0,
      "step": 540
    },
    {
      "entropy": 0.5127262026071548,
      "epoch": 2.0186567164179103,
      "grad_norm": 0.1592157632112503,
      "learning_rate": 0.0002,
      "loss": 0.5228517651557922,
      "mean_token_accuracy": 0.7881066054105759,
      "num_tokens": 8852520.0,
      "step": 541
    },
    {
      "entropy": 0.5340550392866135,
      "epoch": 2.0223880597014925,
      "grad_norm": 0.17649918794631958,
      "learning_rate": 0.0002,
      "loss": 0.5372640490531921,
      "mean_token_accuracy": 0.7816326469182968,
      "num_tokens": 8868924.0,
      "step": 542
    },
    {
      "entropy": 0.5152621045708656,
      "epoch": 2.0261194029850746,
      "grad_norm": 0.14082451164722443,
      "learning_rate": 0.0002,
      "loss": 0.5027863383293152,
      "mean_token_accuracy": 0.7955759316682816,
      "num_tokens": 8885106.0,
      "step": 543
    },
    {
      "entropy": 0.522370234131813,
      "epoch": 2.029850746268657,
      "grad_norm": 0.14631766080856323,
      "learning_rate": 0.0002,
      "loss": 0.5158503651618958,
      "mean_token_accuracy": 0.7942074984312057,
      "num_tokens": 8901600.0,
      "step": 544
    },
    {
      "entropy": 0.5079411640763283,
      "epoch": 2.033582089552239,
      "grad_norm": 0.13866929709911346,
      "learning_rate": 0.0002,
      "loss": 0.4975392818450928,
      "mean_token_accuracy": 0.7998307049274445,
      "num_tokens": 8918068.0,
      "step": 545
    },
    {
      "entropy": 0.5192210525274277,
      "epoch": 2.0373134328358207,
      "grad_norm": 0.15910667181015015,
      "learning_rate": 0.0002,
      "loss": 0.5216242671012878,
      "mean_token_accuracy": 0.7882837206125259,
      "num_tokens": 8934441.0,
      "step": 546
    },
    {
      "entropy": 0.49886780977249146,
      "epoch": 2.041044776119403,
      "grad_norm": 0.1706812083721161,
      "learning_rate": 0.0002,
      "loss": 0.5059823989868164,
      "mean_token_accuracy": 0.7956507951021194,
      "num_tokens": 8950731.0,
      "step": 547
    },
    {
      "entropy": 0.5080099627375603,
      "epoch": 2.044776119402985,
      "grad_norm": 0.18679271638393402,
      "learning_rate": 0.0002,
      "loss": 0.5219873189926147,
      "mean_token_accuracy": 0.7893132120370865,
      "num_tokens": 8966920.0,
      "step": 548
    },
    {
      "entropy": 0.5161280035972595,
      "epoch": 2.048507462686567,
      "grad_norm": 0.15718749165534973,
      "learning_rate": 0.0002,
      "loss": 0.5155758857727051,
      "mean_token_accuracy": 0.7914890348911285,
      "num_tokens": 8983333.0,
      "step": 549
    },
    {
      "entropy": 0.5230599641799927,
      "epoch": 2.0522388059701493,
      "grad_norm": 0.16827784478664398,
      "learning_rate": 0.0002,
      "loss": 0.5158221125602722,
      "mean_token_accuracy": 0.7917670756578445,
      "num_tokens": 8999615.0,
      "step": 550
    },
    {
      "entropy": 0.5276069939136505,
      "epoch": 2.0559701492537314,
      "grad_norm": 0.15786929428577423,
      "learning_rate": 0.0002,
      "loss": 0.5228967070579529,
      "mean_token_accuracy": 0.7900322675704956,
      "num_tokens": 9016070.0,
      "step": 551
    },
    {
      "entropy": 0.5261552929878235,
      "epoch": 2.0597014925373136,
      "grad_norm": 0.179435133934021,
      "learning_rate": 0.0002,
      "loss": 0.5274342894554138,
      "mean_token_accuracy": 0.7874346524477005,
      "num_tokens": 9032325.0,
      "step": 552
    },
    {
      "entropy": 0.508784145116806,
      "epoch": 2.0634328358208953,
      "grad_norm": 0.18862098455429077,
      "learning_rate": 0.0002,
      "loss": 0.5171835422515869,
      "mean_token_accuracy": 0.7909427434206009,
      "num_tokens": 9048544.0,
      "step": 553
    },
    {
      "entropy": 0.5254177302122116,
      "epoch": 2.0671641791044775,
      "grad_norm": 0.15294437110424042,
      "learning_rate": 0.0002,
      "loss": 0.5267332792282104,
      "mean_token_accuracy": 0.7861658930778503,
      "num_tokens": 9064928.0,
      "step": 554
    },
    {
      "entropy": 0.5087190493941307,
      "epoch": 2.0708955223880596,
      "grad_norm": 0.14013507962226868,
      "learning_rate": 0.0002,
      "loss": 0.5050034523010254,
      "mean_token_accuracy": 0.7958011031150818,
      "num_tokens": 9081236.0,
      "step": 555
    },
    {
      "entropy": 0.5136034488677979,
      "epoch": 2.074626865671642,
      "grad_norm": 0.17949984967708588,
      "learning_rate": 0.0002,
      "loss": 0.5207123756408691,
      "mean_token_accuracy": 0.7871850281953812,
      "num_tokens": 9097632.0,
      "step": 556
    },
    {
      "entropy": 0.5167653933167458,
      "epoch": 2.078358208955224,
      "grad_norm": 0.17791901528835297,
      "learning_rate": 0.0002,
      "loss": 0.5193839073181152,
      "mean_token_accuracy": 0.7898633778095245,
      "num_tokens": 9114047.0,
      "step": 557
    },
    {
      "entropy": 0.5242370218038559,
      "epoch": 2.082089552238806,
      "grad_norm": 0.18635064363479614,
      "learning_rate": 0.0002,
      "loss": 0.5251120328903198,
      "mean_token_accuracy": 0.7863645255565643,
      "num_tokens": 9130395.0,
      "step": 558
    },
    {
      "entropy": 0.5116435959935188,
      "epoch": 2.0858208955223883,
      "grad_norm": 0.17594675719738007,
      "learning_rate": 0.0002,
      "loss": 0.5087394714355469,
      "mean_token_accuracy": 0.7953201085329056,
      "num_tokens": 9146761.0,
      "step": 559
    },
    {
      "entropy": 0.5367572158575058,
      "epoch": 2.08955223880597,
      "grad_norm": 0.16770021617412567,
      "learning_rate": 0.0002,
      "loss": 0.5259926915168762,
      "mean_token_accuracy": 0.7842537760734558,
      "num_tokens": 9163292.0,
      "step": 560
    },
    {
      "entropy": 0.5182780101895332,
      "epoch": 2.093283582089552,
      "grad_norm": 0.15339073538780212,
      "learning_rate": 0.0002,
      "loss": 0.5159294009208679,
      "mean_token_accuracy": 0.7939708232879639,
      "num_tokens": 9179681.0,
      "step": 561
    },
    {
      "entropy": 0.5029266253113747,
      "epoch": 2.0970149253731343,
      "grad_norm": 0.2085951417684555,
      "learning_rate": 0.0002,
      "loss": 0.5057911276817322,
      "mean_token_accuracy": 0.7953764498233795,
      "num_tokens": 9196009.0,
      "step": 562
    },
    {
      "entropy": 0.5088721960783005,
      "epoch": 2.1007462686567164,
      "grad_norm": 0.23663504421710968,
      "learning_rate": 0.0002,
      "loss": 0.5137120485305786,
      "mean_token_accuracy": 0.7930423617362976,
      "num_tokens": 9212477.0,
      "step": 563
    },
    {
      "entropy": 0.5090882033109665,
      "epoch": 2.1044776119402986,
      "grad_norm": 0.17074984312057495,
      "learning_rate": 0.0002,
      "loss": 0.5111451148986816,
      "mean_token_accuracy": 0.7952387928962708,
      "num_tokens": 9228991.0,
      "step": 564
    },
    {
      "entropy": 0.5197137892246246,
      "epoch": 2.1082089552238807,
      "grad_norm": 0.18768630921840668,
      "learning_rate": 0.0002,
      "loss": 0.5150706171989441,
      "mean_token_accuracy": 0.7908357232809067,
      "num_tokens": 9245501.0,
      "step": 565
    },
    {
      "entropy": 0.5140221863985062,
      "epoch": 2.111940298507463,
      "grad_norm": 0.1833381950855255,
      "learning_rate": 0.0002,
      "loss": 0.5144209861755371,
      "mean_token_accuracy": 0.7924097031354904,
      "num_tokens": 9261721.0,
      "step": 566
    },
    {
      "entropy": 0.5247924327850342,
      "epoch": 2.1156716417910446,
      "grad_norm": 0.19777074456214905,
      "learning_rate": 0.0002,
      "loss": 0.5174472332000732,
      "mean_token_accuracy": 0.7918887138366699,
      "num_tokens": 9278242.0,
      "step": 567
    },
    {
      "entropy": 0.4977032169699669,
      "epoch": 2.1194029850746268,
      "grad_norm": 0.19118627905845642,
      "learning_rate": 0.0002,
      "loss": 0.5060408115386963,
      "mean_token_accuracy": 0.79521843791008,
      "num_tokens": 9294474.0,
      "step": 568
    },
    {
      "entropy": 0.4930529072880745,
      "epoch": 2.123134328358209,
      "grad_norm": 0.17287500202655792,
      "learning_rate": 0.0002,
      "loss": 0.5063955783843994,
      "mean_token_accuracy": 0.7964753955602646,
      "num_tokens": 9310448.0,
      "step": 569
    },
    {
      "entropy": 0.4997175335884094,
      "epoch": 2.126865671641791,
      "grad_norm": 0.19182001054286957,
      "learning_rate": 0.0002,
      "loss": 0.511003315448761,
      "mean_token_accuracy": 0.7929231375455856,
      "num_tokens": 9326574.0,
      "step": 570
    },
    {
      "entropy": 0.5178787037730217,
      "epoch": 2.1305970149253732,
      "grad_norm": 0.16347914934158325,
      "learning_rate": 0.0002,
      "loss": 0.5087844729423523,
      "mean_token_accuracy": 0.7940262705087662,
      "num_tokens": 9342970.0,
      "step": 571
    },
    {
      "entropy": 0.5362161993980408,
      "epoch": 2.1343283582089554,
      "grad_norm": 0.16588380932807922,
      "learning_rate": 0.0002,
      "loss": 0.5250737071037292,
      "mean_token_accuracy": 0.7876630574464798,
      "num_tokens": 9359168.0,
      "step": 572
    },
    {
      "entropy": 0.532064899802208,
      "epoch": 2.138059701492537,
      "grad_norm": 0.15581250190734863,
      "learning_rate": 0.0002,
      "loss": 0.5236944556236267,
      "mean_token_accuracy": 0.790793389081955,
      "num_tokens": 9375380.0,
      "step": 573
    },
    {
      "entropy": 0.5272438824176788,
      "epoch": 2.1417910447761193,
      "grad_norm": 0.20483332872390747,
      "learning_rate": 0.0002,
      "loss": 0.5327727198600769,
      "mean_token_accuracy": 0.7848416119813919,
      "num_tokens": 9391851.0,
      "step": 574
    },
    {
      "entropy": 0.5178545862436295,
      "epoch": 2.1455223880597014,
      "grad_norm": 0.17937517166137695,
      "learning_rate": 0.0002,
      "loss": 0.520628809928894,
      "mean_token_accuracy": 0.7914203107357025,
      "num_tokens": 9408515.0,
      "step": 575
    },
    {
      "entropy": 0.5144930109381676,
      "epoch": 2.1492537313432836,
      "grad_norm": 0.20648981630802155,
      "learning_rate": 0.0002,
      "loss": 0.5262653827667236,
      "mean_token_accuracy": 0.7883694916963577,
      "num_tokens": 9424748.0,
      "step": 576
    },
    {
      "entropy": 0.5081476420164108,
      "epoch": 2.1529850746268657,
      "grad_norm": 0.15983696281909943,
      "learning_rate": 0.0002,
      "loss": 0.5091989040374756,
      "mean_token_accuracy": 0.7932504862546921,
      "num_tokens": 9440987.0,
      "step": 577
    },
    {
      "entropy": 0.5091744139790535,
      "epoch": 2.156716417910448,
      "grad_norm": 0.143747940659523,
      "learning_rate": 0.0002,
      "loss": 0.507089376449585,
      "mean_token_accuracy": 0.794156089425087,
      "num_tokens": 9457349.0,
      "step": 578
    },
    {
      "entropy": 0.5104940757155418,
      "epoch": 2.16044776119403,
      "grad_norm": 0.16569356620311737,
      "learning_rate": 0.0002,
      "loss": 0.5061535239219666,
      "mean_token_accuracy": 0.7964262366294861,
      "num_tokens": 9473817.0,
      "step": 579
    },
    {
      "entropy": 0.5086464136838913,
      "epoch": 2.1641791044776117,
      "grad_norm": 0.179887592792511,
      "learning_rate": 0.0002,
      "loss": 0.5004557371139526,
      "mean_token_accuracy": 0.7990415841341019,
      "num_tokens": 9490110.0,
      "step": 580
    },
    {
      "entropy": 0.5114818289875984,
      "epoch": 2.167910447761194,
      "grad_norm": 0.2178039252758026,
      "learning_rate": 0.0002,
      "loss": 0.5228156447410583,
      "mean_token_accuracy": 0.7900251597166061,
      "num_tokens": 9506558.0,
      "step": 581
    },
    {
      "entropy": 0.5055092200636864,
      "epoch": 2.171641791044776,
      "grad_norm": 0.17143648862838745,
      "learning_rate": 0.0002,
      "loss": 0.5139644145965576,
      "mean_token_accuracy": 0.7933343797922134,
      "num_tokens": 9522674.0,
      "step": 582
    },
    {
      "entropy": 0.5209199637174606,
      "epoch": 2.175373134328358,
      "grad_norm": 0.1963517814874649,
      "learning_rate": 0.0002,
      "loss": 0.5282465219497681,
      "mean_token_accuracy": 0.7860098034143448,
      "num_tokens": 9538967.0,
      "step": 583
    },
    {
      "entropy": 0.5108934044837952,
      "epoch": 2.1791044776119404,
      "grad_norm": 0.14624883234500885,
      "learning_rate": 0.0002,
      "loss": 0.5060119032859802,
      "mean_token_accuracy": 0.7967799454927444,
      "num_tokens": 9555308.0,
      "step": 584
    },
    {
      "entropy": 0.5247770547866821,
      "epoch": 2.1828358208955225,
      "grad_norm": 0.16273266077041626,
      "learning_rate": 0.0002,
      "loss": 0.5228431224822998,
      "mean_token_accuracy": 0.788749948143959,
      "num_tokens": 9571783.0,
      "step": 585
    },
    {
      "entropy": 0.5204054489731789,
      "epoch": 2.1865671641791047,
      "grad_norm": 0.14925257861614227,
      "learning_rate": 0.0002,
      "loss": 0.5232412219047546,
      "mean_token_accuracy": 0.7892403602600098,
      "num_tokens": 9588235.0,
      "step": 586
    },
    {
      "entropy": 0.5165363550186157,
      "epoch": 2.1902985074626864,
      "grad_norm": 0.16366349160671234,
      "learning_rate": 0.0002,
      "loss": 0.5210464000701904,
      "mean_token_accuracy": 0.7882984429597855,
      "num_tokens": 9604408.0,
      "step": 587
    },
    {
      "entropy": 0.5106817856431007,
      "epoch": 2.1940298507462686,
      "grad_norm": 0.18039654195308685,
      "learning_rate": 0.0002,
      "loss": 0.5134552717208862,
      "mean_token_accuracy": 0.7908795177936554,
      "num_tokens": 9620552.0,
      "step": 588
    },
    {
      "entropy": 0.5111690908670425,
      "epoch": 2.1977611940298507,
      "grad_norm": 0.1635941118001938,
      "learning_rate": 0.0002,
      "loss": 0.5129784941673279,
      "mean_token_accuracy": 0.7909968048334122,
      "num_tokens": 9636913.0,
      "step": 589
    },
    {
      "entropy": 0.5287658274173737,
      "epoch": 2.201492537313433,
      "grad_norm": 0.18495672941207886,
      "learning_rate": 0.0002,
      "loss": 0.5281989574432373,
      "mean_token_accuracy": 0.784828245639801,
      "num_tokens": 9653251.0,
      "step": 590
    },
    {
      "entropy": 0.5366753935813904,
      "epoch": 2.205223880597015,
      "grad_norm": 0.17156296968460083,
      "learning_rate": 0.0002,
      "loss": 0.5293815732002258,
      "mean_token_accuracy": 0.7864246368408203,
      "num_tokens": 9669560.0,
      "step": 591
    },
    {
      "entropy": 0.5325558334589005,
      "epoch": 2.208955223880597,
      "grad_norm": 0.19953666627407074,
      "learning_rate": 0.0002,
      "loss": 0.5299435257911682,
      "mean_token_accuracy": 0.7853131592273712,
      "num_tokens": 9686058.0,
      "step": 592
    },
    {
      "entropy": 0.5221890658140182,
      "epoch": 2.2126865671641793,
      "grad_norm": 0.19219942390918732,
      "learning_rate": 0.0002,
      "loss": 0.5268661975860596,
      "mean_token_accuracy": 0.784967839717865,
      "num_tokens": 9702350.0,
      "step": 593
    },
    {
      "entropy": 0.5307768136262894,
      "epoch": 2.216417910447761,
      "grad_norm": 0.17264924943447113,
      "learning_rate": 0.0002,
      "loss": 0.5293731689453125,
      "mean_token_accuracy": 0.7857524007558823,
      "num_tokens": 9718930.0,
      "step": 594
    },
    {
      "entropy": 0.5272602289915085,
      "epoch": 2.220149253731343,
      "grad_norm": 0.19435909390449524,
      "learning_rate": 0.0002,
      "loss": 0.5310930609703064,
      "mean_token_accuracy": 0.786889910697937,
      "num_tokens": 9735287.0,
      "step": 595
    },
    {
      "entropy": 0.5084069296717644,
      "epoch": 2.2238805970149254,
      "grad_norm": 0.17865370213985443,
      "learning_rate": 0.0002,
      "loss": 0.5123623609542847,
      "mean_token_accuracy": 0.7948104739189148,
      "num_tokens": 9751646.0,
      "step": 596
    },
    {
      "entropy": 0.5117323100566864,
      "epoch": 2.2276119402985075,
      "grad_norm": 0.17900389432907104,
      "learning_rate": 0.0002,
      "loss": 0.5196022987365723,
      "mean_token_accuracy": 0.7908018082380295,
      "num_tokens": 9767886.0,
      "step": 597
    },
    {
      "entropy": 0.5059385448694229,
      "epoch": 2.2313432835820897,
      "grad_norm": 0.17656244337558746,
      "learning_rate": 0.0002,
      "loss": 0.5047177672386169,
      "mean_token_accuracy": 0.7968951612710953,
      "num_tokens": 9784353.0,
      "step": 598
    },
    {
      "entropy": 0.515342965722084,
      "epoch": 2.235074626865672,
      "grad_norm": 0.17201915383338928,
      "learning_rate": 0.0002,
      "loss": 0.5150702595710754,
      "mean_token_accuracy": 0.7913296669721603,
      "num_tokens": 9800514.0,
      "step": 599
    },
    {
      "entropy": 0.5440414994955063,
      "epoch": 2.2388059701492535,
      "grad_norm": 0.17017365992069244,
      "learning_rate": 0.0002,
      "loss": 0.5365599989891052,
      "mean_token_accuracy": 0.7842509299516678,
      "num_tokens": 9817052.0,
      "step": 600
    },
    {
      "entropy": 0.5261255204677582,
      "epoch": 2.2425373134328357,
      "grad_norm": 0.17153213918209076,
      "learning_rate": 0.0002,
      "loss": 0.5241469740867615,
      "mean_token_accuracy": 0.7881980240345001,
      "num_tokens": 9833301.0,
      "step": 601
    },
    {
      "entropy": 0.5091669335961342,
      "epoch": 2.246268656716418,
      "grad_norm": 0.17429564893245697,
      "learning_rate": 0.0002,
      "loss": 0.5136046409606934,
      "mean_token_accuracy": 0.7911395728588104,
      "num_tokens": 9849553.0,
      "step": 602
    },
    {
      "entropy": 0.5217274948954582,
      "epoch": 2.25,
      "grad_norm": 0.18611721694469452,
      "learning_rate": 0.0002,
      "loss": 0.5315898656845093,
      "mean_token_accuracy": 0.7870735377073288,
      "num_tokens": 9865877.0,
      "step": 603
    },
    {
      "entropy": 0.5227696150541306,
      "epoch": 2.253731343283582,
      "grad_norm": 0.18880440294742584,
      "learning_rate": 0.0002,
      "loss": 0.5296134352684021,
      "mean_token_accuracy": 0.7852792292833328,
      "num_tokens": 9882313.0,
      "step": 604
    },
    {
      "entropy": 0.5258260294795036,
      "epoch": 2.2574626865671643,
      "grad_norm": 0.14611785113811493,
      "learning_rate": 0.0002,
      "loss": 0.5206936597824097,
      "mean_token_accuracy": 0.7903030216693878,
      "num_tokens": 9898839.0,
      "step": 605
    },
    {
      "entropy": 0.5399209856987,
      "epoch": 2.2611940298507465,
      "grad_norm": 0.14919111132621765,
      "learning_rate": 0.0002,
      "loss": 0.5303294062614441,
      "mean_token_accuracy": 0.7863423973321915,
      "num_tokens": 9915421.0,
      "step": 606
    },
    {
      "entropy": 0.5277860313653946,
      "epoch": 2.264925373134328,
      "grad_norm": 0.14488378167152405,
      "learning_rate": 0.0002,
      "loss": 0.519542396068573,
      "mean_token_accuracy": 0.7928669452667236,
      "num_tokens": 9931727.0,
      "step": 607
    },
    {
      "entropy": 0.4945507273077965,
      "epoch": 2.2686567164179103,
      "grad_norm": 0.1835111826658249,
      "learning_rate": 0.0002,
      "loss": 0.5029417276382446,
      "mean_token_accuracy": 0.7954000681638718,
      "num_tokens": 9947880.0,
      "step": 608
    },
    {
      "entropy": 0.5122526064515114,
      "epoch": 2.2723880597014925,
      "grad_norm": 0.17832306027412415,
      "learning_rate": 0.0002,
      "loss": 0.5254662036895752,
      "mean_token_accuracy": 0.7880468964576721,
      "num_tokens": 9964273.0,
      "step": 609
    },
    {
      "entropy": 0.5167535543441772,
      "epoch": 2.2761194029850746,
      "grad_norm": 0.1590995043516159,
      "learning_rate": 0.0002,
      "loss": 0.5195916295051575,
      "mean_token_accuracy": 0.787218302488327,
      "num_tokens": 9980854.0,
      "step": 610
    },
    {
      "entropy": 0.5098580792546272,
      "epoch": 2.279850746268657,
      "grad_norm": 0.15836626291275024,
      "learning_rate": 0.0002,
      "loss": 0.5124033689498901,
      "mean_token_accuracy": 0.7913301140069962,
      "num_tokens": 9996905.0,
      "step": 611
    },
    {
      "entropy": 0.5259643197059631,
      "epoch": 2.283582089552239,
      "grad_norm": 0.1523635983467102,
      "learning_rate": 0.0002,
      "loss": 0.5187619924545288,
      "mean_token_accuracy": 0.7937401235103607,
      "num_tokens": 10013405.0,
      "step": 612
    },
    {
      "entropy": 0.5277784913778305,
      "epoch": 2.2873134328358207,
      "grad_norm": 0.1649775356054306,
      "learning_rate": 0.0002,
      "loss": 0.5315808653831482,
      "mean_token_accuracy": 0.7855904698371887,
      "num_tokens": 10029667.0,
      "step": 613
    },
    {
      "entropy": 0.5262410789728165,
      "epoch": 2.291044776119403,
      "grad_norm": 0.15203334391117096,
      "learning_rate": 0.0002,
      "loss": 0.5255904197692871,
      "mean_token_accuracy": 0.7881879210472107,
      "num_tokens": 10046160.0,
      "step": 614
    },
    {
      "entropy": 0.5241505354642868,
      "epoch": 2.294776119402985,
      "grad_norm": 0.18789614737033844,
      "learning_rate": 0.0002,
      "loss": 0.5238255858421326,
      "mean_token_accuracy": 0.7843707501888275,
      "num_tokens": 10062453.0,
      "step": 615
    },
    {
      "entropy": 0.5166228264570236,
      "epoch": 2.298507462686567,
      "grad_norm": 0.1664339303970337,
      "learning_rate": 0.0002,
      "loss": 0.5110273361206055,
      "mean_token_accuracy": 0.7905164808034897,
      "num_tokens": 10078529.0,
      "step": 616
    },
    {
      "entropy": 0.5075492858886719,
      "epoch": 2.3022388059701493,
      "grad_norm": 0.22521668672561646,
      "learning_rate": 0.0002,
      "loss": 0.5120429396629333,
      "mean_token_accuracy": 0.7934232652187347,
      "num_tokens": 10094762.0,
      "step": 617
    },
    {
      "entropy": 0.506694033741951,
      "epoch": 2.3059701492537314,
      "grad_norm": 0.22079938650131226,
      "learning_rate": 0.0002,
      "loss": 0.5139238238334656,
      "mean_token_accuracy": 0.787568524479866,
      "num_tokens": 10110659.0,
      "step": 618
    },
    {
      "entropy": 0.5242674350738525,
      "epoch": 2.3097014925373136,
      "grad_norm": 0.1853271722793579,
      "learning_rate": 0.0002,
      "loss": 0.5255372524261475,
      "mean_token_accuracy": 0.7875581830739975,
      "num_tokens": 10127103.0,
      "step": 619
    },
    {
      "entropy": 0.5136331543326378,
      "epoch": 2.3134328358208958,
      "grad_norm": 0.1589234173297882,
      "learning_rate": 0.0002,
      "loss": 0.5141199827194214,
      "mean_token_accuracy": 0.7937167435884476,
      "num_tokens": 10143612.0,
      "step": 620
    },
    {
      "entropy": 0.524729534983635,
      "epoch": 2.3171641791044775,
      "grad_norm": 0.205324187874794,
      "learning_rate": 0.0002,
      "loss": 0.5216213464736938,
      "mean_token_accuracy": 0.7882188111543655,
      "num_tokens": 10159941.0,
      "step": 621
    },
    {
      "entropy": 0.5130689889192581,
      "epoch": 2.3208955223880596,
      "grad_norm": 0.16606491804122925,
      "learning_rate": 0.0002,
      "loss": 0.5043811798095703,
      "mean_token_accuracy": 0.7958787977695465,
      "num_tokens": 10176187.0,
      "step": 622
    },
    {
      "entropy": 0.5039496794342995,
      "epoch": 2.324626865671642,
      "grad_norm": 0.19112904369831085,
      "learning_rate": 0.0002,
      "loss": 0.5113927721977234,
      "mean_token_accuracy": 0.7904049158096313,
      "num_tokens": 10192620.0,
      "step": 623
    },
    {
      "entropy": 0.49748485535383224,
      "epoch": 2.328358208955224,
      "grad_norm": 0.19301722943782806,
      "learning_rate": 0.0002,
      "loss": 0.5107299089431763,
      "mean_token_accuracy": 0.7948975116014481,
      "num_tokens": 10208980.0,
      "step": 624
    },
    {
      "entropy": 0.5088533237576485,
      "epoch": 2.332089552238806,
      "grad_norm": 0.20937587320804596,
      "learning_rate": 0.0002,
      "loss": 0.5153738260269165,
      "mean_token_accuracy": 0.7925478518009186,
      "num_tokens": 10225269.0,
      "step": 625
    },
    {
      "entropy": 0.5225254744291306,
      "epoch": 2.3358208955223883,
      "grad_norm": 0.15447191894054413,
      "learning_rate": 0.0002,
      "loss": 0.5196532607078552,
      "mean_token_accuracy": 0.7922751158475876,
      "num_tokens": 10241676.0,
      "step": 626
    },
    {
      "entropy": 0.5326119512319565,
      "epoch": 2.33955223880597,
      "grad_norm": 0.1772749423980713,
      "learning_rate": 0.0002,
      "loss": 0.5275702476501465,
      "mean_token_accuracy": 0.7883079051971436,
      "num_tokens": 10258282.0,
      "step": 627
    },
    {
      "entropy": 0.5241412222385406,
      "epoch": 2.343283582089552,
      "grad_norm": 0.15652857720851898,
      "learning_rate": 0.0002,
      "loss": 0.5206055641174316,
      "mean_token_accuracy": 0.7899506539106369,
      "num_tokens": 10274618.0,
      "step": 628
    },
    {
      "entropy": 0.5035678222775459,
      "epoch": 2.3470149253731343,
      "grad_norm": 0.1610826700925827,
      "learning_rate": 0.0002,
      "loss": 0.5049140453338623,
      "mean_token_accuracy": 0.7945234477519989,
      "num_tokens": 10290740.0,
      "step": 629
    },
    {
      "entropy": 0.5056828930974007,
      "epoch": 2.3507462686567164,
      "grad_norm": 0.19641247391700745,
      "learning_rate": 0.0002,
      "loss": 0.5203961730003357,
      "mean_token_accuracy": 0.7882252931594849,
      "num_tokens": 10306908.0,
      "step": 630
    },
    {
      "entropy": 0.5376667827367783,
      "epoch": 2.3544776119402986,
      "grad_norm": 0.1892770230770111,
      "learning_rate": 0.0002,
      "loss": 0.5438809394836426,
      "mean_token_accuracy": 0.7808854281902313,
      "num_tokens": 10323600.0,
      "step": 631
    },
    {
      "entropy": 0.5257938951253891,
      "epoch": 2.3582089552238807,
      "grad_norm": 0.1734701544046402,
      "learning_rate": 0.0002,
      "loss": 0.5274419188499451,
      "mean_token_accuracy": 0.7882835865020752,
      "num_tokens": 10339842.0,
      "step": 632
    },
    {
      "entropy": 0.5335331857204437,
      "epoch": 2.361940298507463,
      "grad_norm": 0.15732470154762268,
      "learning_rate": 0.0002,
      "loss": 0.5302885174751282,
      "mean_token_accuracy": 0.7879149913787842,
      "num_tokens": 10356146.0,
      "step": 633
    },
    {
      "entropy": 0.5305976718664169,
      "epoch": 2.3656716417910446,
      "grad_norm": 0.14950533211231232,
      "learning_rate": 0.0002,
      "loss": 0.5217748880386353,
      "mean_token_accuracy": 0.7901571691036224,
      "num_tokens": 10372774.0,
      "step": 634
    },
    {
      "entropy": 0.5027497857809067,
      "epoch": 2.3694029850746268,
      "grad_norm": 0.1668434888124466,
      "learning_rate": 0.0002,
      "loss": 0.49668481945991516,
      "mean_token_accuracy": 0.7978196144104004,
      "num_tokens": 10389233.0,
      "step": 635
    },
    {
      "entropy": 0.5358419269323349,
      "epoch": 2.373134328358209,
      "grad_norm": 0.17945551872253418,
      "learning_rate": 0.0002,
      "loss": 0.5389994382858276,
      "mean_token_accuracy": 0.7802686244249344,
      "num_tokens": 10405879.0,
      "step": 636
    },
    {
      "entropy": 0.5104084759950638,
      "epoch": 2.376865671641791,
      "grad_norm": 0.17669576406478882,
      "learning_rate": 0.0002,
      "loss": 0.5137860178947449,
      "mean_token_accuracy": 0.7914855033159256,
      "num_tokens": 10422059.0,
      "step": 637
    },
    {
      "entropy": 0.5348759889602661,
      "epoch": 2.3805970149253732,
      "grad_norm": 0.17230568826198578,
      "learning_rate": 0.0002,
      "loss": 0.5333091020584106,
      "mean_token_accuracy": 0.7840609103441238,
      "num_tokens": 10438486.0,
      "step": 638
    },
    {
      "entropy": 0.5165744200348854,
      "epoch": 2.3843283582089554,
      "grad_norm": 0.17449212074279785,
      "learning_rate": 0.0002,
      "loss": 0.5179176330566406,
      "mean_token_accuracy": 0.7892343103885651,
      "num_tokens": 10454418.0,
      "step": 639
    },
    {
      "entropy": 0.5385671108961105,
      "epoch": 2.388059701492537,
      "grad_norm": 0.1890452355146408,
      "learning_rate": 0.0002,
      "loss": 0.5375429391860962,
      "mean_token_accuracy": 0.7834752649068832,
      "num_tokens": 10470842.0,
      "step": 640
    },
    {
      "entropy": 0.49661771208047867,
      "epoch": 2.3917910447761193,
      "grad_norm": 0.1732311248779297,
      "learning_rate": 0.0002,
      "loss": 0.49945348501205444,
      "mean_token_accuracy": 0.8011882454156876,
      "num_tokens": 10487044.0,
      "step": 641
    },
    {
      "entropy": 0.519076332449913,
      "epoch": 2.3955223880597014,
      "grad_norm": 0.18893247842788696,
      "learning_rate": 0.0002,
      "loss": 0.5197221040725708,
      "mean_token_accuracy": 0.7879463732242584,
      "num_tokens": 10502951.0,
      "step": 642
    },
    {
      "entropy": 0.5214255601167679,
      "epoch": 2.3992537313432836,
      "grad_norm": 0.20484893023967743,
      "learning_rate": 0.0002,
      "loss": 0.5267422795295715,
      "mean_token_accuracy": 0.7866235077381134,
      "num_tokens": 10519055.0,
      "step": 643
    },
    {
      "entropy": 0.5253995954990387,
      "epoch": 2.4029850746268657,
      "grad_norm": 0.1876876801252365,
      "learning_rate": 0.0002,
      "loss": 0.5317081212997437,
      "mean_token_accuracy": 0.7831747829914093,
      "num_tokens": 10535333.0,
      "step": 644
    },
    {
      "entropy": 0.5186711996793747,
      "epoch": 2.406716417910448,
      "grad_norm": 0.18439075350761414,
      "learning_rate": 0.0002,
      "loss": 0.5136252045631409,
      "mean_token_accuracy": 0.7919325232505798,
      "num_tokens": 10551621.0,
      "step": 645
    },
    {
      "entropy": 0.5383310168981552,
      "epoch": 2.41044776119403,
      "grad_norm": 0.19391457736492157,
      "learning_rate": 0.0002,
      "loss": 0.536406397819519,
      "mean_token_accuracy": 0.7821435630321503,
      "num_tokens": 10567850.0,
      "step": 646
    },
    {
      "entropy": 0.5125755220651627,
      "epoch": 2.4141791044776117,
      "grad_norm": 0.20207205414772034,
      "learning_rate": 0.0002,
      "loss": 0.5210078954696655,
      "mean_token_accuracy": 0.7870661616325378,
      "num_tokens": 10583999.0,
      "step": 647
    },
    {
      "entropy": 0.5297990292310715,
      "epoch": 2.417910447761194,
      "grad_norm": 0.20189954340457916,
      "learning_rate": 0.0002,
      "loss": 0.5326197743415833,
      "mean_token_accuracy": 0.7850360125303268,
      "num_tokens": 10600525.0,
      "step": 648
    },
    {
      "entropy": 0.5246636644005775,
      "epoch": 2.421641791044776,
      "grad_norm": 0.1728029102087021,
      "learning_rate": 0.0002,
      "loss": 0.525002121925354,
      "mean_token_accuracy": 0.7875626981258392,
      "num_tokens": 10616849.0,
      "step": 649
    },
    {
      "entropy": 0.5095944106578827,
      "epoch": 2.425373134328358,
      "grad_norm": 0.2387600839138031,
      "learning_rate": 0.0002,
      "loss": 0.5112872123718262,
      "mean_token_accuracy": 0.7921067625284195,
      "num_tokens": 10633233.0,
      "step": 650
    },
    {
      "entropy": 0.5263043940067291,
      "epoch": 2.4291044776119404,
      "grad_norm": 0.15833254158496857,
      "learning_rate": 0.0002,
      "loss": 0.5244647264480591,
      "mean_token_accuracy": 0.7867362052202225,
      "num_tokens": 10649827.0,
      "step": 651
    },
    {
      "entropy": 0.5192963778972626,
      "epoch": 2.4328358208955225,
      "grad_norm": 0.20977173745632172,
      "learning_rate": 0.0002,
      "loss": 0.5227269530296326,
      "mean_token_accuracy": 0.7878894209861755,
      "num_tokens": 10666321.0,
      "step": 652
    },
    {
      "entropy": 0.5367371439933777,
      "epoch": 2.4365671641791042,
      "grad_norm": 0.15084603428840637,
      "learning_rate": 0.0002,
      "loss": 0.5376932621002197,
      "mean_token_accuracy": 0.7824547588825226,
      "num_tokens": 10682965.0,
      "step": 653
    },
    {
      "entropy": 0.5382617115974426,
      "epoch": 2.4402985074626864,
      "grad_norm": 0.1801973283290863,
      "learning_rate": 0.0002,
      "loss": 0.5432254076004028,
      "mean_token_accuracy": 0.7805831581354141,
      "num_tokens": 10699544.0,
      "step": 654
    },
    {
      "entropy": 0.5095302909612656,
      "epoch": 2.4440298507462686,
      "grad_norm": 0.18439368903636932,
      "learning_rate": 0.0002,
      "loss": 0.5105661749839783,
      "mean_token_accuracy": 0.7942203730344772,
      "num_tokens": 10715786.0,
      "step": 655
    },
    {
      "entropy": 0.5031231418251991,
      "epoch": 2.4477611940298507,
      "grad_norm": 0.2009238451719284,
      "learning_rate": 0.0002,
      "loss": 0.5032652020454407,
      "mean_token_accuracy": 0.796208992600441,
      "num_tokens": 10731876.0,
      "step": 656
    },
    {
      "entropy": 0.5191497802734375,
      "epoch": 2.451492537313433,
      "grad_norm": 0.15563416481018066,
      "learning_rate": 0.0002,
      "loss": 0.5171559453010559,
      "mean_token_accuracy": 0.7906945198774338,
      "num_tokens": 10748315.0,
      "step": 657
    },
    {
      "entropy": 0.534915566444397,
      "epoch": 2.455223880597015,
      "grad_norm": 0.1711549013853073,
      "learning_rate": 0.0002,
      "loss": 0.5389001965522766,
      "mean_token_accuracy": 0.7799905091524124,
      "num_tokens": 10764790.0,
      "step": 658
    },
    {
      "entropy": 0.5258138328790665,
      "epoch": 2.458955223880597,
      "grad_norm": 0.1676245927810669,
      "learning_rate": 0.0002,
      "loss": 0.5202907919883728,
      "mean_token_accuracy": 0.7907141149044037,
      "num_tokens": 10781180.0,
      "step": 659
    },
    {
      "entropy": 0.5184210613369942,
      "epoch": 2.4626865671641793,
      "grad_norm": 0.19272486865520477,
      "learning_rate": 0.0002,
      "loss": 0.5211130380630493,
      "mean_token_accuracy": 0.7910874783992767,
      "num_tokens": 10797415.0,
      "step": 660
    },
    {
      "entropy": 0.5086240172386169,
      "epoch": 2.466417910447761,
      "grad_norm": 0.15660832822322845,
      "learning_rate": 0.0002,
      "loss": 0.5071555376052856,
      "mean_token_accuracy": 0.7956086099147797,
      "num_tokens": 10813817.0,
      "step": 661
    },
    {
      "entropy": 0.5204941183328629,
      "epoch": 2.470149253731343,
      "grad_norm": 0.1729627549648285,
      "learning_rate": 0.0002,
      "loss": 0.5216123461723328,
      "mean_token_accuracy": 0.7889334261417389,
      "num_tokens": 10830306.0,
      "step": 662
    },
    {
      "entropy": 0.5246492028236389,
      "epoch": 2.4738805970149254,
      "grad_norm": 0.16240543127059937,
      "learning_rate": 0.0002,
      "loss": 0.5221917629241943,
      "mean_token_accuracy": 0.7887895107269287,
      "num_tokens": 10846665.0,
      "step": 663
    },
    {
      "entropy": 0.5247633457183838,
      "epoch": 2.4776119402985075,
      "grad_norm": 0.17897000908851624,
      "learning_rate": 0.0002,
      "loss": 0.5230647325515747,
      "mean_token_accuracy": 0.7872501760721207,
      "num_tokens": 10863063.0,
      "step": 664
    },
    {
      "entropy": 0.5195020511746407,
      "epoch": 2.4813432835820897,
      "grad_norm": 0.17342959344387054,
      "learning_rate": 0.0002,
      "loss": 0.5188111066818237,
      "mean_token_accuracy": 0.7901908159255981,
      "num_tokens": 10879309.0,
      "step": 665
    },
    {
      "entropy": 0.5102927386760712,
      "epoch": 2.485074626865672,
      "grad_norm": 0.19114357233047485,
      "learning_rate": 0.0002,
      "loss": 0.5075148940086365,
      "mean_token_accuracy": 0.793827474117279,
      "num_tokens": 10895754.0,
      "step": 666
    },
    {
      "entropy": 0.5079594776034355,
      "epoch": 2.4888059701492535,
      "grad_norm": 0.17865923047065735,
      "learning_rate": 0.0002,
      "loss": 0.5172625184059143,
      "mean_token_accuracy": 0.7915907651185989,
      "num_tokens": 10912075.0,
      "step": 667
    },
    {
      "entropy": 0.5225436985492706,
      "epoch": 2.4925373134328357,
      "grad_norm": 0.16307690739631653,
      "learning_rate": 0.0002,
      "loss": 0.5252382159233093,
      "mean_token_accuracy": 0.7875044196844101,
      "num_tokens": 10928580.0,
      "step": 668
    },
    {
      "entropy": 0.5093511343002319,
      "epoch": 2.496268656716418,
      "grad_norm": 0.1806548833847046,
      "learning_rate": 0.0002,
      "loss": 0.5136329531669617,
      "mean_token_accuracy": 0.7932980954647064,
      "num_tokens": 10944938.0,
      "step": 669
    },
    {
      "entropy": 0.5392017215490341,
      "epoch": 2.5,
      "grad_norm": 0.172809436917305,
      "learning_rate": 0.0002,
      "loss": 0.5332465767860413,
      "mean_token_accuracy": 0.7860011905431747,
      "num_tokens": 10961259.0,
      "step": 670
    },
    {
      "entropy": 0.5107344835996628,
      "epoch": 2.503731343283582,
      "grad_norm": 0.16989955306053162,
      "learning_rate": 0.0002,
      "loss": 0.515048623085022,
      "mean_token_accuracy": 0.7915125340223312,
      "num_tokens": 10977526.0,
      "step": 671
    },
    {
      "entropy": 0.5134129077196121,
      "epoch": 2.5074626865671643,
      "grad_norm": 0.1751825511455536,
      "learning_rate": 0.0002,
      "loss": 0.5138539671897888,
      "mean_token_accuracy": 0.7937446385622025,
      "num_tokens": 10993852.0,
      "step": 672
    },
    {
      "entropy": 0.5191601738333702,
      "epoch": 2.5111940298507465,
      "grad_norm": 0.20491214096546173,
      "learning_rate": 0.0002,
      "loss": 0.5287593603134155,
      "mean_token_accuracy": 0.7851235568523407,
      "num_tokens": 11010234.0,
      "step": 673
    },
    {
      "entropy": 0.5235985666513443,
      "epoch": 2.5149253731343286,
      "grad_norm": 0.18775786459445953,
      "learning_rate": 0.0002,
      "loss": 0.522467315196991,
      "mean_token_accuracy": 0.7902757078409195,
      "num_tokens": 11026702.0,
      "step": 674
    },
    {
      "entropy": 0.5197082608938217,
      "epoch": 2.5186567164179103,
      "grad_norm": 0.1559353768825531,
      "learning_rate": 0.0002,
      "loss": 0.5209258794784546,
      "mean_token_accuracy": 0.7914120852947235,
      "num_tokens": 11042983.0,
      "step": 675
    },
    {
      "entropy": 0.5311167538166046,
      "epoch": 2.5223880597014925,
      "grad_norm": 0.3643738329410553,
      "learning_rate": 0.0002,
      "loss": 0.5349550247192383,
      "mean_token_accuracy": 0.7879150658845901,
      "num_tokens": 11059310.0,
      "step": 676
    },
    {
      "entropy": 0.5285512655973434,
      "epoch": 2.5261194029850746,
      "grad_norm": 0.35899150371551514,
      "learning_rate": 0.0002,
      "loss": 0.5323964953422546,
      "mean_token_accuracy": 0.7864739298820496,
      "num_tokens": 11075633.0,
      "step": 677
    },
    {
      "entropy": 0.5160254240036011,
      "epoch": 2.529850746268657,
      "grad_norm": 0.1786215752363205,
      "learning_rate": 0.0002,
      "loss": 0.5121774077415466,
      "mean_token_accuracy": 0.7930204421281815,
      "num_tokens": 11091937.0,
      "step": 678
    },
    {
      "entropy": 0.5221347957849503,
      "epoch": 2.533582089552239,
      "grad_norm": 0.17423835396766663,
      "learning_rate": 0.0002,
      "loss": 0.521117091178894,
      "mean_token_accuracy": 0.7917880117893219,
      "num_tokens": 11108434.0,
      "step": 679
    },
    {
      "entropy": 0.5129313766956329,
      "epoch": 2.5373134328358207,
      "grad_norm": 0.2339075803756714,
      "learning_rate": 0.0002,
      "loss": 0.5248045325279236,
      "mean_token_accuracy": 0.7886321395635605,
      "num_tokens": 11124838.0,
      "step": 680
    },
    {
      "entropy": 0.5080573558807373,
      "epoch": 2.541044776119403,
      "grad_norm": 0.16493134200572968,
      "learning_rate": 0.0002,
      "loss": 0.5047459602355957,
      "mean_token_accuracy": 0.7968858331441879,
      "num_tokens": 11141150.0,
      "step": 681
    },
    {
      "entropy": 0.5269840210676193,
      "epoch": 2.544776119402985,
      "grad_norm": 0.18450985848903656,
      "learning_rate": 0.0002,
      "loss": 0.519035816192627,
      "mean_token_accuracy": 0.7893852144479752,
      "num_tokens": 11157626.0,
      "step": 682
    },
    {
      "entropy": 0.5483007431030273,
      "epoch": 2.548507462686567,
      "grad_norm": 0.18199008703231812,
      "learning_rate": 0.0002,
      "loss": 0.5438345074653625,
      "mean_token_accuracy": 0.7794067114591599,
      "num_tokens": 11174029.0,
      "step": 683
    },
    {
      "entropy": 0.513416476547718,
      "epoch": 2.5522388059701493,
      "grad_norm": 0.21075013279914856,
      "learning_rate": 0.0002,
      "loss": 0.5135990977287292,
      "mean_token_accuracy": 0.7913079857826233,
      "num_tokens": 11190225.0,
      "step": 684
    },
    {
      "entropy": 0.5416000038385391,
      "epoch": 2.5559701492537314,
      "grad_norm": 0.17459255456924438,
      "learning_rate": 0.0002,
      "loss": 0.5411964654922485,
      "mean_token_accuracy": 0.7781907021999359,
      "num_tokens": 11206277.0,
      "step": 685
    },
    {
      "entropy": 0.5139903053641319,
      "epoch": 2.5597014925373136,
      "grad_norm": 0.2360483855009079,
      "learning_rate": 0.0002,
      "loss": 0.5156392455101013,
      "mean_token_accuracy": 0.7910273224115372,
      "num_tokens": 11222661.0,
      "step": 686
    },
    {
      "entropy": 0.5297715365886688,
      "epoch": 2.5634328358208958,
      "grad_norm": 0.16449929773807526,
      "learning_rate": 0.0002,
      "loss": 0.526314914226532,
      "mean_token_accuracy": 0.7875741422176361,
      "num_tokens": 11239309.0,
      "step": 687
    },
    {
      "entropy": 0.5207169353961945,
      "epoch": 2.5671641791044775,
      "grad_norm": 0.18443575501441956,
      "learning_rate": 0.0002,
      "loss": 0.5246210694313049,
      "mean_token_accuracy": 0.7876296937465668,
      "num_tokens": 11255544.0,
      "step": 688
    },
    {
      "entropy": 0.5143519788980484,
      "epoch": 2.5708955223880596,
      "grad_norm": 0.18113332986831665,
      "learning_rate": 0.0002,
      "loss": 0.5267289876937866,
      "mean_token_accuracy": 0.7861593663692474,
      "num_tokens": 11272128.0,
      "step": 689
    },
    {
      "entropy": 0.5073645934462547,
      "epoch": 2.574626865671642,
      "grad_norm": 0.1688588708639145,
      "learning_rate": 0.0002,
      "loss": 0.5128133893013,
      "mean_token_accuracy": 0.7928483635187149,
      "num_tokens": 11288485.0,
      "step": 690
    },
    {
      "entropy": 0.5352355241775513,
      "epoch": 2.578358208955224,
      "grad_norm": 0.2043209820985794,
      "learning_rate": 0.0002,
      "loss": 0.5441552996635437,
      "mean_token_accuracy": 0.7808532267808914,
      "num_tokens": 11304943.0,
      "step": 691
    },
    {
      "entropy": 0.5129336938261986,
      "epoch": 2.582089552238806,
      "grad_norm": 0.17394010722637177,
      "learning_rate": 0.0002,
      "loss": 0.5214764475822449,
      "mean_token_accuracy": 0.7900305837392807,
      "num_tokens": 11321326.0,
      "step": 692
    },
    {
      "entropy": 0.5380357950925827,
      "epoch": 2.585820895522388,
      "grad_norm": 0.17667418718338013,
      "learning_rate": 0.0002,
      "loss": 0.5346931219100952,
      "mean_token_accuracy": 0.7854123413562775,
      "num_tokens": 11337848.0,
      "step": 693
    },
    {
      "entropy": 0.5276758819818497,
      "epoch": 2.58955223880597,
      "grad_norm": 0.1541767120361328,
      "learning_rate": 0.0002,
      "loss": 0.5159422755241394,
      "mean_token_accuracy": 0.7916016578674316,
      "num_tokens": 11354276.0,
      "step": 694
    },
    {
      "entropy": 0.5228015929460526,
      "epoch": 2.593283582089552,
      "grad_norm": 0.15557631850242615,
      "learning_rate": 0.0002,
      "loss": 0.5210080146789551,
      "mean_token_accuracy": 0.7870455980300903,
      "num_tokens": 11370495.0,
      "step": 695
    },
    {
      "entropy": 0.5402127206325531,
      "epoch": 2.5970149253731343,
      "grad_norm": 0.1675378531217575,
      "learning_rate": 0.0002,
      "loss": 0.5398890376091003,
      "mean_token_accuracy": 0.7822866439819336,
      "num_tokens": 11387311.0,
      "step": 696
    },
    {
      "entropy": 0.5136967226862907,
      "epoch": 2.6007462686567164,
      "grad_norm": 0.19592520594596863,
      "learning_rate": 0.0002,
      "loss": 0.5270940065383911,
      "mean_token_accuracy": 0.7860371470451355,
      "num_tokens": 11403707.0,
      "step": 697
    },
    {
      "entropy": 0.5126481875777245,
      "epoch": 2.6044776119402986,
      "grad_norm": 0.15868966281414032,
      "learning_rate": 0.0002,
      "loss": 0.5106173157691956,
      "mean_token_accuracy": 0.7943423539400101,
      "num_tokens": 11420475.0,
      "step": 698
    },
    {
      "entropy": 0.5322592258453369,
      "epoch": 2.6082089552238807,
      "grad_norm": 0.1745113879442215,
      "learning_rate": 0.0002,
      "loss": 0.5285219550132751,
      "mean_token_accuracy": 0.7842403054237366,
      "num_tokens": 11436972.0,
      "step": 699
    },
    {
      "entropy": 0.5454751402139664,
      "epoch": 2.611940298507463,
      "grad_norm": 0.17534111440181732,
      "learning_rate": 0.0002,
      "loss": 0.5420178174972534,
      "mean_token_accuracy": 0.7788625806570053,
      "num_tokens": 11453287.0,
      "step": 700
    },
    {
      "entropy": 0.5207047313451767,
      "epoch": 2.6156716417910446,
      "grad_norm": 0.19825099408626556,
      "learning_rate": 0.0002,
      "loss": 0.5221109390258789,
      "mean_token_accuracy": 0.7887770980596542,
      "num_tokens": 11469763.0,
      "step": 701
    },
    {
      "entropy": 0.5161542892456055,
      "epoch": 2.6194029850746268,
      "grad_norm": 0.20169204473495483,
      "learning_rate": 0.0002,
      "loss": 0.5191456079483032,
      "mean_token_accuracy": 0.7898091673851013,
      "num_tokens": 11486183.0,
      "step": 702
    },
    {
      "entropy": 0.5139896869659424,
      "epoch": 2.623134328358209,
      "grad_norm": 0.1553078144788742,
      "learning_rate": 0.0002,
      "loss": 0.5166129469871521,
      "mean_token_accuracy": 0.7905794680118561,
      "num_tokens": 11502580.0,
      "step": 703
    },
    {
      "entropy": 0.5328216999769211,
      "epoch": 2.626865671641791,
      "grad_norm": 0.16720637679100037,
      "learning_rate": 0.0002,
      "loss": 0.5284329652786255,
      "mean_token_accuracy": 0.7845450043678284,
      "num_tokens": 11519139.0,
      "step": 704
    },
    {
      "entropy": 0.5221163928508759,
      "epoch": 2.6305970149253732,
      "grad_norm": 0.18718595802783966,
      "learning_rate": 0.0002,
      "loss": 0.5283530354499817,
      "mean_token_accuracy": 0.7861590385437012,
      "num_tokens": 11535331.0,
      "step": 705
    },
    {
      "entropy": 0.5037321597337723,
      "epoch": 2.6343283582089554,
      "grad_norm": 0.18179404735565186,
      "learning_rate": 0.0002,
      "loss": 0.5100149512290955,
      "mean_token_accuracy": 0.7948807179927826,
      "num_tokens": 11551521.0,
      "step": 706
    },
    {
      "entropy": 0.5192279741168022,
      "epoch": 2.638059701492537,
      "grad_norm": 0.20862863957881927,
      "learning_rate": 0.0002,
      "loss": 0.5236419439315796,
      "mean_token_accuracy": 0.78767329454422,
      "num_tokens": 11567724.0,
      "step": 707
    },
    {
      "entropy": 0.5220971703529358,
      "epoch": 2.6417910447761193,
      "grad_norm": 0.1864498108625412,
      "learning_rate": 0.0002,
      "loss": 0.5215906500816345,
      "mean_token_accuracy": 0.7895607203245163,
      "num_tokens": 11584312.0,
      "step": 708
    },
    {
      "entropy": 0.5166686177253723,
      "epoch": 2.6455223880597014,
      "grad_norm": 0.19756394624710083,
      "learning_rate": 0.0002,
      "loss": 0.5216847658157349,
      "mean_token_accuracy": 0.7885107100009918,
      "num_tokens": 11600627.0,
      "step": 709
    },
    {
      "entropy": 0.5282913744449615,
      "epoch": 2.6492537313432836,
      "grad_norm": 0.17049254477024078,
      "learning_rate": 0.0002,
      "loss": 0.5236992835998535,
      "mean_token_accuracy": 0.7891093492507935,
      "num_tokens": 11617078.0,
      "step": 710
    },
    {
      "entropy": 0.535365641117096,
      "epoch": 2.6529850746268657,
      "grad_norm": 0.18149472773075104,
      "learning_rate": 0.0002,
      "loss": 0.5293483734130859,
      "mean_token_accuracy": 0.7839226573705673,
      "num_tokens": 11633150.0,
      "step": 711
    },
    {
      "entropy": 0.5354510843753815,
      "epoch": 2.656716417910448,
      "grad_norm": 0.14960937201976776,
      "learning_rate": 0.0002,
      "loss": 0.5259742736816406,
      "mean_token_accuracy": 0.7883720546960831,
      "num_tokens": 11649869.0,
      "step": 712
    },
    {
      "entropy": 0.5009367614984512,
      "epoch": 2.66044776119403,
      "grad_norm": 0.1922633796930313,
      "learning_rate": 0.0002,
      "loss": 0.5144891142845154,
      "mean_token_accuracy": 0.7931148111820221,
      "num_tokens": 11666213.0,
      "step": 713
    },
    {
      "entropy": 0.5240390002727509,
      "epoch": 2.664179104477612,
      "grad_norm": 0.18613417446613312,
      "learning_rate": 0.0002,
      "loss": 0.5301287174224854,
      "mean_token_accuracy": 0.7852264195680618,
      "num_tokens": 11682401.0,
      "step": 714
    },
    {
      "entropy": 0.5058765560388565,
      "epoch": 2.667910447761194,
      "grad_norm": 0.18312235176563263,
      "learning_rate": 0.0002,
      "loss": 0.5103883743286133,
      "mean_token_accuracy": 0.7938342690467834,
      "num_tokens": 11698526.0,
      "step": 715
    },
    {
      "entropy": 0.5158815458416939,
      "epoch": 2.671641791044776,
      "grad_norm": 0.17897051572799683,
      "learning_rate": 0.0002,
      "loss": 0.5187905430793762,
      "mean_token_accuracy": 0.7909149527549744,
      "num_tokens": 11714969.0,
      "step": 716
    },
    {
      "entropy": 0.5180910006165504,
      "epoch": 2.675373134328358,
      "grad_norm": 0.17239928245544434,
      "learning_rate": 0.0002,
      "loss": 0.5176571011543274,
      "mean_token_accuracy": 0.7897525131702423,
      "num_tokens": 11731465.0,
      "step": 717
    },
    {
      "entropy": 0.5389165133237839,
      "epoch": 2.6791044776119404,
      "grad_norm": 0.187730610370636,
      "learning_rate": 0.0002,
      "loss": 0.5379902720451355,
      "mean_token_accuracy": 0.7827825844287872,
      "num_tokens": 11747857.0,
      "step": 718
    },
    {
      "entropy": 0.5241931825876236,
      "epoch": 2.6828358208955225,
      "grad_norm": 0.17442777752876282,
      "learning_rate": 0.0002,
      "loss": 0.5209442973136902,
      "mean_token_accuracy": 0.7903313934803009,
      "num_tokens": 11764348.0,
      "step": 719
    },
    {
      "entropy": 0.5140400677919388,
      "epoch": 2.6865671641791042,
      "grad_norm": 0.17288851737976074,
      "learning_rate": 0.0002,
      "loss": 0.5115959048271179,
      "mean_token_accuracy": 0.7920527160167694,
      "num_tokens": 11780822.0,
      "step": 720
    },
    {
      "entropy": 0.5226388201117516,
      "epoch": 2.6902985074626864,
      "grad_norm": 0.20361240208148956,
      "learning_rate": 0.0002,
      "loss": 0.5352538824081421,
      "mean_token_accuracy": 0.7822130769491196,
      "num_tokens": 11797173.0,
      "step": 721
    },
    {
      "entropy": 0.5000582486391068,
      "epoch": 2.6940298507462686,
      "grad_norm": 0.15772968530654907,
      "learning_rate": 0.0002,
      "loss": 0.5034382939338684,
      "mean_token_accuracy": 0.796001210808754,
      "num_tokens": 11813674.0,
      "step": 722
    },
    {
      "entropy": 0.5309499502182007,
      "epoch": 2.6977611940298507,
      "grad_norm": 0.18732546269893646,
      "learning_rate": 0.0002,
      "loss": 0.5299732685089111,
      "mean_token_accuracy": 0.7848151177167892,
      "num_tokens": 11830402.0,
      "step": 723
    },
    {
      "entropy": 0.5357850790023804,
      "epoch": 2.701492537313433,
      "grad_norm": 0.18165066838264465,
      "learning_rate": 0.0002,
      "loss": 0.5308882594108582,
      "mean_token_accuracy": 0.7862505465745926,
      "num_tokens": 11846707.0,
      "step": 724
    },
    {
      "entropy": 0.5284833014011383,
      "epoch": 2.705223880597015,
      "grad_norm": 0.172774076461792,
      "learning_rate": 0.0002,
      "loss": 0.5216760635375977,
      "mean_token_accuracy": 0.7896177619695663,
      "num_tokens": 11863081.0,
      "step": 725
    },
    {
      "entropy": 0.5200880020856857,
      "epoch": 2.708955223880597,
      "grad_norm": 0.2079714685678482,
      "learning_rate": 0.0002,
      "loss": 0.5182638168334961,
      "mean_token_accuracy": 0.7900623828172684,
      "num_tokens": 11879482.0,
      "step": 726
    },
    {
      "entropy": 0.5209276676177979,
      "epoch": 2.7126865671641793,
      "grad_norm": 0.19981354475021362,
      "learning_rate": 0.0002,
      "loss": 0.533184289932251,
      "mean_token_accuracy": 0.7846023589372635,
      "num_tokens": 11895845.0,
      "step": 727
    },
    {
      "entropy": 0.5259044617414474,
      "epoch": 2.716417910447761,
      "grad_norm": 0.25463277101516724,
      "learning_rate": 0.0002,
      "loss": 0.5362758636474609,
      "mean_token_accuracy": 0.7840900421142578,
      "num_tokens": 11912248.0,
      "step": 728
    },
    {
      "entropy": 0.5323220193386078,
      "epoch": 2.720149253731343,
      "grad_norm": 0.1765269637107849,
      "learning_rate": 0.0002,
      "loss": 0.5269978046417236,
      "mean_token_accuracy": 0.7877814024686813,
      "num_tokens": 11928507.0,
      "step": 729
    },
    {
      "entropy": 0.5406877994537354,
      "epoch": 2.7238805970149254,
      "grad_norm": 0.19709576666355133,
      "learning_rate": 0.0002,
      "loss": 0.5350936651229858,
      "mean_token_accuracy": 0.7793232202529907,
      "num_tokens": 11944765.0,
      "step": 730
    },
    {
      "entropy": 0.5157359838485718,
      "epoch": 2.7276119402985075,
      "grad_norm": 0.17762304842472076,
      "learning_rate": 0.0002,
      "loss": 0.5103439092636108,
      "mean_token_accuracy": 0.7935812622308731,
      "num_tokens": 11961061.0,
      "step": 731
    },
    {
      "entropy": 0.5088659226894379,
      "epoch": 2.7313432835820897,
      "grad_norm": 0.2038097232580185,
      "learning_rate": 0.0002,
      "loss": 0.5155967473983765,
      "mean_token_accuracy": 0.7911150306463242,
      "num_tokens": 11977255.0,
      "step": 732
    },
    {
      "entropy": 0.5112843066453934,
      "epoch": 2.7350746268656714,
      "grad_norm": 0.19635888934135437,
      "learning_rate": 0.0002,
      "loss": 0.5193964242935181,
      "mean_token_accuracy": 0.790523573756218,
      "num_tokens": 11993549.0,
      "step": 733
    },
    {
      "entropy": 0.5119920894503593,
      "epoch": 2.7388059701492535,
      "grad_norm": 0.17737363278865814,
      "learning_rate": 0.0002,
      "loss": 0.5203633308410645,
      "mean_token_accuracy": 0.7897647768259048,
      "num_tokens": 12009626.0,
      "step": 734
    },
    {
      "entropy": 0.5212873220443726,
      "epoch": 2.7425373134328357,
      "grad_norm": 0.17992158234119415,
      "learning_rate": 0.0002,
      "loss": 0.5268858671188354,
      "mean_token_accuracy": 0.7881815582513809,
      "num_tokens": 12026060.0,
      "step": 735
    },
    {
      "entropy": 0.5266913175582886,
      "epoch": 2.746268656716418,
      "grad_norm": 0.1624017059803009,
      "learning_rate": 0.0002,
      "loss": 0.525255560874939,
      "mean_token_accuracy": 0.7902468144893646,
      "num_tokens": 12042407.0,
      "step": 736
    },
    {
      "entropy": 0.5043733343482018,
      "epoch": 2.75,
      "grad_norm": 0.17971543967723846,
      "learning_rate": 0.0002,
      "loss": 0.5028079152107239,
      "mean_token_accuracy": 0.7998141497373581,
      "num_tokens": 12058820.0,
      "step": 737
    },
    {
      "entropy": 0.5165901780128479,
      "epoch": 2.753731343283582,
      "grad_norm": 0.18306542932987213,
      "learning_rate": 0.0002,
      "loss": 0.514511227607727,
      "mean_token_accuracy": 0.7911152690649033,
      "num_tokens": 12075073.0,
      "step": 738
    },
    {
      "entropy": 0.521696463227272,
      "epoch": 2.7574626865671643,
      "grad_norm": 0.2015487104654312,
      "learning_rate": 0.0002,
      "loss": 0.5179941058158875,
      "mean_token_accuracy": 0.789873868227005,
      "num_tokens": 12091362.0,
      "step": 739
    },
    {
      "entropy": 0.52324378490448,
      "epoch": 2.7611940298507465,
      "grad_norm": 0.1782568246126175,
      "learning_rate": 0.0002,
      "loss": 0.5252962112426758,
      "mean_token_accuracy": 0.7859823405742645,
      "num_tokens": 12107787.0,
      "step": 740
    },
    {
      "entropy": 0.5147482007741928,
      "epoch": 2.7649253731343286,
      "grad_norm": 0.19460241496562958,
      "learning_rate": 0.0002,
      "loss": 0.5198803544044495,
      "mean_token_accuracy": 0.7908852845430374,
      "num_tokens": 12123988.0,
      "step": 741
    },
    {
      "entropy": 0.532531350851059,
      "epoch": 2.7686567164179103,
      "grad_norm": 0.21107251942157745,
      "learning_rate": 0.0002,
      "loss": 0.5386015176773071,
      "mean_token_accuracy": 0.7829076945781708,
      "num_tokens": 12140336.0,
      "step": 742
    },
    {
      "entropy": 0.5097082331776619,
      "epoch": 2.7723880597014925,
      "grad_norm": 0.18247488141059875,
      "learning_rate": 0.0002,
      "loss": 0.5075528025627136,
      "mean_token_accuracy": 0.7967899888753891,
      "num_tokens": 12156517.0,
      "step": 743
    },
    {
      "entropy": 0.5386329740285873,
      "epoch": 2.7761194029850746,
      "grad_norm": 0.1703251600265503,
      "learning_rate": 0.0002,
      "loss": 0.5331581830978394,
      "mean_token_accuracy": 0.7818602025508881,
      "num_tokens": 12172890.0,
      "step": 744
    },
    {
      "entropy": 0.5344566106796265,
      "epoch": 2.779850746268657,
      "grad_norm": 0.1912059634923935,
      "learning_rate": 0.0002,
      "loss": 0.5289663076400757,
      "mean_token_accuracy": 0.7873619496822357,
      "num_tokens": 12189337.0,
      "step": 745
    },
    {
      "entropy": 0.519013062119484,
      "epoch": 2.783582089552239,
      "grad_norm": 0.1737934947013855,
      "learning_rate": 0.0002,
      "loss": 0.5253188014030457,
      "mean_token_accuracy": 0.7876646816730499,
      "num_tokens": 12205610.0,
      "step": 746
    },
    {
      "entropy": 0.5199131593108177,
      "epoch": 2.7873134328358207,
      "grad_norm": 0.20281687378883362,
      "learning_rate": 0.0002,
      "loss": 0.5254766941070557,
      "mean_token_accuracy": 0.785002738237381,
      "num_tokens": 12221943.0,
      "step": 747
    },
    {
      "entropy": 0.5261564403772354,
      "epoch": 2.791044776119403,
      "grad_norm": 0.17186403274536133,
      "learning_rate": 0.0002,
      "loss": 0.52431321144104,
      "mean_token_accuracy": 0.7875388860702515,
      "num_tokens": 12238419.0,
      "step": 748
    },
    {
      "entropy": 0.5311170220375061,
      "epoch": 2.794776119402985,
      "grad_norm": 0.2272450029850006,
      "learning_rate": 0.0002,
      "loss": 0.5422958135604858,
      "mean_token_accuracy": 0.7811820060014725,
      "num_tokens": 12254810.0,
      "step": 749
    },
    {
      "entropy": 0.5110429674386978,
      "epoch": 2.798507462686567,
      "grad_norm": 0.1752447783946991,
      "learning_rate": 0.0002,
      "loss": 0.511790931224823,
      "mean_token_accuracy": 0.7926972359418869,
      "num_tokens": 12271003.0,
      "step": 750
    },
    {
      "entropy": 0.5179876983165741,
      "epoch": 2.8022388059701493,
      "grad_norm": 0.17473958432674408,
      "learning_rate": 0.0002,
      "loss": 0.5154609084129333,
      "mean_token_accuracy": 0.7912380993366241,
      "num_tokens": 12287481.0,
      "step": 751
    },
    {
      "entropy": 0.5093216001987457,
      "epoch": 2.8059701492537314,
      "grad_norm": 0.18723160028457642,
      "learning_rate": 0.0002,
      "loss": 0.5158673524856567,
      "mean_token_accuracy": 0.7871368080377579,
      "num_tokens": 12303930.0,
      "step": 752
    },
    {
      "entropy": 0.508473701775074,
      "epoch": 2.8097014925373136,
      "grad_norm": 0.16832517087459564,
      "learning_rate": 0.0002,
      "loss": 0.5054484605789185,
      "mean_token_accuracy": 0.7960272431373596,
      "num_tokens": 12320104.0,
      "step": 753
    },
    {
      "entropy": 0.5237400829792023,
      "epoch": 2.8134328358208958,
      "grad_norm": 0.16328011453151703,
      "learning_rate": 0.0002,
      "loss": 0.521124541759491,
      "mean_token_accuracy": 0.7856553643941879,
      "num_tokens": 12336433.0,
      "step": 754
    },
    {
      "entropy": 0.5281547904014587,
      "epoch": 2.8171641791044775,
      "grad_norm": 0.1548423022031784,
      "learning_rate": 0.0002,
      "loss": 0.5290802121162415,
      "mean_token_accuracy": 0.786285325884819,
      "num_tokens": 12353044.0,
      "step": 755
    },
    {
      "entropy": 0.5319864898920059,
      "epoch": 2.8208955223880596,
      "grad_norm": 0.17879781126976013,
      "learning_rate": 0.0002,
      "loss": 0.5398349761962891,
      "mean_token_accuracy": 0.7838435024023056,
      "num_tokens": 12369412.0,
      "step": 756
    },
    {
      "entropy": 0.5187028869986534,
      "epoch": 2.824626865671642,
      "grad_norm": 0.1700705736875534,
      "learning_rate": 0.0002,
      "loss": 0.5216772556304932,
      "mean_token_accuracy": 0.7881180942058563,
      "num_tokens": 12385842.0,
      "step": 757
    },
    {
      "entropy": 0.5375183820724487,
      "epoch": 2.828358208955224,
      "grad_norm": 0.18571849167346954,
      "learning_rate": 0.0002,
      "loss": 0.5363373160362244,
      "mean_token_accuracy": 0.7835224717855453,
      "num_tokens": 12401843.0,
      "step": 758
    },
    {
      "entropy": 0.5308329612016678,
      "epoch": 2.832089552238806,
      "grad_norm": 0.17223785817623138,
      "learning_rate": 0.0002,
      "loss": 0.526056170463562,
      "mean_token_accuracy": 0.7878949195146561,
      "num_tokens": 12418294.0,
      "step": 759
    },
    {
      "entropy": 0.5180955529212952,
      "epoch": 2.835820895522388,
      "grad_norm": 0.19426262378692627,
      "learning_rate": 0.0002,
      "loss": 0.5233747363090515,
      "mean_token_accuracy": 0.7847504019737244,
      "num_tokens": 12434639.0,
      "step": 760
    },
    {
      "entropy": 0.5211240500211716,
      "epoch": 2.83955223880597,
      "grad_norm": 0.17702415585517883,
      "learning_rate": 0.0002,
      "loss": 0.5200589299201965,
      "mean_token_accuracy": 0.7906626909971237,
      "num_tokens": 12450966.0,
      "step": 761
    },
    {
      "entropy": 0.5131405591964722,
      "epoch": 2.843283582089552,
      "grad_norm": 0.16737323999404907,
      "learning_rate": 0.0002,
      "loss": 0.5157369375228882,
      "mean_token_accuracy": 0.7910412847995758,
      "num_tokens": 12467267.0,
      "step": 762
    },
    {
      "entropy": 0.5366750806570053,
      "epoch": 2.8470149253731343,
      "grad_norm": 0.16437119245529175,
      "learning_rate": 0.0002,
      "loss": 0.5366208553314209,
      "mean_token_accuracy": 0.7785362154245377,
      "num_tokens": 12483825.0,
      "step": 763
    },
    {
      "entropy": 0.5352065116167068,
      "epoch": 2.8507462686567164,
      "grad_norm": 0.14709708094596863,
      "learning_rate": 0.0002,
      "loss": 0.5321803092956543,
      "mean_token_accuracy": 0.7876356840133667,
      "num_tokens": 12500282.0,
      "step": 764
    },
    {
      "entropy": 0.5344839543104172,
      "epoch": 2.8544776119402986,
      "grad_norm": 0.14966800808906555,
      "learning_rate": 0.0002,
      "loss": 0.531541109085083,
      "mean_token_accuracy": 0.7864569425582886,
      "num_tokens": 12516745.0,
      "step": 765
    },
    {
      "entropy": 0.5019206777215004,
      "epoch": 2.8582089552238807,
      "grad_norm": 0.18061211705207825,
      "learning_rate": 0.0002,
      "loss": 0.5042290687561035,
      "mean_token_accuracy": 0.7964124828577042,
      "num_tokens": 12532956.0,
      "step": 766
    },
    {
      "entropy": 0.5074614435434341,
      "epoch": 2.861940298507463,
      "grad_norm": 0.18505603075027466,
      "learning_rate": 0.0002,
      "loss": 0.5167686939239502,
      "mean_token_accuracy": 0.7910870015621185,
      "num_tokens": 12549228.0,
      "step": 767
    },
    {
      "entropy": 0.5183932334184647,
      "epoch": 2.8656716417910446,
      "grad_norm": 0.1724204570055008,
      "learning_rate": 0.0002,
      "loss": 0.524673581123352,
      "mean_token_accuracy": 0.7895944714546204,
      "num_tokens": 12565584.0,
      "step": 768
    },
    {
      "entropy": 0.5218961760401726,
      "epoch": 2.8694029850746268,
      "grad_norm": 0.15690268576145172,
      "learning_rate": 0.0002,
      "loss": 0.5237387418746948,
      "mean_token_accuracy": 0.7879291921854019,
      "num_tokens": 12582028.0,
      "step": 769
    },
    {
      "entropy": 0.522913858294487,
      "epoch": 2.873134328358209,
      "grad_norm": 0.21558451652526855,
      "learning_rate": 0.0002,
      "loss": 0.5252016186714172,
      "mean_token_accuracy": 0.7892911732196808,
      "num_tokens": 12598433.0,
      "step": 770
    },
    {
      "entropy": 0.5268243104219437,
      "epoch": 2.876865671641791,
      "grad_norm": 0.14720359444618225,
      "learning_rate": 0.0002,
      "loss": 0.5126805901527405,
      "mean_token_accuracy": 0.7932652682065964,
      "num_tokens": 12615016.0,
      "step": 771
    },
    {
      "entropy": 0.5350854992866516,
      "epoch": 2.8805970149253732,
      "grad_norm": 0.2269367277622223,
      "learning_rate": 0.0002,
      "loss": 0.5303336977958679,
      "mean_token_accuracy": 0.7860198318958282,
      "num_tokens": 12631417.0,
      "step": 772
    },
    {
      "entropy": 0.5364827066659927,
      "epoch": 2.8843283582089554,
      "grad_norm": 0.174966961145401,
      "learning_rate": 0.0002,
      "loss": 0.539239227771759,
      "mean_token_accuracy": 0.7799928337335587,
      "num_tokens": 12647945.0,
      "step": 773
    },
    {
      "entropy": 0.5175943821668625,
      "epoch": 2.888059701492537,
      "grad_norm": 0.239835724234581,
      "learning_rate": 0.0002,
      "loss": 0.5245868563652039,
      "mean_token_accuracy": 0.7879510670900345,
      "num_tokens": 12664357.0,
      "step": 774
    },
    {
      "entropy": 0.525666281580925,
      "epoch": 2.8917910447761193,
      "grad_norm": 0.19655871391296387,
      "learning_rate": 0.0002,
      "loss": 0.5303555727005005,
      "mean_token_accuracy": 0.7861410826444626,
      "num_tokens": 12680982.0,
      "step": 775
    },
    {
      "entropy": 0.5001505762338638,
      "epoch": 2.8955223880597014,
      "grad_norm": 0.2157313972711563,
      "learning_rate": 0.0002,
      "loss": 0.5040432810783386,
      "mean_token_accuracy": 0.7954959124326706,
      "num_tokens": 12697247.0,
      "step": 776
    },
    {
      "entropy": 0.5350449979305267,
      "epoch": 2.8992537313432836,
      "grad_norm": 0.20097671449184418,
      "learning_rate": 0.0002,
      "loss": 0.5355807542800903,
      "mean_token_accuracy": 0.7829754054546356,
      "num_tokens": 12713704.0,
      "step": 777
    },
    {
      "entropy": 0.5254835188388824,
      "epoch": 2.9029850746268657,
      "grad_norm": 0.16286872327327728,
      "learning_rate": 0.0002,
      "loss": 0.5268586874008179,
      "mean_token_accuracy": 0.7856743782758713,
      "num_tokens": 12730121.0,
      "step": 778
    },
    {
      "entropy": 0.5580068975687027,
      "epoch": 2.906716417910448,
      "grad_norm": 0.1875162571668625,
      "learning_rate": 0.0002,
      "loss": 0.5552141666412354,
      "mean_token_accuracy": 0.7768043726682663,
      "num_tokens": 12746743.0,
      "step": 779
    },
    {
      "entropy": 0.5280749648809433,
      "epoch": 2.91044776119403,
      "grad_norm": 0.1497505009174347,
      "learning_rate": 0.0002,
      "loss": 0.5217434167861938,
      "mean_token_accuracy": 0.7901686578989029,
      "num_tokens": 12763132.0,
      "step": 780
    },
    {
      "entropy": 0.5150540545582771,
      "epoch": 2.914179104477612,
      "grad_norm": 0.1816144436597824,
      "learning_rate": 0.0002,
      "loss": 0.518502414226532,
      "mean_token_accuracy": 0.787481039762497,
      "num_tokens": 12779424.0,
      "step": 781
    },
    {
      "entropy": 0.5084019601345062,
      "epoch": 2.917910447761194,
      "grad_norm": 0.18711701035499573,
      "learning_rate": 0.0002,
      "loss": 0.513791024684906,
      "mean_token_accuracy": 0.792212188243866,
      "num_tokens": 12795613.0,
      "step": 782
    },
    {
      "entropy": 0.5053626373410225,
      "epoch": 2.921641791044776,
      "grad_norm": 0.18591050803661346,
      "learning_rate": 0.0002,
      "loss": 0.5184471607208252,
      "mean_token_accuracy": 0.7897629290819168,
      "num_tokens": 12812042.0,
      "step": 783
    },
    {
      "entropy": 0.5152643695473671,
      "epoch": 2.925373134328358,
      "grad_norm": 0.1889844536781311,
      "learning_rate": 0.0002,
      "loss": 0.5239505171775818,
      "mean_token_accuracy": 0.7899060547351837,
      "num_tokens": 12828347.0,
      "step": 784
    },
    {
      "entropy": 0.5273154824972153,
      "epoch": 2.9291044776119404,
      "grad_norm": 0.1580808460712433,
      "learning_rate": 0.0002,
      "loss": 0.5230674147605896,
      "mean_token_accuracy": 0.7892739921808243,
      "num_tokens": 12844649.0,
      "step": 785
    },
    {
      "entropy": 0.5427146404981613,
      "epoch": 2.9328358208955225,
      "grad_norm": 0.1609242558479309,
      "learning_rate": 0.0002,
      "loss": 0.5328511595726013,
      "mean_token_accuracy": 0.7870279252529144,
      "num_tokens": 12860925.0,
      "step": 786
    },
    {
      "entropy": 0.5395708978176117,
      "epoch": 2.9365671641791042,
      "grad_norm": 0.16566500067710876,
      "learning_rate": 0.0002,
      "loss": 0.5288230180740356,
      "mean_token_accuracy": 0.7857491821050644,
      "num_tokens": 12877421.0,
      "step": 787
    },
    {
      "entropy": 0.5150002017617226,
      "epoch": 2.9402985074626864,
      "grad_norm": 0.15979735553264618,
      "learning_rate": 0.0002,
      "loss": 0.5151138305664062,
      "mean_token_accuracy": 0.7929274588823318,
      "num_tokens": 12893884.0,
      "step": 788
    },
    {
      "entropy": 0.5136501267552376,
      "epoch": 2.9440298507462686,
      "grad_norm": 0.22995848953723907,
      "learning_rate": 0.0002,
      "loss": 0.5289221405982971,
      "mean_token_accuracy": 0.7861058861017227,
      "num_tokens": 12910392.0,
      "step": 789
    },
    {
      "entropy": 0.5414163321256638,
      "epoch": 2.9477611940298507,
      "grad_norm": 0.18121406435966492,
      "learning_rate": 0.0002,
      "loss": 0.5471609830856323,
      "mean_token_accuracy": 0.7791576832532883,
      "num_tokens": 12927008.0,
      "step": 790
    },
    {
      "entropy": 0.5124125629663467,
      "epoch": 2.951492537313433,
      "grad_norm": 0.20011217892169952,
      "learning_rate": 0.0002,
      "loss": 0.5082826614379883,
      "mean_token_accuracy": 0.7953236252069473,
      "num_tokens": 12943455.0,
      "step": 791
    },
    {
      "entropy": 0.5297976434230804,
      "epoch": 2.955223880597015,
      "grad_norm": 0.15697196125984192,
      "learning_rate": 0.0002,
      "loss": 0.52349853515625,
      "mean_token_accuracy": 0.7878834158182144,
      "num_tokens": 12959738.0,
      "step": 792
    },
    {
      "entropy": 0.5208505988121033,
      "epoch": 2.958955223880597,
      "grad_norm": 0.1743849664926529,
      "learning_rate": 0.0002,
      "loss": 0.5164416432380676,
      "mean_token_accuracy": 0.7905778139829636,
      "num_tokens": 12975959.0,
      "step": 793
    },
    {
      "entropy": 0.5202152505517006,
      "epoch": 2.9626865671641793,
      "grad_norm": 0.2171812653541565,
      "learning_rate": 0.0002,
      "loss": 0.5232265591621399,
      "mean_token_accuracy": 0.7891267985105515,
      "num_tokens": 12992210.0,
      "step": 794
    },
    {
      "entropy": 0.5199966579675674,
      "epoch": 2.966417910447761,
      "grad_norm": 0.22469930350780487,
      "learning_rate": 0.0002,
      "loss": 0.5291239619255066,
      "mean_token_accuracy": 0.7850851565599442,
      "num_tokens": 13008651.0,
      "step": 795
    },
    {
      "entropy": 0.5255369544029236,
      "epoch": 2.970149253731343,
      "grad_norm": 0.2371702492237091,
      "learning_rate": 0.0002,
      "loss": 0.5344793796539307,
      "mean_token_accuracy": 0.7816057652235031,
      "num_tokens": 13024835.0,
      "step": 796
    },
    {
      "entropy": 0.533539667725563,
      "epoch": 2.9738805970149254,
      "grad_norm": 0.16759631037712097,
      "learning_rate": 0.0002,
      "loss": 0.5290622711181641,
      "mean_token_accuracy": 0.7863010764122009,
      "num_tokens": 13041317.0,
      "step": 797
    },
    {
      "entropy": 0.5440120846033096,
      "epoch": 2.9776119402985075,
      "grad_norm": 0.18567466735839844,
      "learning_rate": 0.0002,
      "loss": 0.5380448698997498,
      "mean_token_accuracy": 0.7847449481487274,
      "num_tokens": 13057786.0,
      "step": 798
    },
    {
      "entropy": 0.5316625982522964,
      "epoch": 2.9813432835820897,
      "grad_norm": 0.16756805777549744,
      "learning_rate": 0.0002,
      "loss": 0.530038058757782,
      "mean_token_accuracy": 0.7850957661867142,
      "num_tokens": 13074101.0,
      "step": 799
    },
    {
      "entropy": 0.5209571197628975,
      "epoch": 2.9850746268656714,
      "grad_norm": 0.15372832119464874,
      "learning_rate": 0.0002,
      "loss": 0.5174048542976379,
      "mean_token_accuracy": 0.7904630899429321,
      "num_tokens": 13090657.0,
      "step": 800
    },
    {
      "entropy": 0.4961584433913231,
      "epoch": 2.9888059701492535,
      "grad_norm": 0.20615528523921967,
      "learning_rate": 0.0002,
      "loss": 0.5061897039413452,
      "mean_token_accuracy": 0.7948594838380814,
      "num_tokens": 13106809.0,
      "step": 801
    },
    {
      "entropy": 0.5250304043292999,
      "epoch": 2.9925373134328357,
      "grad_norm": 0.19997206330299377,
      "learning_rate": 0.0002,
      "loss": 0.5354763269424438,
      "mean_token_accuracy": 0.7813558727502823,
      "num_tokens": 13123157.0,
      "step": 802
    },
    {
      "entropy": 0.5133228674530983,
      "epoch": 2.996268656716418,
      "grad_norm": 0.1635276973247528,
      "learning_rate": 0.0002,
      "loss": 0.5166719555854797,
      "mean_token_accuracy": 0.7908363342285156,
      "num_tokens": 13139302.0,
      "step": 803
    },
    {
      "entropy": 0.5134190171957016,
      "epoch": 3.0,
      "grad_norm": 0.15589290857315063,
      "learning_rate": 0.0002,
      "loss": 0.5092208981513977,
      "mean_token_accuracy": 0.7965500056743622,
      "num_tokens": 13155671.0,
      "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": 1.2262397699277455e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}