{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2191,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.098961293697357,
      "epoch": 0.032,
      "grad_norm": 1.891703486442566,
      "learning_rate": 1.9148936170212767e-06,
      "loss": 2.0828,
      "mean_token_accuracy": 0.530680388212204,
      "num_tokens": 72723.0,
      "step": 10
    },
    {
      "entropy": 2.119775766134262,
      "epoch": 0.064,
      "grad_norm": 1.2044862508773804,
      "learning_rate": 4.042553191489362e-06,
      "loss": 2.0093,
      "mean_token_accuracy": 0.5355814293026924,
      "num_tokens": 146392.0,
      "step": 20
    },
    {
      "entropy": 2.220579963922501,
      "epoch": 0.096,
      "grad_norm": 0.9982365369796753,
      "learning_rate": 6.170212765957447e-06,
      "loss": 1.8939,
      "mean_token_accuracy": 0.5451944440603256,
      "num_tokens": 223711.0,
      "step": 30
    },
    {
      "entropy": 2.382017892599106,
      "epoch": 0.128,
      "grad_norm": 0.7386544346809387,
      "learning_rate": 8.297872340425532e-06,
      "loss": 1.9066,
      "mean_token_accuracy": 0.5411656655371189,
      "num_tokens": 300889.0,
      "step": 40
    },
    {
      "entropy": 2.274736815690994,
      "epoch": 0.16,
      "grad_norm": 0.6412256956100464,
      "learning_rate": 1.0425531914893619e-05,
      "loss": 1.7387,
      "mean_token_accuracy": 0.5679451540112496,
      "num_tokens": 377362.0,
      "step": 50
    },
    {
      "entropy": 2.3663365960121157,
      "epoch": 0.192,
      "grad_norm": 0.6228290796279907,
      "learning_rate": 1.2553191489361702e-05,
      "loss": 1.7492,
      "mean_token_accuracy": 0.5746532663702965,
      "num_tokens": 449594.0,
      "step": 60
    },
    {
      "entropy": 2.315044218301773,
      "epoch": 0.224,
      "grad_norm": 0.6034156680107117,
      "learning_rate": 1.4680851063829789e-05,
      "loss": 1.7111,
      "mean_token_accuracy": 0.5675176709890366,
      "num_tokens": 523439.0,
      "step": 70
    },
    {
      "entropy": 2.288265961408615,
      "epoch": 0.256,
      "grad_norm": 0.45914268493652344,
      "learning_rate": 1.6808510638297873e-05,
      "loss": 1.6931,
      "mean_token_accuracy": 0.5713589735329151,
      "num_tokens": 599650.0,
      "step": 80
    },
    {
      "entropy": 2.2693382859230042,
      "epoch": 0.288,
      "grad_norm": 0.6197793483734131,
      "learning_rate": 1.893617021276596e-05,
      "loss": 1.6542,
      "mean_token_accuracy": 0.578165066242218,
      "num_tokens": 675377.0,
      "step": 90
    },
    {
      "entropy": 2.293796479701996,
      "epoch": 0.32,
      "grad_norm": 0.5502006411552429,
      "learning_rate": 1.9999866154043656e-05,
      "loss": 1.7108,
      "mean_token_accuracy": 0.5681634023785591,
      "num_tokens": 751838.0,
      "step": 100
    },
    {
      "entropy": 2.2658903509378434,
      "epoch": 0.352,
      "grad_norm": 0.5713317394256592,
      "learning_rate": 1.9998795407890486e-05,
      "loss": 1.6168,
      "mean_token_accuracy": 0.5843982398509979,
      "num_tokens": 825539.0,
      "step": 110
    },
    {
      "entropy": 2.270280033349991,
      "epoch": 0.384,
      "grad_norm": 0.5967482924461365,
      "learning_rate": 1.999665403023542e-05,
      "loss": 1.6194,
      "mean_token_accuracy": 0.5839526921510696,
      "num_tokens": 897258.0,
      "step": 120
    },
    {
      "entropy": 2.2349284648895265,
      "epoch": 0.416,
      "grad_norm": 0.4899630844593048,
      "learning_rate": 1.9993442250368708e-05,
      "loss": 1.6313,
      "mean_token_accuracy": 0.5815729826688767,
      "num_tokens": 973142.0,
      "step": 130
    },
    {
      "entropy": 2.245553806424141,
      "epoch": 0.448,
      "grad_norm": 0.6546034812927246,
      "learning_rate": 1.9989160412195047e-05,
      "loss": 1.6395,
      "mean_token_accuracy": 0.5780692532658577,
      "num_tokens": 1046762.0,
      "step": 140
    },
    {
      "entropy": 2.288555932044983,
      "epoch": 0.48,
      "grad_norm": 0.5528404116630554,
      "learning_rate": 1.9983808974196752e-05,
      "loss": 1.7118,
      "mean_token_accuracy": 0.5686657652258873,
      "num_tokens": 1125167.0,
      "step": 150
    },
    {
      "entropy": 2.232080355286598,
      "epoch": 0.512,
      "grad_norm": 0.5887461304664612,
      "learning_rate": 1.9977388509384656e-05,
      "loss": 1.6339,
      "mean_token_accuracy": 0.5838325396180153,
      "num_tokens": 1199589.0,
      "step": 160
    },
    {
      "entropy": 2.2232475757598875,
      "epoch": 0.544,
      "grad_norm": 0.5764511823654175,
      "learning_rate": 1.9969899705236763e-05,
      "loss": 1.6173,
      "mean_token_accuracy": 0.5848860442638397,
      "num_tokens": 1276431.0,
      "step": 170
    },
    {
      "entropy": 2.244092071056366,
      "epoch": 0.576,
      "grad_norm": 0.6295827627182007,
      "learning_rate": 1.9961343363624626e-05,
      "loss": 1.6017,
      "mean_token_accuracy": 0.5818701103329659,
      "num_tokens": 1350012.0,
      "step": 180
    },
    {
      "entropy": 2.237305074930191,
      "epoch": 0.608,
      "grad_norm": 0.5939638018608093,
      "learning_rate": 1.9951720400727495e-05,
      "loss": 1.6704,
      "mean_token_accuracy": 0.5779796853661537,
      "num_tokens": 1423391.0,
      "step": 190
    },
    {
      "entropy": 2.211505854129791,
      "epoch": 0.64,
      "grad_norm": 0.6119778156280518,
      "learning_rate": 1.9941031846934213e-05,
      "loss": 1.6223,
      "mean_token_accuracy": 0.5848233133554459,
      "num_tokens": 1499124.0,
      "step": 200
    },
    {
      "entropy": 2.2195493161678312,
      "epoch": 0.672,
      "grad_norm": 0.6129831671714783,
      "learning_rate": 1.9929278846732883e-05,
      "loss": 1.5886,
      "mean_token_accuracy": 0.5897421136498451,
      "num_tokens": 1573541.0,
      "step": 210
    },
    {
      "entropy": 2.2096123576164244,
      "epoch": 0.704,
      "grad_norm": 0.6091306209564209,
      "learning_rate": 1.9916462658588328e-05,
      "loss": 1.6031,
      "mean_token_accuracy": 0.5894487425684929,
      "num_tokens": 1649546.0,
      "step": 220
    },
    {
      "entropy": 2.249841979146004,
      "epoch": 0.736,
      "grad_norm": 0.570816695690155,
      "learning_rate": 1.9902584654807325e-05,
      "loss": 1.5876,
      "mean_token_accuracy": 0.5911228567361831,
      "num_tokens": 1722199.0,
      "step": 230
    },
    {
      "entropy": 2.1915894985198974,
      "epoch": 0.768,
      "grad_norm": 0.5748864412307739,
      "learning_rate": 1.988764632139168e-05,
      "loss": 1.5963,
      "mean_token_accuracy": 0.5891387596726417,
      "num_tokens": 1797304.0,
      "step": 240
    },
    {
      "entropy": 2.2358563423156737,
      "epoch": 0.8,
      "grad_norm": 0.6511492729187012,
      "learning_rate": 1.9871649257879115e-05,
      "loss": 1.6453,
      "mean_token_accuracy": 0.5792816638946533,
      "num_tokens": 1870113.0,
      "step": 250
    },
    {
      "entropy": 2.2169984579086304,
      "epoch": 0.832,
      "grad_norm": 0.5317641496658325,
      "learning_rate": 1.9854595177171968e-05,
      "loss": 1.6594,
      "mean_token_accuracy": 0.577045065164566,
      "num_tokens": 1947405.0,
      "step": 260
    },
    {
      "entropy": 2.2434292674064635,
      "epoch": 0.864,
      "grad_norm": 0.5399971604347229,
      "learning_rate": 1.9836485905353823e-05,
      "loss": 1.7158,
      "mean_token_accuracy": 0.5683416239917278,
      "num_tokens": 2026284.0,
      "step": 270
    },
    {
      "entropy": 2.227828550338745,
      "epoch": 0.896,
      "grad_norm": 0.5378643870353699,
      "learning_rate": 1.9817323381493933e-05,
      "loss": 1.6714,
      "mean_token_accuracy": 0.5818367518484593,
      "num_tokens": 2103986.0,
      "step": 280
    },
    {
      "entropy": 2.2110894501209257,
      "epoch": 0.928,
      "grad_norm": 0.5195969343185425,
      "learning_rate": 1.979710965743964e-05,
      "loss": 1.6239,
      "mean_token_accuracy": 0.5819958478212357,
      "num_tokens": 2177010.0,
      "step": 290
    },
    {
      "entropy": 2.1666628658771514,
      "epoch": 0.96,
      "grad_norm": 0.5663164258003235,
      "learning_rate": 1.977584689759664e-05,
      "loss": 1.6024,
      "mean_token_accuracy": 0.5876665830612182,
      "num_tokens": 2251285.0,
      "step": 300
    },
    {
      "entropy": 2.214203083515167,
      "epoch": 0.992,
      "grad_norm": 0.6764860153198242,
      "learning_rate": 1.9753537378697237e-05,
      "loss": 1.6446,
      "mean_token_accuracy": 0.5818003416061401,
      "num_tokens": 2325752.0,
      "step": 310
    },
    {
      "entropy": 2.16783396821273,
      "epoch": 1.0224,
      "grad_norm": 0.5795008540153503,
      "learning_rate": 1.9730183489556563e-05,
      "loss": 1.594,
      "mean_token_accuracy": 0.5867547392845154,
      "num_tokens": 2396254.0,
      "step": 320
    },
    {
      "entropy": 2.172953352332115,
      "epoch": 1.0544,
      "grad_norm": 0.6686444282531738,
      "learning_rate": 1.9705787730816776e-05,
      "loss": 1.613,
      "mean_token_accuracy": 0.5906373374164104,
      "num_tokens": 2470123.0,
      "step": 330
    },
    {
      "entropy": 2.2217346757650374,
      "epoch": 1.0864,
      "grad_norm": 0.6389091610908508,
      "learning_rate": 1.9680352714679324e-05,
      "loss": 1.7053,
      "mean_token_accuracy": 0.577599074691534,
      "num_tokens": 2545749.0,
      "step": 340
    },
    {
      "entropy": 2.138428696990013,
      "epoch": 1.1184,
      "grad_norm": 0.7369883060455322,
      "learning_rate": 1.9653881164625234e-05,
      "loss": 1.5599,
      "mean_token_accuracy": 0.5946489304304123,
      "num_tokens": 2623270.0,
      "step": 350
    },
    {
      "entropy": 2.147254040837288,
      "epoch": 1.1504,
      "grad_norm": 0.6707085967063904,
      "learning_rate": 1.9626375915123473e-05,
      "loss": 1.5843,
      "mean_token_accuracy": 0.5965728983283043,
      "num_tokens": 2697616.0,
      "step": 360
    },
    {
      "entropy": 2.1412769109010696,
      "epoch": 1.1824,
      "grad_norm": 0.7201400995254517,
      "learning_rate": 1.9597839911327475e-05,
      "loss": 1.58,
      "mean_token_accuracy": 0.5957784004509449,
      "num_tokens": 2771426.0,
      "step": 370
    },
    {
      "entropy": 2.164059528708458,
      "epoch": 1.2144,
      "grad_norm": 0.7561144232749939,
      "learning_rate": 1.9568276208759772e-05,
      "loss": 1.5872,
      "mean_token_accuracy": 0.5874110117554665,
      "num_tokens": 2846711.0,
      "step": 380
    },
    {
      "entropy": 2.205427420139313,
      "epoch": 1.2464,
      "grad_norm": 0.691585898399353,
      "learning_rate": 1.9537687972984804e-05,
      "loss": 1.625,
      "mean_token_accuracy": 0.5892911069095135,
      "num_tokens": 2920916.0,
      "step": 390
    },
    {
      "entropy": 2.1242104679346085,
      "epoch": 1.2784,
      "grad_norm": 0.6999676823616028,
      "learning_rate": 1.950607847926999e-05,
      "loss": 1.5606,
      "mean_token_accuracy": 0.5917269751429558,
      "num_tokens": 2996056.0,
      "step": 400
    },
    {
      "entropy": 2.114223065972328,
      "epoch": 1.3104,
      "grad_norm": 0.7616406679153442,
      "learning_rate": 1.947345111223502e-05,
      "loss": 1.5296,
      "mean_token_accuracy": 0.5938275754451752,
      "num_tokens": 3072912.0,
      "step": 410
    },
    {
      "entropy": 2.1418962299823763,
      "epoch": 1.3424,
      "grad_norm": 0.7253025770187378,
      "learning_rate": 1.943980936548942e-05,
      "loss": 1.575,
      "mean_token_accuracy": 0.5945621818304062,
      "num_tokens": 3148498.0,
      "step": 420
    },
    {
      "entropy": 2.109667718410492,
      "epoch": 1.3744,
      "grad_norm": 0.8988682627677917,
      "learning_rate": 1.9405156841258498e-05,
      "loss": 1.5796,
      "mean_token_accuracy": 0.5901263400912284,
      "num_tokens": 3224741.0,
      "step": 430
    },
    {
      "entropy": 2.179358023405075,
      "epoch": 1.4064,
      "grad_norm": 0.741558849811554,
      "learning_rate": 1.936949724999762e-05,
      "loss": 1.6507,
      "mean_token_accuracy": 0.581992793083191,
      "num_tokens": 3299366.0,
      "step": 440
    },
    {
      "entropy": 2.1574251472949983,
      "epoch": 1.4384000000000001,
      "grad_norm": 0.7538727521896362,
      "learning_rate": 1.9332834409994906e-05,
      "loss": 1.5771,
      "mean_token_accuracy": 0.5888051658868789,
      "num_tokens": 3374162.0,
      "step": 450
    },
    {
      "entropy": 2.1186763852834702,
      "epoch": 1.4704,
      "grad_norm": 0.7905173301696777,
      "learning_rate": 1.929517224696239e-05,
      "loss": 1.6138,
      "mean_token_accuracy": 0.584889967739582,
      "num_tokens": 3452582.0,
      "step": 460
    },
    {
      "entropy": 2.1135365635156633,
      "epoch": 1.5024,
      "grad_norm": 0.7416484951972961,
      "learning_rate": 1.9256514793615674e-05,
      "loss": 1.5623,
      "mean_token_accuracy": 0.5928735345602035,
      "num_tokens": 3527694.0,
      "step": 470
    },
    {
      "entropy": 2.146635016798973,
      "epoch": 1.5344,
      "grad_norm": 0.731999397277832,
      "learning_rate": 1.9216866189242095e-05,
      "loss": 1.5634,
      "mean_token_accuracy": 0.5988615363836288,
      "num_tokens": 3600277.0,
      "step": 480
    },
    {
      "entropy": 2.1472962319850923,
      "epoch": 1.5664,
      "grad_norm": 0.7493702173233032,
      "learning_rate": 1.9176230679257547e-05,
      "loss": 1.5891,
      "mean_token_accuracy": 0.5858126983046532,
      "num_tokens": 3674781.0,
      "step": 490
    },
    {
      "entropy": 2.1530486762523653,
      "epoch": 1.5984,
      "grad_norm": 0.8006687164306641,
      "learning_rate": 1.9134612614751865e-05,
      "loss": 1.5674,
      "mean_token_accuracy": 0.5904534175992012,
      "num_tokens": 3748434.0,
      "step": 500
    },
    {
      "entropy": 2.169738906621933,
      "epoch": 1.6303999999999998,
      "grad_norm": 0.9293455481529236,
      "learning_rate": 1.909201645202294e-05,
      "loss": 1.6104,
      "mean_token_accuracy": 0.5860036969184875,
      "num_tokens": 3823982.0,
      "step": 510
    },
    {
      "entropy": 2.178475347161293,
      "epoch": 1.6623999999999999,
      "grad_norm": 0.7716575860977173,
      "learning_rate": 1.904844675209956e-05,
      "loss": 1.6432,
      "mean_token_accuracy": 0.5838924221694469,
      "num_tokens": 3900064.0,
      "step": 520
    },
    {
      "entropy": 2.1585603266954423,
      "epoch": 1.6944,
      "grad_norm": 0.8225084543228149,
      "learning_rate": 1.9003908180253027e-05,
      "loss": 1.5957,
      "mean_token_accuracy": 0.5880116850137711,
      "num_tokens": 3974029.0,
      "step": 530
    },
    {
      "entropy": 2.111869788169861,
      "epoch": 1.7264,
      "grad_norm": 0.7035638093948364,
      "learning_rate": 1.8958405505497613e-05,
      "loss": 1.579,
      "mean_token_accuracy": 0.5890362292528153,
      "num_tokens": 4049974.0,
      "step": 540
    },
    {
      "entropy": 2.144411253929138,
      "epoch": 1.7584,
      "grad_norm": 0.7046850919723511,
      "learning_rate": 1.8911943600079934e-05,
      "loss": 1.5926,
      "mean_token_accuracy": 0.5874261602759361,
      "num_tokens": 4125206.0,
      "step": 550
    },
    {
      "entropy": 2.1093025386333464,
      "epoch": 1.7904,
      "grad_norm": 0.807727575302124,
      "learning_rate": 1.8864527438957223e-05,
      "loss": 1.5367,
      "mean_token_accuracy": 0.5988967984914779,
      "num_tokens": 4199365.0,
      "step": 560
    },
    {
      "entropy": 2.097169244289398,
      "epoch": 1.8224,
      "grad_norm": 0.7856780886650085,
      "learning_rate": 1.881616209926465e-05,
      "loss": 1.561,
      "mean_token_accuracy": 0.5948230788111687,
      "num_tokens": 4275889.0,
      "step": 570
    },
    {
      "entropy": 2.088553088903427,
      "epoch": 1.8544,
      "grad_norm": 0.8993458151817322,
      "learning_rate": 1.876685275977167e-05,
      "loss": 1.5557,
      "mean_token_accuracy": 0.5941933646798134,
      "num_tokens": 4350502.0,
      "step": 580
    },
    {
      "entropy": 2.132419008016586,
      "epoch": 1.8864,
      "grad_norm": 0.7769711017608643,
      "learning_rate": 1.8716604700327516e-05,
      "loss": 1.6105,
      "mean_token_accuracy": 0.5815305605530738,
      "num_tokens": 4426429.0,
      "step": 590
    },
    {
      "entropy": 2.1076891005039213,
      "epoch": 1.9184,
      "grad_norm": 0.9261249899864197,
      "learning_rate": 1.866542330129583e-05,
      "loss": 1.5307,
      "mean_token_accuracy": 0.5964644759893417,
      "num_tokens": 4500147.0,
      "step": 600
    },
    {
      "entropy": 2.114642283320427,
      "epoch": 1.9504000000000001,
      "grad_norm": 0.806425929069519,
      "learning_rate": 1.8613314042978576e-05,
      "loss": 1.5809,
      "mean_token_accuracy": 0.5901800125837326,
      "num_tokens": 4573438.0,
      "step": 610
    },
    {
      "entropy": 2.1167576968669892,
      "epoch": 1.9824000000000002,
      "grad_norm": 0.8191499710083008,
      "learning_rate": 1.856028250502923e-05,
      "loss": 1.6031,
      "mean_token_accuracy": 0.5843381330370903,
      "num_tokens": 4648156.0,
      "step": 620
    },
    {
      "entropy": 2.0566249019221257,
      "epoch": 2.0128,
      "grad_norm": 0.7406135201454163,
      "learning_rate": 1.8506334365855315e-05,
      "loss": 1.5187,
      "mean_token_accuracy": 0.6027438483740154,
      "num_tokens": 4719492.0,
      "step": 630
    },
    {
      "entropy": 2.0126763731241226,
      "epoch": 2.0448,
      "grad_norm": 0.8845784068107605,
      "learning_rate": 1.8451475402010405e-05,
      "loss": 1.4841,
      "mean_token_accuracy": 0.6069207280874253,
      "num_tokens": 4796271.0,
      "step": 640
    },
    {
      "entropy": 2.0516900300979612,
      "epoch": 2.0768,
      "grad_norm": 0.9927017092704773,
      "learning_rate": 1.8395711487575564e-05,
      "loss": 1.512,
      "mean_token_accuracy": 0.6031922519207,
      "num_tokens": 4870202.0,
      "step": 650
    },
    {
      "entropy": 2.0824343889951704,
      "epoch": 2.1088,
      "grad_norm": 0.927236795425415,
      "learning_rate": 1.8339048593530406e-05,
      "loss": 1.5843,
      "mean_token_accuracy": 0.5952437989413738,
      "num_tokens": 4945568.0,
      "step": 660
    },
    {
      "entropy": 2.0304481953382494,
      "epoch": 2.1408,
      "grad_norm": 0.874019205570221,
      "learning_rate": 1.8281492787113707e-05,
      "loss": 1.5096,
      "mean_token_accuracy": 0.5992600306868553,
      "num_tokens": 5020723.0,
      "step": 670
    },
    {
      "entropy": 2.0402441143989565,
      "epoch": 2.1728,
      "grad_norm": 0.8746942281723022,
      "learning_rate": 1.8223050231173802e-05,
      "loss": 1.5119,
      "mean_token_accuracy": 0.5994458049535751,
      "num_tokens": 5095780.0,
      "step": 680
    },
    {
      "entropy": 2.018441066145897,
      "epoch": 2.2048,
      "grad_norm": 1.063180923461914,
      "learning_rate": 1.816372718350864e-05,
      "loss": 1.4923,
      "mean_token_accuracy": 0.6064845189452172,
      "num_tokens": 5169733.0,
      "step": 690
    },
    {
      "entropy": 2.0563316702842713,
      "epoch": 2.2368,
      "grad_norm": 1.0281789302825928,
      "learning_rate": 1.810352999619574e-05,
      "loss": 1.5505,
      "mean_token_accuracy": 0.602813882380724,
      "num_tokens": 5246393.0,
      "step": 700
    },
    {
      "entropy": 2.0298285841941834,
      "epoch": 2.2688,
      "grad_norm": 1.070520281791687,
      "learning_rate": 1.804246511491206e-05,
      "loss": 1.5159,
      "mean_token_accuracy": 0.6006126523017883,
      "num_tokens": 5322244.0,
      "step": 710
    },
    {
      "entropy": 2.0195819228887557,
      "epoch": 2.3008,
      "grad_norm": 0.9672983884811401,
      "learning_rate": 1.7980539078243783e-05,
      "loss": 1.5166,
      "mean_token_accuracy": 0.6054230839014053,
      "num_tokens": 5399317.0,
      "step": 720
    },
    {
      "entropy": 2.045917159318924,
      "epoch": 2.3327999999999998,
      "grad_norm": 1.1228744983673096,
      "learning_rate": 1.791775851698622e-05,
      "loss": 1.5096,
      "mean_token_accuracy": 0.6015639662742615,
      "num_tokens": 5473195.0,
      "step": 730
    },
    {
      "entropy": 2.0935415983200074,
      "epoch": 2.3648,
      "grad_norm": 1.149794578552246,
      "learning_rate": 1.7854130153433785e-05,
      "loss": 1.5583,
      "mean_token_accuracy": 0.5921522840857506,
      "num_tokens": 5548357.0,
      "step": 740
    },
    {
      "entropy": 2.044076007604599,
      "epoch": 2.3968,
      "grad_norm": 1.063625693321228,
      "learning_rate": 1.7789660800660222e-05,
      "loss": 1.5013,
      "mean_token_accuracy": 0.5974589124321937,
      "num_tokens": 5620915.0,
      "step": 750
    },
    {
      "entropy": 2.092478734254837,
      "epoch": 2.4288,
      "grad_norm": 1.1822012662887573,
      "learning_rate": 1.7724357361789075e-05,
      "loss": 1.5552,
      "mean_token_accuracy": 0.5929681301116944,
      "num_tokens": 5693406.0,
      "step": 760
    },
    {
      "entropy": 2.0430804908275606,
      "epoch": 2.4608,
      "grad_norm": 0.9921984076499939,
      "learning_rate": 1.765822682925453e-05,
      "loss": 1.4944,
      "mean_token_accuracy": 0.6029774472117424,
      "num_tokens": 5770143.0,
      "step": 770
    },
    {
      "entropy": 2.049290281534195,
      "epoch": 2.4928,
      "grad_norm": 1.0144131183624268,
      "learning_rate": 1.7591276284052695e-05,
      "loss": 1.5437,
      "mean_token_accuracy": 0.5986773043870925,
      "num_tokens": 5844022.0,
      "step": 780
    },
    {
      "entropy": 2.033898201584816,
      "epoch": 2.5248,
      "grad_norm": 1.1700315475463867,
      "learning_rate": 1.7523512894983396e-05,
      "loss": 1.5197,
      "mean_token_accuracy": 0.5972102269530296,
      "num_tokens": 5919099.0,
      "step": 790
    },
    {
      "entropy": 2.03344586789608,
      "epoch": 2.5568,
      "grad_norm": 1.0503427982330322,
      "learning_rate": 1.745494391788257e-05,
      "loss": 1.5456,
      "mean_token_accuracy": 0.6011263683438302,
      "num_tokens": 5997797.0,
      "step": 800
    },
    {
      "entropy": 2.0796399265527725,
      "epoch": 2.5888,
      "grad_norm": 1.0316176414489746,
      "learning_rate": 1.7385576694845324e-05,
      "loss": 1.608,
      "mean_token_accuracy": 0.6024919278919697,
      "num_tokens": 6075434.0,
      "step": 810
    },
    {
      "entropy": 2.0257797837257385,
      "epoch": 2.6208,
      "grad_norm": 1.048309087753296,
      "learning_rate": 1.7315418653439802e-05,
      "loss": 1.4876,
      "mean_token_accuracy": 0.6070949509739876,
      "num_tokens": 6149232.0,
      "step": 820
    },
    {
      "entropy": 2.024846690893173,
      "epoch": 2.6528,
      "grad_norm": 1.186710000038147,
      "learning_rate": 1.7244477305911845e-05,
      "loss": 1.499,
      "mean_token_accuracy": 0.6022308841347694,
      "num_tokens": 6222180.0,
      "step": 830
    },
    {
      "entropy": 1.9938248336315154,
      "epoch": 2.6848,
      "grad_norm": 1.1091604232788086,
      "learning_rate": 1.717276024838062e-05,
      "loss": 1.4795,
      "mean_token_accuracy": 0.6044012248516083,
      "num_tokens": 6296902.0,
      "step": 840
    },
    {
      "entropy": 1.9988998174667358,
      "epoch": 2.7168,
      "grad_norm": 1.0359690189361572,
      "learning_rate": 1.710027516002526e-05,
      "loss": 1.5173,
      "mean_token_accuracy": 0.6025070771574974,
      "num_tokens": 6373494.0,
      "step": 850
    },
    {
      "entropy": 2.02343093752861,
      "epoch": 2.7488,
      "grad_norm": 1.1783568859100342,
      "learning_rate": 1.7027029802262598e-05,
      "loss": 1.5146,
      "mean_token_accuracy": 0.6033479735255242,
      "num_tokens": 6449229.0,
      "step": 860
    },
    {
      "entropy": 2.0429257422685625,
      "epoch": 2.7808,
      "grad_norm": 0.9909568428993225,
      "learning_rate": 1.6953032017916115e-05,
      "loss": 1.5473,
      "mean_token_accuracy": 0.5932901218533516,
      "num_tokens": 6525728.0,
      "step": 870
    },
    {
      "entropy": 2.0058376491069794,
      "epoch": 2.8128,
      "grad_norm": 1.0904430150985718,
      "learning_rate": 1.687828973037615e-05,
      "loss": 1.4545,
      "mean_token_accuracy": 0.6120153024792672,
      "num_tokens": 6599335.0,
      "step": 880
    },
    {
      "entropy": 2.005480855703354,
      "epoch": 2.8448,
      "grad_norm": 1.1638548374176025,
      "learning_rate": 1.6802810942751514e-05,
      "loss": 1.4887,
      "mean_token_accuracy": 0.6060751393437386,
      "num_tokens": 6672722.0,
      "step": 890
    },
    {
      "entropy": 2.0311779022216796,
      "epoch": 2.8768000000000002,
      "grad_norm": 1.1404571533203125,
      "learning_rate": 1.6726603737012527e-05,
      "loss": 1.5238,
      "mean_token_accuracy": 0.6015868663787842,
      "num_tokens": 6748069.0,
      "step": 900
    },
    {
      "entropy": 2.0126856863498688,
      "epoch": 2.9088000000000003,
      "grad_norm": 1.0942543745040894,
      "learning_rate": 1.6649676273125647e-05,
      "loss": 1.4984,
      "mean_token_accuracy": 0.6019899815320968,
      "num_tokens": 6820935.0,
      "step": 910
    },
    {
      "entropy": 1.9961138010025024,
      "epoch": 2.9408,
      "grad_norm": 1.0870610475540161,
      "learning_rate": 1.6572036788179728e-05,
      "loss": 1.4962,
      "mean_token_accuracy": 0.6030571654438972,
      "num_tokens": 6896286.0,
      "step": 920
    },
    {
      "entropy": 2.035824549198151,
      "epoch": 2.9728,
      "grad_norm": 1.0822633504867554,
      "learning_rate": 1.6493693595504022e-05,
      "loss": 1.5354,
      "mean_token_accuracy": 0.5986709952354431,
      "num_tokens": 6971854.0,
      "step": 930
    },
    {
      "entropy": 2.0243908260997974,
      "epoch": 3.0032,
      "grad_norm": 1.0899602174758911,
      "learning_rate": 1.6414655083778027e-05,
      "loss": 1.5032,
      "mean_token_accuracy": 0.5983682243447555,
      "num_tokens": 7041122.0,
      "step": 940
    },
    {
      "entropy": 1.9538823068141937,
      "epoch": 3.0352,
      "grad_norm": 1.3042237758636475,
      "learning_rate": 1.633492971613326e-05,
      "loss": 1.4604,
      "mean_token_accuracy": 0.6146818101406097,
      "num_tokens": 7116032.0,
      "step": 950
    },
    {
      "entropy": 1.9383916020393372,
      "epoch": 3.0672,
      "grad_norm": 1.397078037261963,
      "learning_rate": 1.6254526029247048e-05,
      "loss": 1.4019,
      "mean_token_accuracy": 0.6210932344198227,
      "num_tokens": 7189009.0,
      "step": 960
    },
    {
      "entropy": 1.9460978150367736,
      "epoch": 3.0992,
      "grad_norm": 1.2756887674331665,
      "learning_rate": 1.617345263242847e-05,
      "loss": 1.4623,
      "mean_token_accuracy": 0.6121616646647453,
      "num_tokens": 7263068.0,
      "step": 970
    },
    {
      "entropy": 1.9156711965799331,
      "epoch": 3.1312,
      "grad_norm": 1.1937649250030518,
      "learning_rate": 1.609171820669649e-05,
      "loss": 1.4301,
      "mean_token_accuracy": 0.6136599197983742,
      "num_tokens": 7338652.0,
      "step": 980
    },
    {
      "entropy": 1.9247682303190232,
      "epoch": 3.1632,
      "grad_norm": 1.3291118144989014,
      "learning_rate": 1.6009331503850448e-05,
      "loss": 1.4545,
      "mean_token_accuracy": 0.6153608947992325,
      "num_tokens": 7414529.0,
      "step": 990
    },
    {
      "entropy": 1.9066543668508529,
      "epoch": 3.1952,
      "grad_norm": 1.4356389045715332,
      "learning_rate": 1.5926301345532925e-05,
      "loss": 1.4413,
      "mean_token_accuracy": 0.612147618830204,
      "num_tokens": 7489106.0,
      "step": 1000
    },
    {
      "entropy": 1.895160937309265,
      "epoch": 3.2272,
      "grad_norm": 1.4345523118972778,
      "learning_rate": 1.5842636622285187e-05,
      "loss": 1.4207,
      "mean_token_accuracy": 0.6138400137424469,
      "num_tokens": 7564304.0,
      "step": 1010
    },
    {
      "entropy": 1.9546802312135696,
      "epoch": 3.2592,
      "grad_norm": 1.5242680311203003,
      "learning_rate": 1.575834629259519e-05,
      "loss": 1.4435,
      "mean_token_accuracy": 0.6153354361653328,
      "num_tokens": 7637409.0,
      "step": 1020
    },
    {
      "entropy": 1.912938117980957,
      "epoch": 3.2912,
      "grad_norm": 1.529726505279541,
      "learning_rate": 1.5673439381938365e-05,
      "loss": 1.4409,
      "mean_token_accuracy": 0.6191004544496537,
      "num_tokens": 7711595.0,
      "step": 1030
    },
    {
      "entropy": 1.8989770442247391,
      "epoch": 3.3232,
      "grad_norm": 1.3367948532104492,
      "learning_rate": 1.5587924981811196e-05,
      "loss": 1.394,
      "mean_token_accuracy": 0.624155393242836,
      "num_tokens": 7785750.0,
      "step": 1040
    },
    {
      "entropy": 1.932333904504776,
      "epoch": 3.3552,
      "grad_norm": 1.4732215404510498,
      "learning_rate": 1.5501812248757734e-05,
      "loss": 1.3959,
      "mean_token_accuracy": 0.6221834555268287,
      "num_tokens": 7859036.0,
      "step": 1050
    },
    {
      "entropy": 1.9492982983589173,
      "epoch": 3.3872,
      "grad_norm": 1.4499313831329346,
      "learning_rate": 1.5415110403389166e-05,
      "loss": 1.4633,
      "mean_token_accuracy": 0.6100246667861938,
      "num_tokens": 7933165.0,
      "step": 1060
    },
    {
      "entropy": 1.9063653618097305,
      "epoch": 3.4192,
      "grad_norm": 1.4364317655563354,
      "learning_rate": 1.5327828729396482e-05,
      "loss": 1.4216,
      "mean_token_accuracy": 0.6210869938135147,
      "num_tokens": 8009376.0,
      "step": 1070
    },
    {
      "entropy": 1.9919361650943757,
      "epoch": 3.4512,
      "grad_norm": 1.5573089122772217,
      "learning_rate": 1.5239976572556438e-05,
      "loss": 1.5899,
      "mean_token_accuracy": 0.5991086520254612,
      "num_tokens": 8086825.0,
      "step": 1080
    },
    {
      "entropy": 1.922476476430893,
      "epoch": 3.4832,
      "grad_norm": 1.3339344263076782,
      "learning_rate": 1.5151563339730849e-05,
      "loss": 1.4162,
      "mean_token_accuracy": 0.6182018965482712,
      "num_tokens": 8161726.0,
      "step": 1090
    },
    {
      "entropy": 1.9143129527568816,
      "epoch": 3.5152,
      "grad_norm": 1.4425708055496216,
      "learning_rate": 1.506259849785931e-05,
      "loss": 1.4076,
      "mean_token_accuracy": 0.6197950705885887,
      "num_tokens": 8237046.0,
      "step": 1100
    },
    {
      "entropy": 1.9093267023563385,
      "epoch": 3.5472,
      "grad_norm": 1.5437992811203003,
      "learning_rate": 1.497309157294555e-05,
      "loss": 1.4339,
      "mean_token_accuracy": 0.6177847877144813,
      "num_tokens": 8315350.0,
      "step": 1110
    },
    {
      "entropy": 1.9121424347162246,
      "epoch": 3.5792,
      "grad_norm": 1.3761622905731201,
      "learning_rate": 1.4883052149037395e-05,
      "loss": 1.4175,
      "mean_token_accuracy": 0.6171463698148727,
      "num_tokens": 8390383.0,
      "step": 1120
    },
    {
      "entropy": 1.883551675081253,
      "epoch": 3.6112,
      "grad_norm": 1.36739182472229,
      "learning_rate": 1.479248986720057e-05,
      "loss": 1.4158,
      "mean_token_accuracy": 0.6186214044690133,
      "num_tokens": 8468414.0,
      "step": 1130
    },
    {
      "entropy": 1.988349151611328,
      "epoch": 3.6432,
      "grad_norm": 1.4566738605499268,
      "learning_rate": 1.4701414424486353e-05,
      "loss": 1.5366,
      "mean_token_accuracy": 0.6110676810145378,
      "num_tokens": 8541715.0,
      "step": 1140
    },
    {
      "entropy": 1.9057112097740174,
      "epoch": 3.6752000000000002,
      "grad_norm": 1.499079704284668,
      "learning_rate": 1.4609835572893266e-05,
      "loss": 1.3991,
      "mean_token_accuracy": 0.6208718970417977,
      "num_tokens": 8615694.0,
      "step": 1150
    },
    {
      "entropy": 1.9219326049089431,
      "epoch": 3.7072000000000003,
      "grad_norm": 1.3865621089935303,
      "learning_rate": 1.4517763118322861e-05,
      "loss": 1.431,
      "mean_token_accuracy": 0.6143050745129586,
      "num_tokens": 8692473.0,
      "step": 1160
    },
    {
      "entropy": 1.9036399960517882,
      "epoch": 3.7392,
      "grad_norm": 1.5362603664398193,
      "learning_rate": 1.4425206919529747e-05,
      "loss": 1.4156,
      "mean_token_accuracy": 0.6199175521731377,
      "num_tokens": 8767618.0,
      "step": 1170
    },
    {
      "entropy": 1.9499989479780198,
      "epoch": 3.7712,
      "grad_norm": 1.663404941558838,
      "learning_rate": 1.4332176887065955e-05,
      "loss": 1.4668,
      "mean_token_accuracy": 0.605186915397644,
      "num_tokens": 8843100.0,
      "step": 1180
    },
    {
      "entropy": 1.9545456051826477,
      "epoch": 3.8032,
      "grad_norm": 1.6169345378875732,
      "learning_rate": 1.4238682982219753e-05,
      "loss": 1.4241,
      "mean_token_accuracy": 0.6206902250647545,
      "num_tokens": 8914604.0,
      "step": 1190
    },
    {
      "entropy": 1.9130536198616028,
      "epoch": 3.8352,
      "grad_norm": 1.472740650177002,
      "learning_rate": 1.4144735215949028e-05,
      "loss": 1.4271,
      "mean_token_accuracy": 0.6139126420021057,
      "num_tokens": 8989305.0,
      "step": 1200
    },
    {
      "entropy": 1.938635140657425,
      "epoch": 3.8672,
      "grad_norm": 1.4194226264953613,
      "learning_rate": 1.4050343647809354e-05,
      "loss": 1.4538,
      "mean_token_accuracy": 0.6131341770291329,
      "num_tokens": 9065589.0,
      "step": 1210
    },
    {
      "entropy": 1.9123675346374511,
      "epoch": 3.8992,
      "grad_norm": 1.5208053588867188,
      "learning_rate": 1.3955518384876863e-05,
      "loss": 1.4309,
      "mean_token_accuracy": 0.6139545351266861,
      "num_tokens": 9140150.0,
      "step": 1220
    },
    {
      "entropy": 1.9148090302944183,
      "epoch": 3.9312,
      "grad_norm": 1.6418218612670898,
      "learning_rate": 1.3860269580666004e-05,
      "loss": 1.4269,
      "mean_token_accuracy": 0.6169310078024864,
      "num_tokens": 9215796.0,
      "step": 1230
    },
    {
      "entropy": 1.9157740741968154,
      "epoch": 3.9632,
      "grad_norm": 1.4638084173202515,
      "learning_rate": 1.3764607434042353e-05,
      "loss": 1.4509,
      "mean_token_accuracy": 0.6164968460798264,
      "num_tokens": 9291010.0,
      "step": 1240
    },
    {
      "entropy": 1.9184510678052902,
      "epoch": 3.9952,
      "grad_norm": 1.5152716636657715,
      "learning_rate": 1.3668542188130567e-05,
      "loss": 1.4453,
      "mean_token_accuracy": 0.6112410992383956,
      "num_tokens": 9367186.0,
      "step": 1250
    },
    {
      "entropy": 1.9016748384425515,
      "epoch": 4.0256,
      "grad_norm": 1.490628719329834,
      "learning_rate": 1.3572084129217566e-05,
      "loss": 1.382,
      "mean_token_accuracy": 0.623968276538347,
      "num_tokens": 9439028.0,
      "step": 1260
    },
    {
      "entropy": 1.8026290327310561,
      "epoch": 4.0576,
      "grad_norm": 1.8969308137893677,
      "learning_rate": 1.347524358565115e-05,
      "loss": 1.3128,
      "mean_token_accuracy": 0.6386646836996078,
      "num_tokens": 9513855.0,
      "step": 1270
    },
    {
      "entropy": 1.8283424764871596,
      "epoch": 4.0896,
      "grad_norm": 1.5952194929122925,
      "learning_rate": 1.3378030926734052e-05,
      "loss": 1.3362,
      "mean_token_accuracy": 0.6328515768051147,
      "num_tokens": 9589080.0,
      "step": 1280
    },
    {
      "entropy": 1.8405955344438554,
      "epoch": 4.1216,
      "grad_norm": 1.6057584285736084,
      "learning_rate": 1.3280456561613653e-05,
      "loss": 1.4151,
      "mean_token_accuracy": 0.6261398203670978,
      "num_tokens": 9666808.0,
      "step": 1290
    },
    {
      "entropy": 1.8390818655490875,
      "epoch": 4.1536,
      "grad_norm": 1.8149824142456055,
      "learning_rate": 1.3182530938167409e-05,
      "loss": 1.3455,
      "mean_token_accuracy": 0.6318597674369812,
      "num_tokens": 9740267.0,
      "step": 1300
    },
    {
      "entropy": 1.8203887075185776,
      "epoch": 4.1856,
      "grad_norm": 1.6102676391601562,
      "learning_rate": 1.3084264541884118e-05,
      "loss": 1.3255,
      "mean_token_accuracy": 0.6316933467984199,
      "num_tokens": 9816400.0,
      "step": 1310
    },
    {
      "entropy": 1.8592366576194763,
      "epoch": 4.2176,
      "grad_norm": 1.9501773118972778,
      "learning_rate": 1.2985667894741197e-05,
      "loss": 1.3521,
      "mean_token_accuracy": 0.6301594719290733,
      "num_tokens": 9889311.0,
      "step": 1320
    },
    {
      "entropy": 1.8420085966587068,
      "epoch": 4.2496,
      "grad_norm": 1.6526106595993042,
      "learning_rate": 1.2886751554078015e-05,
      "loss": 1.3662,
      "mean_token_accuracy": 0.6302071824669838,
      "num_tokens": 9965339.0,
      "step": 1330
    },
    {
      "entropy": 1.8313881188631058,
      "epoch": 4.2816,
      "grad_norm": 1.6269904375076294,
      "learning_rate": 1.2787526111465453e-05,
      "loss": 1.3579,
      "mean_token_accuracy": 0.6328388035297394,
      "num_tokens": 10039668.0,
      "step": 1340
    },
    {
      "entropy": 1.858151137828827,
      "epoch": 4.3136,
      "grad_norm": 1.9028024673461914,
      "learning_rate": 1.2688002191571829e-05,
      "loss": 1.421,
      "mean_token_accuracy": 0.6276688367128372,
      "num_tokens": 10115387.0,
      "step": 1350
    },
    {
      "entropy": 1.8273844957351684,
      "epoch": 4.3456,
      "grad_norm": 1.7530555725097656,
      "learning_rate": 1.2588190451025209e-05,
      "loss": 1.3527,
      "mean_token_accuracy": 0.6345869660377502,
      "num_tokens": 10191506.0,
      "step": 1360
    },
    {
      "entropy": 1.8732422679662704,
      "epoch": 4.3776,
      "grad_norm": 1.7372691631317139,
      "learning_rate": 1.248810157727236e-05,
      "loss": 1.4132,
      "mean_token_accuracy": 0.6252246856689453,
      "num_tokens": 10268756.0,
      "step": 1370
    },
    {
      "entropy": 1.8583054572343827,
      "epoch": 4.4096,
      "grad_norm": 1.6993470191955566,
      "learning_rate": 1.2387746287434385e-05,
      "loss": 1.3638,
      "mean_token_accuracy": 0.6286717876791954,
      "num_tokens": 10341779.0,
      "step": 1380
    },
    {
      "entropy": 1.8324467271566391,
      "epoch": 4.4416,
      "grad_norm": 1.7818169593811035,
      "learning_rate": 1.2287135327159165e-05,
      "loss": 1.3372,
      "mean_token_accuracy": 0.6361263945698739,
      "num_tokens": 10414642.0,
      "step": 1390
    },
    {
      "entropy": 1.8514392852783204,
      "epoch": 4.4736,
      "grad_norm": 1.7585517168045044,
      "learning_rate": 1.2186279469470757e-05,
      "loss": 1.3697,
      "mean_token_accuracy": 0.628801380097866,
      "num_tokens": 10489517.0,
      "step": 1400
    },
    {
      "entropy": 1.8218136370182036,
      "epoch": 4.5056,
      "grad_norm": 1.9843116998672485,
      "learning_rate": 1.2085189513615872e-05,
      "loss": 1.3628,
      "mean_token_accuracy": 0.6295172199606895,
      "num_tokens": 10565467.0,
      "step": 1410
    },
    {
      "entropy": 1.8919565021991729,
      "epoch": 4.5376,
      "grad_norm": 1.9309132099151611,
      "learning_rate": 1.1983876283907522e-05,
      "loss": 1.4467,
      "mean_token_accuracy": 0.6263746194541454,
      "num_tokens": 10641283.0,
      "step": 1420
    },
    {
      "entropy": 1.8356508910655975,
      "epoch": 4.5696,
      "grad_norm": 1.7685068845748901,
      "learning_rate": 1.1882350628566008e-05,
      "loss": 1.3631,
      "mean_token_accuracy": 0.624418406188488,
      "num_tokens": 10716701.0,
      "step": 1430
    },
    {
      "entropy": 1.8288098931312562,
      "epoch": 4.6016,
      "grad_norm": 1.8276050090789795,
      "learning_rate": 1.178062341855732e-05,
      "loss": 1.3619,
      "mean_token_accuracy": 0.6286922857165337,
      "num_tokens": 10791427.0,
      "step": 1440
    },
    {
      "entropy": 1.8557640790939331,
      "epoch": 4.6336,
      "grad_norm": 1.7773240804672241,
      "learning_rate": 1.1678705546429132e-05,
      "loss": 1.383,
      "mean_token_accuracy": 0.6216814562678337,
      "num_tokens": 10866356.0,
      "step": 1450
    },
    {
      "entropy": 1.8483826667070389,
      "epoch": 4.6655999999999995,
      "grad_norm": 1.831931710243225,
      "learning_rate": 1.1576607925144456e-05,
      "loss": 1.355,
      "mean_token_accuracy": 0.6278511002659798,
      "num_tokens": 10940772.0,
      "step": 1460
    },
    {
      "entropy": 1.8824394553899766,
      "epoch": 4.6975999999999996,
      "grad_norm": 1.9213542938232422,
      "learning_rate": 1.1474341486913146e-05,
      "loss": 1.3767,
      "mean_token_accuracy": 0.6256057649850846,
      "num_tokens": 11016144.0,
      "step": 1470
    },
    {
      "entropy": 1.8709469974040984,
      "epoch": 4.7296,
      "grad_norm": 1.8768925666809082,
      "learning_rate": 1.1371917182021297e-05,
      "loss": 1.3734,
      "mean_token_accuracy": 0.6317574754357338,
      "num_tokens": 11089939.0,
      "step": 1480
    },
    {
      "entropy": 1.8673742085695266,
      "epoch": 4.7616,
      "grad_norm": 1.796302318572998,
      "learning_rate": 1.1269345977658747e-05,
      "loss": 1.3707,
      "mean_token_accuracy": 0.6238353416323662,
      "num_tokens": 11166087.0,
      "step": 1490
    },
    {
      "entropy": 1.8310889720916748,
      "epoch": 4.7936,
      "grad_norm": 1.8969939947128296,
      "learning_rate": 1.1166638856744747e-05,
      "loss": 1.3373,
      "mean_token_accuracy": 0.6348015949130058,
      "num_tokens": 11240732.0,
      "step": 1500
    },
    {
      "entropy": 1.8809226244688033,
      "epoch": 4.8256,
      "grad_norm": 1.642104983329773,
      "learning_rate": 1.1063806816751957e-05,
      "loss": 1.3792,
      "mean_token_accuracy": 0.6250617265701294,
      "num_tokens": 11316878.0,
      "step": 1510
    },
    {
      "entropy": 1.8715822875499726,
      "epoch": 4.8576,
      "grad_norm": 1.962158441543579,
      "learning_rate": 1.0960860868528872e-05,
      "loss": 1.3711,
      "mean_token_accuracy": 0.6293752744793892,
      "num_tokens": 11389042.0,
      "step": 1520
    },
    {
      "entropy": 1.8657191127538681,
      "epoch": 4.8896,
      "grad_norm": 1.9577444791793823,
      "learning_rate": 1.0857812035120845e-05,
      "loss": 1.379,
      "mean_token_accuracy": 0.6259156972169876,
      "num_tokens": 11464215.0,
      "step": 1530
    },
    {
      "entropy": 1.8811951220035552,
      "epoch": 4.9216,
      "grad_norm": 2.015150785446167,
      "learning_rate": 1.0754671350589752e-05,
      "loss": 1.4155,
      "mean_token_accuracy": 0.626779156178236,
      "num_tokens": 11539122.0,
      "step": 1540
    },
    {
      "entropy": 1.863905319571495,
      "epoch": 4.9536,
      "grad_norm": 1.8474093675613403,
      "learning_rate": 1.065144985883253e-05,
      "loss": 1.3409,
      "mean_token_accuracy": 0.6319419264793396,
      "num_tokens": 11613016.0,
      "step": 1550
    },
    {
      "entropy": 1.836970153450966,
      "epoch": 4.9856,
      "grad_norm": 1.8822177648544312,
      "learning_rate": 1.054815861239864e-05,
      "loss": 1.3514,
      "mean_token_accuracy": 0.6292115703225136,
      "num_tokens": 11688143.0,
      "step": 1560
    },
    {
      "entropy": 1.8377465950815302,
      "epoch": 5.016,
      "grad_norm": 1.8221346139907837,
      "learning_rate": 1.0444808671306588e-05,
      "loss": 1.3028,
      "mean_token_accuracy": 0.6413120329380035,
      "num_tokens": 11758768.0,
      "step": 1570
    },
    {
      "entropy": 1.7883025139570237,
      "epoch": 5.048,
      "grad_norm": 2.1959595680236816,
      "learning_rate": 1.034141110185968e-05,
      "loss": 1.2797,
      "mean_token_accuracy": 0.6479741290211678,
      "num_tokens": 11832210.0,
      "step": 1580
    },
    {
      "entropy": 1.7955584406852723,
      "epoch": 5.08,
      "grad_norm": 2.106905698776245,
      "learning_rate": 1.0237976975461074e-05,
      "loss": 1.2881,
      "mean_token_accuracy": 0.6424632370471954,
      "num_tokens": 11906115.0,
      "step": 1590
    },
    {
      "entropy": 1.7998322755098344,
      "epoch": 5.112,
      "grad_norm": 2.327314615249634,
      "learning_rate": 1.0134517367428309e-05,
      "loss": 1.3324,
      "mean_token_accuracy": 0.6460248224437237,
      "num_tokens": 11981328.0,
      "step": 1600
    },
    {
      "entropy": 1.7885828018188477,
      "epoch": 5.144,
      "grad_norm": 2.1001713275909424,
      "learning_rate": 1.0031043355807386e-05,
      "loss": 1.3098,
      "mean_token_accuracy": 0.63900758177042,
      "num_tokens": 12056453.0,
      "step": 1610
    },
    {
      "entropy": 1.769435602426529,
      "epoch": 5.176,
      "grad_norm": 2.1210567951202393,
      "learning_rate": 9.927566020186592e-06,
      "loss": 1.2892,
      "mean_token_accuracy": 0.6432970002293587,
      "num_tokens": 12133433.0,
      "step": 1620
    },
    {
      "entropy": 1.7907766073942184,
      "epoch": 5.208,
      "grad_norm": 2.1842658519744873,
      "learning_rate": 9.82409644051013e-06,
      "loss": 1.2856,
      "mean_token_accuracy": 0.6423615619540215,
      "num_tokens": 12207150.0,
      "step": 1630
    },
    {
      "entropy": 1.7834827870130538,
      "epoch": 5.24,
      "grad_norm": 2.2503459453582764,
      "learning_rate": 9.720645695891733e-06,
      "loss": 1.3066,
      "mean_token_accuracy": 0.6417693704366684,
      "num_tokens": 12282584.0,
      "step": 1640
    },
    {
      "entropy": 1.763256973028183,
      "epoch": 5.272,
      "grad_norm": 1.9505388736724854,
      "learning_rate": 9.617224863428346e-06,
      "loss": 1.2951,
      "mean_token_accuracy": 0.6429389595985413,
      "num_tokens": 12359793.0,
      "step": 1650
    },
    {
      "entropy": 1.8142763644456863,
      "epoch": 5.304,
      "grad_norm": 1.9957698583602905,
      "learning_rate": 9.513845017014048e-06,
      "loss": 1.33,
      "mean_token_accuracy": 0.6413653999567032,
      "num_tokens": 12434251.0,
      "step": 1660
    },
    {
      "entropy": 1.797221601009369,
      "epoch": 5.336,
      "grad_norm": 2.5095462799072266,
      "learning_rate": 9.410517226154276e-06,
      "loss": 1.2978,
      "mean_token_accuracy": 0.6389835774898529,
      "num_tokens": 12508416.0,
      "step": 1670
    },
    {
      "entropy": 1.8157870292663574,
      "epoch": 5.368,
      "grad_norm": 2.1890602111816406,
      "learning_rate": 9.30725255478058e-06,
      "loss": 1.3183,
      "mean_token_accuracy": 0.6448161751031876,
      "num_tokens": 12582896.0,
      "step": 1680
    },
    {
      "entropy": 1.7990054041147232,
      "epoch": 5.4,
      "grad_norm": 2.3904025554656982,
      "learning_rate": 9.204062060065915e-06,
      "loss": 1.3318,
      "mean_token_accuracy": 0.636146479845047,
      "num_tokens": 12656802.0,
      "step": 1690
    },
    {
      "entropy": 1.8003453463315964,
      "epoch": 5.432,
      "grad_norm": 1.9204304218292236,
      "learning_rate": 9.100956791240699e-06,
      "loss": 1.3186,
      "mean_token_accuracy": 0.6372130662202835,
      "num_tokens": 12733283.0,
      "step": 1700
    },
    {
      "entropy": 1.8101116061210631,
      "epoch": 5.464,
      "grad_norm": 2.009500026702881,
      "learning_rate": 8.997947788409696e-06,
      "loss": 1.3247,
      "mean_token_accuracy": 0.6406339526176452,
      "num_tokens": 12810272.0,
      "step": 1710
    },
    {
      "entropy": 1.764935952425003,
      "epoch": 5.496,
      "grad_norm": 2.2038798332214355,
      "learning_rate": 8.89504608136989e-06,
      "loss": 1.2792,
      "mean_token_accuracy": 0.6445836886763573,
      "num_tokens": 12885633.0,
      "step": 1720
    },
    {
      "entropy": 1.7950240582227708,
      "epoch": 5.5280000000000005,
      "grad_norm": 2.0160531997680664,
      "learning_rate": 8.792262688429445e-06,
      "loss": 1.2934,
      "mean_token_accuracy": 0.6469692558050155,
      "num_tokens": 12961131.0,
      "step": 1730
    },
    {
      "entropy": 1.7804677098989488,
      "epoch": 5.5600000000000005,
      "grad_norm": 2.1956582069396973,
      "learning_rate": 8.689608615227933e-06,
      "loss": 1.2969,
      "mean_token_accuracy": 0.6438389763236045,
      "num_tokens": 13036481.0,
      "step": 1740
    },
    {
      "entropy": 1.7932062089443206,
      "epoch": 5.592,
      "grad_norm": 2.2215394973754883,
      "learning_rate": 8.587094853557877e-06,
      "loss": 1.2907,
      "mean_token_accuracy": 0.6460438340902328,
      "num_tokens": 13111001.0,
      "step": 1750
    },
    {
      "entropy": 1.8026408910751344,
      "epoch": 5.624,
      "grad_norm": 2.3881425857543945,
      "learning_rate": 8.484732380187785e-06,
      "loss": 1.3049,
      "mean_token_accuracy": 0.6414234206080437,
      "num_tokens": 13186347.0,
      "step": 1760
    },
    {
      "entropy": 1.8440747499465941,
      "epoch": 5.656,
      "grad_norm": 2.2154159545898438,
      "learning_rate": 8.382532155686825e-06,
      "loss": 1.3797,
      "mean_token_accuracy": 0.6365857936441899,
      "num_tokens": 13261455.0,
      "step": 1770
    },
    {
      "entropy": 1.7975190997123718,
      "epoch": 5.688,
      "grad_norm": 2.1991233825683594,
      "learning_rate": 8.280505123251183e-06,
      "loss": 1.3191,
      "mean_token_accuracy": 0.6393151715397835,
      "num_tokens": 13338064.0,
      "step": 1780
    },
    {
      "entropy": 1.8396487146615983,
      "epoch": 5.72,
      "grad_norm": 2.0190858840942383,
      "learning_rate": 8.178662207532343e-06,
      "loss": 1.4052,
      "mean_token_accuracy": 0.629064130038023,
      "num_tokens": 13414806.0,
      "step": 1790
    },
    {
      "entropy": 1.7840806126594544,
      "epoch": 5.752,
      "grad_norm": 2.3335204124450684,
      "learning_rate": 8.077014313467274e-06,
      "loss": 1.2701,
      "mean_token_accuracy": 0.6464540064334869,
      "num_tokens": 13489075.0,
      "step": 1800
    },
    {
      "entropy": 1.7840022534132003,
      "epoch": 5.784,
      "grad_norm": 2.2151618003845215,
      "learning_rate": 7.975572325110819e-06,
      "loss": 1.3248,
      "mean_token_accuracy": 0.6358998969197274,
      "num_tokens": 13565636.0,
      "step": 1810
    },
    {
      "entropy": 1.7677135676145554,
      "epoch": 5.816,
      "grad_norm": 2.11505389213562,
      "learning_rate": 7.874347104470234e-06,
      "loss": 1.2765,
      "mean_token_accuracy": 0.6448719501495361,
      "num_tokens": 13641112.0,
      "step": 1820
    },
    {
      "entropy": 1.7586044907569884,
      "epoch": 5.848,
      "grad_norm": 2.178250789642334,
      "learning_rate": 7.773349490342157e-06,
      "loss": 1.2846,
      "mean_token_accuracy": 0.6450280979275703,
      "num_tokens": 13715158.0,
      "step": 1830
    },
    {
      "entropy": 1.8128920108079911,
      "epoch": 5.88,
      "grad_norm": 2.2499353885650635,
      "learning_rate": 7.672590297152013e-06,
      "loss": 1.3248,
      "mean_token_accuracy": 0.6343795835971833,
      "num_tokens": 13791086.0,
      "step": 1840
    },
    {
      "entropy": 1.7873643577098846,
      "epoch": 5.912,
      "grad_norm": 2.1989104747772217,
      "learning_rate": 7.572080313796064e-06,
      "loss": 1.2907,
      "mean_token_accuracy": 0.6413815975189209,
      "num_tokens": 13865700.0,
      "step": 1850
    },
    {
      "entropy": 1.790488451719284,
      "epoch": 5.944,
      "grad_norm": 2.2605504989624023,
      "learning_rate": 7.471830302486151e-06,
      "loss": 1.2955,
      "mean_token_accuracy": 0.6424889475107193,
      "num_tokens": 13938540.0,
      "step": 1860
    },
    {
      "entropy": 1.7985246628522873,
      "epoch": 5.976,
      "grad_norm": 2.3228533267974854,
      "learning_rate": 7.371850997597355e-06,
      "loss": 1.2854,
      "mean_token_accuracy": 0.6457341402769089,
      "num_tokens": 14011087.0,
      "step": 1870
    },
    {
      "entropy": 1.7890221727521796,
      "epoch": 6.0064,
      "grad_norm": 2.192910671234131,
      "learning_rate": 7.272153104518567e-06,
      "loss": 1.2796,
      "mean_token_accuracy": 0.6482133128141102,
      "num_tokens": 14082075.0,
      "step": 1880
    },
    {
      "entropy": 1.7633086562156677,
      "epoch": 6.0384,
      "grad_norm": 2.368185043334961,
      "learning_rate": 7.172747298506224e-06,
      "loss": 1.2703,
      "mean_token_accuracy": 0.6497290328145027,
      "num_tokens": 14156298.0,
      "step": 1890
    },
    {
      "entropy": 1.750128635764122,
      "epoch": 6.0704,
      "grad_norm": 2.36487078666687,
      "learning_rate": 7.073644223541227e-06,
      "loss": 1.2521,
      "mean_token_accuracy": 0.6534707516431808,
      "num_tokens": 14232528.0,
      "step": 1900
    },
    {
      "entropy": 1.7416553497314453,
      "epoch": 6.1024,
      "grad_norm": 2.3927595615386963,
      "learning_rate": 6.974854491189243e-06,
      "loss": 1.217,
      "mean_token_accuracy": 0.6588135868310928,
      "num_tokens": 14307073.0,
      "step": 1910
    },
    {
      "entropy": 1.7359241485595702,
      "epoch": 6.1344,
      "grad_norm": 2.1107988357543945,
      "learning_rate": 6.876388679464437e-06,
      "loss": 1.2763,
      "mean_token_accuracy": 0.6550255373120308,
      "num_tokens": 14383819.0,
      "step": 1920
    },
    {
      "entropy": 1.7380403220653533,
      "epoch": 6.1664,
      "grad_norm": 2.4158387184143066,
      "learning_rate": 6.7782573316968424e-06,
      "loss": 1.2359,
      "mean_token_accuracy": 0.656632873415947,
      "num_tokens": 14460092.0,
      "step": 1930
    },
    {
      "entropy": 1.7227638810873032,
      "epoch": 6.1984,
      "grad_norm": 2.3467485904693604,
      "learning_rate": 6.6804709554034075e-06,
      "loss": 1.2311,
      "mean_token_accuracy": 0.654091839492321,
      "num_tokens": 14534160.0,
      "step": 1940
    },
    {
      "entropy": 1.7244948148727417,
      "epoch": 6.2304,
      "grad_norm": 2.760057210922241,
      "learning_rate": 6.583040021162905e-06,
      "loss": 1.2189,
      "mean_token_accuracy": 0.6611428812146187,
      "num_tokens": 14608592.0,
      "step": 1950
    },
    {
      "entropy": 1.7471657902002335,
      "epoch": 6.2624,
      "grad_norm": 2.3923745155334473,
      "learning_rate": 6.485974961494772e-06,
      "loss": 1.2631,
      "mean_token_accuracy": 0.6524021357297898,
      "num_tokens": 14683538.0,
      "step": 1960
    },
    {
      "entropy": 1.7506494253873826,
      "epoch": 6.2943999999999996,
      "grad_norm": 2.4149715900421143,
      "learning_rate": 6.389286169742048e-06,
      "loss": 1.2468,
      "mean_token_accuracy": 0.6567713841795921,
      "num_tokens": 14755778.0,
      "step": 1970
    },
    {
      "entropy": 1.7104488879442215,
      "epoch": 6.3264,
      "grad_norm": 2.632632255554199,
      "learning_rate": 6.292983998958478e-06,
      "loss": 1.2267,
      "mean_token_accuracy": 0.6561126798391342,
      "num_tokens": 14831036.0,
      "step": 1980
    },
    {
      "entropy": 1.7591658294200898,
      "epoch": 6.3584,
      "grad_norm": 2.4012722969055176,
      "learning_rate": 6.1970787607999815e-06,
      "loss": 1.2693,
      "mean_token_accuracy": 0.6490694522857666,
      "num_tokens": 14906610.0,
      "step": 1990
    },
    {
      "entropy": 1.7317969173192977,
      "epoch": 6.3904,
      "grad_norm": 2.8288748264312744,
      "learning_rate": 6.101580724420478e-06,
      "loss": 1.235,
      "mean_token_accuracy": 0.6564200609922409,
      "num_tokens": 14980134.0,
      "step": 2000
    },
    {
      "entropy": 1.7617577254772185,
      "epoch": 6.4224,
      "grad_norm": 2.4008944034576416,
      "learning_rate": 6.00650011537235e-06,
      "loss": 1.2773,
      "mean_token_accuracy": 0.6494350075721741,
      "num_tokens": 15054969.0,
      "step": 2010
    },
    {
      "entropy": 1.749829688668251,
      "epoch": 6.4544,
      "grad_norm": 2.5665228366851807,
      "learning_rate": 5.911847114511497e-06,
      "loss": 1.2512,
      "mean_token_accuracy": 0.6512764275074006,
      "num_tokens": 15129421.0,
      "step": 2020
    },
    {
      "entropy": 1.7387797951698303,
      "epoch": 6.4864,
      "grad_norm": 2.6020922660827637,
      "learning_rate": 5.817631856907233e-06,
      "loss": 1.2477,
      "mean_token_accuracy": 0.6530226185917855,
      "num_tokens": 15203465.0,
      "step": 2030
    },
    {
      "entropy": 1.7363551884889603,
      "epoch": 6.5184,
      "grad_norm": 2.161478281021118,
      "learning_rate": 5.723864430757047e-06,
      "loss": 1.2692,
      "mean_token_accuracy": 0.6527093783020973,
      "num_tokens": 15279761.0,
      "step": 2040
    },
    {
      "entropy": 1.7563295543193818,
      "epoch": 6.5504,
      "grad_norm": 2.5587289333343506,
      "learning_rate": 5.630554876306407e-06,
      "loss": 1.2211,
      "mean_token_accuracy": 0.6574550330638885,
      "num_tokens": 15351301.0,
      "step": 2050
    },
    {
      "entropy": 1.7521151036024094,
      "epoch": 6.5824,
      "grad_norm": 2.4042234420776367,
      "learning_rate": 5.537713184773686e-06,
      "loss": 1.271,
      "mean_token_accuracy": 0.6478641331195831,
      "num_tokens": 15427936.0,
      "step": 2060
    },
    {
      "entropy": 1.7270145863294601,
      "epoch": 6.6144,
      "grad_norm": 2.3122522830963135,
      "learning_rate": 5.44534929728036e-06,
      "loss": 1.224,
      "mean_token_accuracy": 0.6566437393426895,
      "num_tokens": 15502561.0,
      "step": 2070
    },
    {
      "entropy": 1.7568089962005615,
      "epoch": 6.6464,
      "grad_norm": 2.461474895477295,
      "learning_rate": 5.353473103786511e-06,
      "loss": 1.2864,
      "mean_token_accuracy": 0.6471308276057244,
      "num_tokens": 15578053.0,
      "step": 2080
    },
    {
      "entropy": 1.7190027862787247,
      "epoch": 6.6784,
      "grad_norm": 2.4412550926208496,
      "learning_rate": 5.262094442031901e-06,
      "loss": 1.2092,
      "mean_token_accuracy": 0.6601713746786118,
      "num_tokens": 15653342.0,
      "step": 2090
    },
    {
      "entropy": 1.717634916305542,
      "epoch": 6.7104,
      "grad_norm": 2.276007890701294,
      "learning_rate": 5.171223096482533e-06,
      "loss": 1.2271,
      "mean_token_accuracy": 0.6595920532941818,
      "num_tokens": 15730387.0,
      "step": 2100
    },
    {
      "entropy": 1.7230647921562194,
      "epoch": 6.7424,
      "grad_norm": 2.480471134185791,
      "learning_rate": 5.080868797283019e-06,
      "loss": 1.229,
      "mean_token_accuracy": 0.6568982020020485,
      "num_tokens": 15804405.0,
      "step": 2110
    },
    {
      "entropy": 1.7535502433776855,
      "epoch": 6.7744,
      "grad_norm": 2.448997974395752,
      "learning_rate": 4.9910412192146795e-06,
      "loss": 1.2584,
      "mean_token_accuracy": 0.648795773088932,
      "num_tokens": 15878537.0,
      "step": 2120
    },
    {
      "entropy": 1.786664029955864,
      "epoch": 6.8064,
      "grad_norm": 2.430039405822754,
      "learning_rate": 4.901749980659617e-06,
      "loss": 1.3358,
      "mean_token_accuracy": 0.6427689291536808,
      "num_tokens": 15952964.0,
      "step": 2130
    },
    {
      "entropy": 1.7594995677471161,
      "epoch": 6.8384,
      "grad_norm": 2.469172716140747,
      "learning_rate": 4.813004642570822e-06,
      "loss": 1.2844,
      "mean_token_accuracy": 0.6534359715878963,
      "num_tokens": 16028086.0,
      "step": 2140
    },
    {
      "entropy": 1.7347292125225067,
      "epoch": 6.8704,
      "grad_norm": 2.6162445545196533,
      "learning_rate": 4.724814707448418e-06,
      "loss": 1.2707,
      "mean_token_accuracy": 0.6447671175003051,
      "num_tokens": 16103263.0,
      "step": 2150
    },
    {
      "entropy": 1.7325938045978546,
      "epoch": 6.9024,
      "grad_norm": 2.416431188583374,
      "learning_rate": 4.637189618322173e-06,
      "loss": 1.2794,
      "mean_token_accuracy": 0.6439008563756943,
      "num_tokens": 16182360.0,
      "step": 2160
    },
    {
      "entropy": 1.7763712674379348,
      "epoch": 6.9344,
      "grad_norm": 2.3447437286376953,
      "learning_rate": 4.550138757740381e-06,
      "loss": 1.3043,
      "mean_token_accuracy": 0.650251479446888,
      "num_tokens": 16256272.0,
      "step": 2170
    },
    {
      "entropy": 1.739478302001953,
      "epoch": 6.9664,
      "grad_norm": 2.650451183319092,
      "learning_rate": 4.463671446765206e-06,
      "loss": 1.259,
      "mean_token_accuracy": 0.6514677822589874,
      "num_tokens": 16330984.0,
      "step": 2180
    },
    {
      "entropy": 1.7292406976222991,
      "epoch": 6.9984,
      "grad_norm": 2.5442306995391846,
      "learning_rate": 4.377796943974641e-06,
      "loss": 1.2554,
      "mean_token_accuracy": 0.6506337657570839,
      "num_tokens": 16406982.0,
      "step": 2190
    }
  ],
  "logging_steps": 10,
  "max_steps": 3130,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5937216534515548e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}