{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1370,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.8819891855120658,
      "epoch": 0.014603870025556773,
      "grad_norm": 0.8598580956459045,
      "learning_rate": 1.7475728155339808e-06,
      "loss": 0.8448,
      "mean_token_accuracy": 0.7740211486816406,
      "num_tokens": 770936.0,
      "step": 10
    },
    {
      "entropy": 0.9709278956055641,
      "epoch": 0.029207740051113547,
      "grad_norm": 0.615368127822876,
      "learning_rate": 3.689320388349515e-06,
      "loss": 0.8205,
      "mean_token_accuracy": 0.7773708969354629,
      "num_tokens": 1541367.0,
      "step": 20
    },
    {
      "entropy": 1.0815545290708541,
      "epoch": 0.04381161007667032,
      "grad_norm": 0.6463781595230103,
      "learning_rate": 5.631067961165049e-06,
      "loss": 0.7713,
      "mean_token_accuracy": 0.7880979612469673,
      "num_tokens": 2318852.0,
      "step": 30
    },
    {
      "entropy": 1.0128332734107972,
      "epoch": 0.058415480102227094,
      "grad_norm": 0.5948992371559143,
      "learning_rate": 7.572815533980583e-06,
      "loss": 0.7764,
      "mean_token_accuracy": 0.7841672986745835,
      "num_tokens": 3084308.0,
      "step": 40
    },
    {
      "entropy": 1.012282955646515,
      "epoch": 0.07301935012778386,
      "grad_norm": 0.6310043931007385,
      "learning_rate": 9.514563106796117e-06,
      "loss": 0.7564,
      "mean_token_accuracy": 0.7881451666355133,
      "num_tokens": 3834841.0,
      "step": 50
    },
    {
      "entropy": 1.0291071116924286,
      "epoch": 0.08762322015334063,
      "grad_norm": 0.6180324554443359,
      "learning_rate": 1.145631067961165e-05,
      "loss": 0.7502,
      "mean_token_accuracy": 0.7902129292488098,
      "num_tokens": 4596954.0,
      "step": 60
    },
    {
      "entropy": 1.0330789506435394,
      "epoch": 0.1022270901788974,
      "grad_norm": 0.6203945279121399,
      "learning_rate": 1.3398058252427187e-05,
      "loss": 0.7596,
      "mean_token_accuracy": 0.7884851574897767,
      "num_tokens": 5367587.0,
      "step": 70
    },
    {
      "entropy": 1.0233020395040513,
      "epoch": 0.11683096020445419,
      "grad_norm": 0.5948684811592102,
      "learning_rate": 1.533980582524272e-05,
      "loss": 0.7538,
      "mean_token_accuracy": 0.788135002553463,
      "num_tokens": 6132539.0,
      "step": 80
    },
    {
      "entropy": 1.022057643532753,
      "epoch": 0.13143483023001096,
      "grad_norm": 0.6073455810546875,
      "learning_rate": 1.7281553398058253e-05,
      "loss": 0.7418,
      "mean_token_accuracy": 0.7916493251919746,
      "num_tokens": 6910021.0,
      "step": 90
    },
    {
      "entropy": 1.0418485432863236,
      "epoch": 0.14603870025556773,
      "grad_norm": 0.56880784034729,
      "learning_rate": 1.922330097087379e-05,
      "loss": 0.7686,
      "mean_token_accuracy": 0.7854298010468483,
      "num_tokens": 7672070.0,
      "step": 100
    },
    {
      "entropy": 1.0249923586845398,
      "epoch": 0.1606425702811245,
      "grad_norm": 0.6060709357261658,
      "learning_rate": 1.9999839020034848e-05,
      "loss": 0.7376,
      "mean_token_accuracy": 0.7922860443592071,
      "num_tokens": 8437851.0,
      "step": 110
    },
    {
      "entropy": 1.0423746079206466,
      "epoch": 0.17524644030668127,
      "grad_norm": 0.5950225591659546,
      "learning_rate": 1.9998855272350457e-05,
      "loss": 0.75,
      "mean_token_accuracy": 0.7883219286799431,
      "num_tokens": 9220959.0,
      "step": 120
    },
    {
      "entropy": 1.058434711396694,
      "epoch": 0.18985031033223804,
      "grad_norm": 0.5805737972259521,
      "learning_rate": 1.99969772981684e-05,
      "loss": 0.754,
      "mean_token_accuracy": 0.7883959293365479,
      "num_tokens": 9998137.0,
      "step": 130
    },
    {
      "entropy": 1.055925799906254,
      "epoch": 0.2044541803577948,
      "grad_norm": 0.5844792723655701,
      "learning_rate": 1.9994205265441328e-05,
      "loss": 0.7443,
      "mean_token_accuracy": 0.7903072774410248,
      "num_tokens": 10776402.0,
      "step": 140
    },
    {
      "entropy": 1.0465368673205375,
      "epoch": 0.21905805038335158,
      "grad_norm": 0.6346914172172546,
      "learning_rate": 1.9990539422080134e-05,
      "loss": 0.7423,
      "mean_token_accuracy": 0.7909073531627655,
      "num_tokens": 11537876.0,
      "step": 150
    },
    {
      "entropy": 1.0467117950320244,
      "epoch": 0.23366192040890837,
      "grad_norm": 0.555743396282196,
      "learning_rate": 1.9985980095931774e-05,
      "loss": 0.7426,
      "mean_token_accuracy": 0.7902399882674217,
      "num_tokens": 12313854.0,
      "step": 160
    },
    {
      "entropy": 1.0474098443984985,
      "epoch": 0.24826579043446514,
      "grad_norm": 0.6585262417793274,
      "learning_rate": 1.9980527694749952e-05,
      "loss": 0.7502,
      "mean_token_accuracy": 0.7889217540621758,
      "num_tokens": 13059556.0,
      "step": 170
    },
    {
      "entropy": 1.046202352643013,
      "epoch": 0.2628696604600219,
      "grad_norm": 0.5887473225593567,
      "learning_rate": 1.9974182706158646e-05,
      "loss": 0.7287,
      "mean_token_accuracy": 0.7938704118132591,
      "num_tokens": 13829911.0,
      "step": 180
    },
    {
      "entropy": 1.0543677926063537,
      "epoch": 0.27747353048557866,
      "grad_norm": 0.5907500982284546,
      "learning_rate": 1.996694569760851e-05,
      "loss": 0.7465,
      "mean_token_accuracy": 0.790279072523117,
      "num_tokens": 14595687.0,
      "step": 190
    },
    {
      "entropy": 1.059071497619152,
      "epoch": 0.29207740051113545,
      "grad_norm": 0.5858122706413269,
      "learning_rate": 1.995881731632611e-05,
      "loss": 0.7376,
      "mean_token_accuracy": 0.7931964993476868,
      "num_tokens": 15362920.0,
      "step": 200
    },
    {
      "entropy": 1.0532679110765457,
      "epoch": 0.3066812705366922,
      "grad_norm": 0.5991299152374268,
      "learning_rate": 1.9949798289256054e-05,
      "loss": 0.7578,
      "mean_token_accuracy": 0.786410291492939,
      "num_tokens": 16128067.0,
      "step": 210
    },
    {
      "entropy": 1.0485742643475533,
      "epoch": 0.321285140562249,
      "grad_norm": 0.5720910429954529,
      "learning_rate": 1.993988942299598e-05,
      "loss": 0.7404,
      "mean_token_accuracy": 0.7914015546441078,
      "num_tokens": 16911462.0,
      "step": 220
    },
    {
      "entropy": 1.0678794473409652,
      "epoch": 0.3358890105878058,
      "grad_norm": 0.6100424528121948,
      "learning_rate": 1.9929091603724404e-05,
      "loss": 0.7479,
      "mean_token_accuracy": 0.7890788897871971,
      "num_tokens": 17697397.0,
      "step": 230
    },
    {
      "entropy": 1.0550393044948578,
      "epoch": 0.35049288061336253,
      "grad_norm": 0.5960067510604858,
      "learning_rate": 1.9917405797121484e-05,
      "loss": 0.7518,
      "mean_token_accuracy": 0.7884634032845497,
      "num_tokens": 18465317.0,
      "step": 240
    },
    {
      "entropy": 1.0647842451930045,
      "epoch": 0.36509675063891933,
      "grad_norm": 0.61286461353302,
      "learning_rate": 1.990483304828264e-05,
      "loss": 0.7568,
      "mean_token_accuracy": 0.7871849820017814,
      "num_tokens": 19218071.0,
      "step": 250
    },
    {
      "entropy": 1.02879488915205,
      "epoch": 0.3797006206644761,
      "grad_norm": 0.5549123883247375,
      "learning_rate": 1.9891374481625112e-05,
      "loss": 0.7221,
      "mean_token_accuracy": 0.7956910878419876,
      "num_tokens": 19985551.0,
      "step": 260
    },
    {
      "entropy": 1.0511873975396155,
      "epoch": 0.39430449069003287,
      "grad_norm": 0.542924702167511,
      "learning_rate": 1.987703130078737e-05,
      "loss": 0.7303,
      "mean_token_accuracy": 0.7930067017674446,
      "num_tokens": 20759854.0,
      "step": 270
    },
    {
      "entropy": 1.046246202290058,
      "epoch": 0.4089083607155896,
      "grad_norm": 0.5551313161849976,
      "learning_rate": 1.986180478852149e-05,
      "loss": 0.7252,
      "mean_token_accuracy": 0.7956920295953751,
      "num_tokens": 21543604.0,
      "step": 280
    },
    {
      "entropy": 1.0610669255256653,
      "epoch": 0.4235122307411464,
      "grad_norm": 0.629205048084259,
      "learning_rate": 1.9845696306578433e-05,
      "loss": 0.7377,
      "mean_token_accuracy": 0.7924546420574188,
      "num_tokens": 22310802.0,
      "step": 290
    },
    {
      "entropy": 1.0590673327445983,
      "epoch": 0.43811610076670315,
      "grad_norm": 0.6042408347129822,
      "learning_rate": 1.9828707295586253e-05,
      "loss": 0.7356,
      "mean_token_accuracy": 0.7922392532229423,
      "num_tokens": 23088849.0,
      "step": 300
    },
    {
      "entropy": 1.059952473640442,
      "epoch": 0.45271997079225995,
      "grad_norm": 0.6013913750648499,
      "learning_rate": 1.981083927492125e-05,
      "loss": 0.7379,
      "mean_token_accuracy": 0.7928509280085564,
      "num_tokens": 23858610.0,
      "step": 310
    },
    {
      "entropy": 1.0460969746112823,
      "epoch": 0.46732384081781675,
      "grad_norm": 0.5834254622459412,
      "learning_rate": 1.9792093842572106e-05,
      "loss": 0.7335,
      "mean_token_accuracy": 0.7924273937940598,
      "num_tokens": 24590316.0,
      "step": 320
    },
    {
      "entropy": 1.0532741829752923,
      "epoch": 0.4819277108433735,
      "grad_norm": 0.5749837160110474,
      "learning_rate": 1.9772472674996962e-05,
      "loss": 0.746,
      "mean_token_accuracy": 0.789176419377327,
      "num_tokens": 25353089.0,
      "step": 330
    },
    {
      "entropy": 1.0500716269016266,
      "epoch": 0.4965315808689303,
      "grad_norm": 0.5746060609817505,
      "learning_rate": 1.975197752697349e-05,
      "loss": 0.7414,
      "mean_token_accuracy": 0.7907495066523552,
      "num_tokens": 26117777.0,
      "step": 340
    },
    {
      "entropy": 1.051976852118969,
      "epoch": 0.5111354508944871,
      "grad_norm": 0.5868931412696838,
      "learning_rate": 1.973061023144196e-05,
      "loss": 0.7256,
      "mean_token_accuracy": 0.7959051370620728,
      "num_tokens": 26885934.0,
      "step": 350
    },
    {
      "entropy": 1.078227588534355,
      "epoch": 0.5257393209200438,
      "grad_norm": 0.5909886360168457,
      "learning_rate": 1.9708372699341297e-05,
      "loss": 0.7568,
      "mean_token_accuracy": 0.7874275401234627,
      "num_tokens": 27648234.0,
      "step": 360
    },
    {
      "entropy": 1.0532238066196442,
      "epoch": 0.5403431909456006,
      "grad_norm": 0.5472894310951233,
      "learning_rate": 1.9685266919438208e-05,
      "loss": 0.7351,
      "mean_token_accuracy": 0.7938356637954712,
      "num_tokens": 28411776.0,
      "step": 370
    },
    {
      "entropy": 1.0588267534971236,
      "epoch": 0.5549470609711573,
      "grad_norm": 0.5929501056671143,
      "learning_rate": 1.9661294958149312e-05,
      "loss": 0.745,
      "mean_token_accuracy": 0.7911377623677254,
      "num_tokens": 29169804.0,
      "step": 380
    },
    {
      "entropy": 1.0762585639953612,
      "epoch": 0.5695509309967142,
      "grad_norm": 0.6019272804260254,
      "learning_rate": 1.963645895935632e-05,
      "loss": 0.7458,
      "mean_token_accuracy": 0.7898744881153107,
      "num_tokens": 29942397.0,
      "step": 390
    },
    {
      "entropy": 1.0671012222766876,
      "epoch": 0.5841548010222709,
      "grad_norm": 0.6131295561790466,
      "learning_rate": 1.9610761144214307e-05,
      "loss": 0.7547,
      "mean_token_accuracy": 0.7888392016291619,
      "num_tokens": 30737478.0,
      "step": 400
    },
    {
      "entropy": 1.0662592992186546,
      "epoch": 0.5987586710478277,
      "grad_norm": 0.579192578792572,
      "learning_rate": 1.9584203810953094e-05,
      "loss": 0.731,
      "mean_token_accuracy": 0.7930623203516006,
      "num_tokens": 31487239.0,
      "step": 410
    },
    {
      "entropy": 1.0672000512480735,
      "epoch": 0.6133625410733844,
      "grad_norm": 0.5550879836082458,
      "learning_rate": 1.9556789334671668e-05,
      "loss": 0.7395,
      "mean_token_accuracy": 0.792091254889965,
      "num_tokens": 32248396.0,
      "step": 420
    },
    {
      "entropy": 1.0656457453966142,
      "epoch": 0.6279664110989412,
      "grad_norm": 1.1255723237991333,
      "learning_rate": 1.9528520167125803e-05,
      "loss": 0.7395,
      "mean_token_accuracy": 0.7922166392207146,
      "num_tokens": 33011134.0,
      "step": 430
    },
    {
      "entropy": 1.0731883138418197,
      "epoch": 0.642570281124498,
      "grad_norm": 0.5882017016410828,
      "learning_rate": 1.9499398836508776e-05,
      "loss": 0.7493,
      "mean_token_accuracy": 0.7877946421504021,
      "num_tokens": 33787966.0,
      "step": 440
    },
    {
      "entropy": 1.0552370175719261,
      "epoch": 0.6571741511500547,
      "grad_norm": 0.6130959987640381,
      "learning_rate": 1.9469427947225267e-05,
      "loss": 0.7111,
      "mean_token_accuracy": 0.7995760783553123,
      "num_tokens": 34559268.0,
      "step": 450
    },
    {
      "entropy": 1.069392178952694,
      "epoch": 0.6717780211756116,
      "grad_norm": 0.5893939733505249,
      "learning_rate": 1.9438610179658447e-05,
      "loss": 0.7364,
      "mean_token_accuracy": 0.7937344864010811,
      "num_tokens": 35306495.0,
      "step": 460
    },
    {
      "entropy": 1.0607215002179147,
      "epoch": 0.6863818912011683,
      "grad_norm": 0.5958935022354126,
      "learning_rate": 1.9406948289930247e-05,
      "loss": 0.7358,
      "mean_token_accuracy": 0.7919678211212158,
      "num_tokens": 36064111.0,
      "step": 470
    },
    {
      "entropy": 1.079184153676033,
      "epoch": 0.7009857612267251,
      "grad_norm": 0.5986253619194031,
      "learning_rate": 1.9374445109654888e-05,
      "loss": 0.7421,
      "mean_token_accuracy": 0.7907416477799416,
      "num_tokens": 36832765.0,
      "step": 480
    },
    {
      "entropy": 1.0671533614397049,
      "epoch": 0.7155896312522818,
      "grad_norm": 0.5463415384292603,
      "learning_rate": 1.9341103545685637e-05,
      "loss": 0.7304,
      "mean_token_accuracy": 0.7941543251276016,
      "num_tokens": 37609612.0,
      "step": 490
    },
    {
      "entropy": 1.0800618380308151,
      "epoch": 0.7301935012778387,
      "grad_norm": 0.7118626832962036,
      "learning_rate": 1.930692657985482e-05,
      "loss": 0.7289,
      "mean_token_accuracy": 0.7952305421233177,
      "num_tokens": 38378623.0,
      "step": 500
    },
    {
      "entropy": 1.051339966058731,
      "epoch": 0.7447973713033954,
      "grad_norm": 0.5645574331283569,
      "learning_rate": 1.927191726870718e-05,
      "loss": 0.7091,
      "mean_token_accuracy": 0.7985311895608902,
      "num_tokens": 39146551.0,
      "step": 510
    },
    {
      "entropy": 1.0769214510917664,
      "epoch": 0.7594012413289521,
      "grad_norm": 0.6002981066703796,
      "learning_rate": 1.9236078743226502e-05,
      "loss": 0.7288,
      "mean_token_accuracy": 0.7942462310194969,
      "num_tokens": 39908848.0,
      "step": 520
    },
    {
      "entropy": 1.0461754769086837,
      "epoch": 0.7740051113545089,
      "grad_norm": 0.5525550246238708,
      "learning_rate": 1.919941420855559e-05,
      "loss": 0.7058,
      "mean_token_accuracy": 0.8000339075922966,
      "num_tokens": 40694755.0,
      "step": 530
    },
    {
      "entropy": 1.0753471747040748,
      "epoch": 0.7886089813800657,
      "grad_norm": 0.5900936722755432,
      "learning_rate": 1.916192694370965e-05,
      "loss": 0.7535,
      "mean_token_accuracy": 0.7890919044613838,
      "num_tokens": 41468815.0,
      "step": 540
    },
    {
      "entropy": 1.0729323342442512,
      "epoch": 0.8032128514056225,
      "grad_norm": 0.6021299958229065,
      "learning_rate": 1.912362030128302e-05,
      "loss": 0.7149,
      "mean_token_accuracy": 0.7978271067142486,
      "num_tokens": 42243180.0,
      "step": 550
    },
    {
      "entropy": 1.0811306938529015,
      "epoch": 0.8178167214311792,
      "grad_norm": 0.5898880958557129,
      "learning_rate": 1.9084497707149337e-05,
      "loss": 0.7383,
      "mean_token_accuracy": 0.7915405228734016,
      "num_tokens": 43002684.0,
      "step": 560
    },
    {
      "entropy": 1.0548559844493866,
      "epoch": 0.8324205914567361,
      "grad_norm": 0.6246165633201599,
      "learning_rate": 1.9044562660155158e-05,
      "loss": 0.7287,
      "mean_token_accuracy": 0.7933633595705032,
      "num_tokens": 43756816.0,
      "step": 570
    },
    {
      "entropy": 1.0436548948287965,
      "epoch": 0.8470244614822928,
      "grad_norm": 0.5454269051551819,
      "learning_rate": 1.900381873180704e-05,
      "loss": 0.7019,
      "mean_token_accuracy": 0.8000532567501069,
      "num_tokens": 44502793.0,
      "step": 580
    },
    {
      "entropy": 1.061595305800438,
      "epoch": 0.8616283315078496,
      "grad_norm": 0.5577073693275452,
      "learning_rate": 1.896226956595214e-05,
      "loss": 0.7347,
      "mean_token_accuracy": 0.7934001550078392,
      "num_tokens": 45260661.0,
      "step": 590
    },
    {
      "entropy": 1.0671459570527078,
      "epoch": 0.8762322015334063,
      "grad_norm": 0.5666953325271606,
      "learning_rate": 1.891991887845233e-05,
      "loss": 0.7157,
      "mean_token_accuracy": 0.7979460313916207,
      "num_tokens": 46040873.0,
      "step": 600
    },
    {
      "entropy": 1.0621700644493104,
      "epoch": 0.8908360715589632,
      "grad_norm": 0.5529576539993286,
      "learning_rate": 1.887677045685188e-05,
      "loss": 0.7252,
      "mean_token_accuracy": 0.7951329663395882,
      "num_tokens": 46818910.0,
      "step": 610
    },
    {
      "entropy": 1.0564261555671692,
      "epoch": 0.9054399415845199,
      "grad_norm": 0.6081776022911072,
      "learning_rate": 1.8832828160038717e-05,
      "loss": 0.7224,
      "mean_token_accuracy": 0.7945129945874214,
      "num_tokens": 47581801.0,
      "step": 620
    },
    {
      "entropy": 1.0867107123136521,
      "epoch": 0.9200438116100766,
      "grad_norm": 0.5485532283782959,
      "learning_rate": 1.8788095917899322e-05,
      "loss": 0.7397,
      "mean_token_accuracy": 0.7922059059143066,
      "num_tokens": 48353228.0,
      "step": 630
    },
    {
      "entropy": 1.0725250199437142,
      "epoch": 0.9346476816356335,
      "grad_norm": 0.5478349924087524,
      "learning_rate": 1.8742577730967275e-05,
      "loss": 0.7282,
      "mean_token_accuracy": 0.7937524914741516,
      "num_tokens": 49128771.0,
      "step": 640
    },
    {
      "entropy": 1.065740318596363,
      "epoch": 0.9492515516611902,
      "grad_norm": 0.5200572609901428,
      "learning_rate": 1.8696277670065453e-05,
      "loss": 0.7158,
      "mean_token_accuracy": 0.7966950073838234,
      "num_tokens": 49898864.0,
      "step": 650
    },
    {
      "entropy": 1.0765223398804664,
      "epoch": 0.963855421686747,
      "grad_norm": 0.5623005032539368,
      "learning_rate": 1.8649199875942e-05,
      "loss": 0.7316,
      "mean_token_accuracy": 0.79328583329916,
      "num_tokens": 50667692.0,
      "step": 660
    },
    {
      "entropy": 1.0664428249001503,
      "epoch": 0.9784592917123037,
      "grad_norm": 0.5457090735435486,
      "learning_rate": 1.860134855889997e-05,
      "loss": 0.7192,
      "mean_token_accuracy": 0.7951617255806923,
      "num_tokens": 51439436.0,
      "step": 670
    },
    {
      "entropy": 1.054172757267952,
      "epoch": 0.9930631617378606,
      "grad_norm": 0.558342456817627,
      "learning_rate": 1.8552727998420815e-05,
      "loss": 0.7284,
      "mean_token_accuracy": 0.795482975244522,
      "num_tokens": 52216680.0,
      "step": 680
    },
    {
      "entropy": 1.0329305483744695,
      "epoch": 1.0073019350127783,
      "grad_norm": 0.519607424736023,
      "learning_rate": 1.850334254278164e-05,
      "loss": 0.6312,
      "mean_token_accuracy": 0.8187721814864721,
      "num_tokens": 52953330.0,
      "step": 690
    },
    {
      "entropy": 0.8502921864390374,
      "epoch": 1.0219058050383352,
      "grad_norm": 0.6058902740478516,
      "learning_rate": 1.845319660866635e-05,
      "loss": 0.5259,
      "mean_token_accuracy": 0.8443149983882904,
      "num_tokens": 53713129.0,
      "step": 700
    },
    {
      "entropy": 0.8117894425988197,
      "epoch": 1.036509675063892,
      "grad_norm": 0.5925081968307495,
      "learning_rate": 1.8402294680770607e-05,
      "loss": 0.5039,
      "mean_token_accuracy": 0.8498010948300362,
      "num_tokens": 54480149.0,
      "step": 710
    },
    {
      "entropy": 0.7792180389165878,
      "epoch": 1.0511135450894487,
      "grad_norm": 0.637313961982727,
      "learning_rate": 1.8350641311400813e-05,
      "loss": 0.4964,
      "mean_token_accuracy": 0.851611290872097,
      "num_tokens": 55232208.0,
      "step": 720
    },
    {
      "entropy": 0.8110285863280297,
      "epoch": 1.0657174151150055,
      "grad_norm": 0.6269355416297913,
      "learning_rate": 1.8298241120066923e-05,
      "loss": 0.5102,
      "mean_token_accuracy": 0.8473430201411247,
      "num_tokens": 56017632.0,
      "step": 730
    },
    {
      "entropy": 0.805873404443264,
      "epoch": 1.0803212851405624,
      "grad_norm": 0.6263405084609985,
      "learning_rate": 1.8245098793069353e-05,
      "loss": 0.4989,
      "mean_token_accuracy": 0.8503431648015976,
      "num_tokens": 56794094.0,
      "step": 740
    },
    {
      "entropy": 0.7984029710292816,
      "epoch": 1.094925155166119,
      "grad_norm": 0.6596930623054504,
      "learning_rate": 1.819121908307985e-05,
      "loss": 0.5029,
      "mean_token_accuracy": 0.8493917599320412,
      "num_tokens": 57569065.0,
      "step": 750
    },
    {
      "entropy": 0.8163655236363411,
      "epoch": 1.1095290251916758,
      "grad_norm": 0.6013880372047424,
      "learning_rate": 1.813660680871645e-05,
      "loss": 0.508,
      "mean_token_accuracy": 0.8476730227470398,
      "num_tokens": 58342947.0,
      "step": 760
    },
    {
      "entropy": 0.8022693067789077,
      "epoch": 1.1241328952172325,
      "grad_norm": 0.6172059774398804,
      "learning_rate": 1.8081266854112536e-05,
      "loss": 0.5066,
      "mean_token_accuracy": 0.847652480006218,
      "num_tokens": 59095076.0,
      "step": 770
    },
    {
      "entropy": 0.8082443997263908,
      "epoch": 1.1387367652427893,
      "grad_norm": 0.5798039436340332,
      "learning_rate": 1.8025204168480036e-05,
      "loss": 0.5084,
      "mean_token_accuracy": 0.8473975166678429,
      "num_tokens": 59855251.0,
      "step": 780
    },
    {
      "entropy": 0.8160747662186623,
      "epoch": 1.1533406352683462,
      "grad_norm": 0.6323524713516235,
      "learning_rate": 1.7968423765666805e-05,
      "loss": 0.5181,
      "mean_token_accuracy": 0.8460124984383584,
      "num_tokens": 60616729.0,
      "step": 790
    },
    {
      "entropy": 0.8220678076148034,
      "epoch": 1.1679445052939028,
      "grad_norm": 0.5704360008239746,
      "learning_rate": 1.7910930723708206e-05,
      "loss": 0.4989,
      "mean_token_accuracy": 0.8500168919563293,
      "num_tokens": 61391607.0,
      "step": 800
    },
    {
      "entropy": 0.8171453341841698,
      "epoch": 1.1825483753194597,
      "grad_norm": 0.6301026940345764,
      "learning_rate": 1.7852730184372996e-05,
      "loss": 0.5094,
      "mean_token_accuracy": 0.847735871374607,
      "num_tokens": 62176326.0,
      "step": 810
    },
    {
      "entropy": 0.8387730494141579,
      "epoch": 1.1971522453450165,
      "grad_norm": 0.6920040845870972,
      "learning_rate": 1.779382735270345e-05,
      "loss": 0.528,
      "mean_token_accuracy": 0.8430045962333679,
      "num_tokens": 62936900.0,
      "step": 820
    },
    {
      "entropy": 0.8173261970281601,
      "epoch": 1.2117561153705732,
      "grad_norm": 0.6174663305282593,
      "learning_rate": 1.773422749654988e-05,
      "loss": 0.505,
      "mean_token_accuracy": 0.849335603415966,
      "num_tokens": 63714937.0,
      "step": 830
    },
    {
      "entropy": 0.8201940849423408,
      "epoch": 1.22635998539613,
      "grad_norm": 0.6521441340446472,
      "learning_rate": 1.7673935946099515e-05,
      "loss": 0.5169,
      "mean_token_accuracy": 0.8452720895409584,
      "num_tokens": 64484592.0,
      "step": 840
    },
    {
      "entropy": 0.8129187062382698,
      "epoch": 1.2409638554216866,
      "grad_norm": 0.6480047106742859,
      "learning_rate": 1.7612958093399793e-05,
      "loss": 0.5101,
      "mean_token_accuracy": 0.847150382399559,
      "num_tokens": 65242523.0,
      "step": 850
    },
    {
      "entropy": 0.797595490515232,
      "epoch": 1.2555677254472435,
      "grad_norm": 0.6072301268577576,
      "learning_rate": 1.7551299391876147e-05,
      "loss": 0.4954,
      "mean_token_accuracy": 0.8505616560578346,
      "num_tokens": 66010969.0,
      "step": 860
    },
    {
      "entropy": 0.8110659316182136,
      "epoch": 1.2701715954728003,
      "grad_norm": 0.6669343709945679,
      "learning_rate": 1.7488965355844293e-05,
      "loss": 0.5132,
      "mean_token_accuracy": 0.8465436458587646,
      "num_tokens": 66767531.0,
      "step": 870
    },
    {
      "entropy": 0.8244558870792389,
      "epoch": 1.284775465498357,
      "grad_norm": 0.6384320855140686,
      "learning_rate": 1.742596156001705e-05,
      "loss": 0.5276,
      "mean_token_accuracy": 0.8408863857388497,
      "num_tokens": 67498692.0,
      "step": 880
    },
    {
      "entropy": 0.8114714965224266,
      "epoch": 1.2993793355239138,
      "grad_norm": 0.5711145401000977,
      "learning_rate": 1.73622936390058e-05,
      "loss": 0.5139,
      "mean_token_accuracy": 0.8463606715202332,
      "num_tokens": 68285940.0,
      "step": 890
    },
    {
      "entropy": 0.8195077747106552,
      "epoch": 1.3139832055494707,
      "grad_norm": 0.6267218589782715,
      "learning_rate": 1.7297967286816553e-05,
      "loss": 0.5154,
      "mean_token_accuracy": 0.8449677467346192,
      "num_tokens": 69039571.0,
      "step": 900
    },
    {
      "entropy": 0.8230465367436409,
      "epoch": 1.3285870755750273,
      "grad_norm": 0.6753976345062256,
      "learning_rate": 1.723298825634072e-05,
      "loss": 0.5357,
      "mean_token_accuracy": 0.8401415064930916,
      "num_tokens": 69793108.0,
      "step": 910
    },
    {
      "entropy": 0.8096449464559555,
      "epoch": 1.3431909456005842,
      "grad_norm": 0.6720008850097656,
      "learning_rate": 1.716736235884062e-05,
      "loss": 0.5098,
      "mean_token_accuracy": 0.8470739260315895,
      "num_tokens": 70568130.0,
      "step": 920
    },
    {
      "entropy": 0.8186160072684288,
      "epoch": 1.357794815626141,
      "grad_norm": 0.5903974771499634,
      "learning_rate": 1.7101095463429748e-05,
      "loss": 0.5198,
      "mean_token_accuracy": 0.8454992339015007,
      "num_tokens": 71349507.0,
      "step": 930
    },
    {
      "entropy": 0.8150006383657455,
      "epoch": 1.3723986856516976,
      "grad_norm": 0.5898988842964172,
      "learning_rate": 1.7034193496547903e-05,
      "loss": 0.5071,
      "mean_token_accuracy": 0.8491646945476532,
      "num_tokens": 72135471.0,
      "step": 940
    },
    {
      "entropy": 0.808189807832241,
      "epoch": 1.3870025556772545,
      "grad_norm": 0.5973511934280396,
      "learning_rate": 1.6966662441431157e-05,
      "loss": 0.5321,
      "mean_token_accuracy": 0.8418431639671325,
      "num_tokens": 72886499.0,
      "step": 950
    },
    {
      "entropy": 0.8094048380851746,
      "epoch": 1.4016064257028114,
      "grad_norm": 0.6130071878433228,
      "learning_rate": 1.6898508337576754e-05,
      "loss": 0.519,
      "mean_token_accuracy": 0.8449132606387139,
      "num_tokens": 73639876.0,
      "step": 960
    },
    {
      "entropy": 0.8094017982482911,
      "epoch": 1.416210295728368,
      "grad_norm": 0.6462528705596924,
      "learning_rate": 1.6829737280203e-05,
      "loss": 0.5158,
      "mean_token_accuracy": 0.8453432083129883,
      "num_tokens": 74401459.0,
      "step": 970
    },
    {
      "entropy": 0.8205527007579804,
      "epoch": 1.4308141657539248,
      "grad_norm": 0.6410425901412964,
      "learning_rate": 1.676035541970411e-05,
      "loss": 0.5366,
      "mean_token_accuracy": 0.8401491552591324,
      "num_tokens": 75180262.0,
      "step": 980
    },
    {
      "entropy": 0.8101997837424278,
      "epoch": 1.4454180357794817,
      "grad_norm": 0.6770428419113159,
      "learning_rate": 1.669036896110021e-05,
      "loss": 0.5278,
      "mean_token_accuracy": 0.8413688018918037,
      "num_tokens": 75935757.0,
      "step": 990
    },
    {
      "entropy": 0.8236281529068947,
      "epoch": 1.4600219058050383,
      "grad_norm": 0.5985122323036194,
      "learning_rate": 1.6619784163482372e-05,
      "loss": 0.5169,
      "mean_token_accuracy": 0.8454629138112069,
      "num_tokens": 76677214.0,
      "step": 1000
    },
    {
      "entropy": 0.8149201571941376,
      "epoch": 1.4746257758305952,
      "grad_norm": 0.6033229231834412,
      "learning_rate": 1.6548607339452853e-05,
      "loss": 0.5157,
      "mean_token_accuracy": 0.8454660639166832,
      "num_tokens": 77429839.0,
      "step": 1010
    },
    {
      "entropy": 0.8358469530940056,
      "epoch": 1.4892296458561518,
      "grad_norm": 0.6250749826431274,
      "learning_rate": 1.6476844854560537e-05,
      "loss": 0.5274,
      "mean_token_accuracy": 0.8430205255746841,
      "num_tokens": 78198665.0,
      "step": 1020
    },
    {
      "entropy": 0.8159400016069412,
      "epoch": 1.5038335158817087,
      "grad_norm": 0.6094375848770142,
      "learning_rate": 1.640450312673166e-05,
      "loss": 0.5237,
      "mean_token_accuracy": 0.8430356681346893,
      "num_tokens": 78963745.0,
      "step": 1030
    },
    {
      "entropy": 0.7948550447821617,
      "epoch": 1.5184373859072653,
      "grad_norm": 0.6190559267997742,
      "learning_rate": 1.6331588625695823e-05,
      "loss": 0.4984,
      "mean_token_accuracy": 0.850845368206501,
      "num_tokens": 79742482.0,
      "step": 1040
    },
    {
      "entropy": 0.8203089535236359,
      "epoch": 1.5330412559328221,
      "grad_norm": 0.627266526222229,
      "learning_rate": 1.6258107872407376e-05,
      "loss": 0.5222,
      "mean_token_accuracy": 0.8443042784929276,
      "num_tokens": 80513082.0,
      "step": 1050
    },
    {
      "entropy": 0.8034705385565758,
      "epoch": 1.547645125958379,
      "grad_norm": 0.5822563767433167,
      "learning_rate": 1.6184067438462268e-05,
      "loss": 0.5108,
      "mean_token_accuracy": 0.8467739015817642,
      "num_tokens": 81287042.0,
      "step": 1060
    },
    {
      "entropy": 0.7996939823031426,
      "epoch": 1.5622489959839356,
      "grad_norm": 0.8322076201438904,
      "learning_rate": 1.6109473945510277e-05,
      "loss": 0.5059,
      "mean_token_accuracy": 0.8484288737177849,
      "num_tokens": 82062918.0,
      "step": 1070
    },
    {
      "entropy": 0.8242404267191887,
      "epoch": 1.5768528660094925,
      "grad_norm": 0.6267576217651367,
      "learning_rate": 1.6034334064662868e-05,
      "loss": 0.5236,
      "mean_token_accuracy": 0.8428681313991546,
      "num_tokens": 82835236.0,
      "step": 1080
    },
    {
      "entropy": 0.8226122260093689,
      "epoch": 1.5914567360350493,
      "grad_norm": 0.578869104385376,
      "learning_rate": 1.595865451589654e-05,
      "loss": 0.522,
      "mean_token_accuracy": 0.8434418380260468,
      "num_tokens": 83610617.0,
      "step": 1090
    },
    {
      "entropy": 0.8142560958862305,
      "epoch": 1.606060606060606,
      "grad_norm": 0.652740478515625,
      "learning_rate": 1.5882442067451856e-05,
      "loss": 0.52,
      "mean_token_accuracy": 0.8449100449681282,
      "num_tokens": 84379036.0,
      "step": 1100
    },
    {
      "entropy": 0.8149377256631851,
      "epoch": 1.6206644760861628,
      "grad_norm": 0.5886592864990234,
      "learning_rate": 1.5805703535228137e-05,
      "loss": 0.5318,
      "mean_token_accuracy": 0.8409554123878479,
      "num_tokens": 85126712.0,
      "step": 1110
    },
    {
      "entropy": 0.8216987118124962,
      "epoch": 1.6352683461117197,
      "grad_norm": 0.625464916229248,
      "learning_rate": 1.5728445782173896e-05,
      "loss": 0.5213,
      "mean_token_accuracy": 0.8435990065336227,
      "num_tokens": 85888279.0,
      "step": 1120
    },
    {
      "entropy": 0.8107628434896469,
      "epoch": 1.6498722161372763,
      "grad_norm": 0.6127883195877075,
      "learning_rate": 1.565067571767306e-05,
      "loss": 0.5307,
      "mean_token_accuracy": 0.8404728651046753,
      "num_tokens": 86651718.0,
      "step": 1130
    },
    {
      "entropy": 0.8122030794620514,
      "epoch": 1.6644760861628332,
      "grad_norm": 0.6447589993476868,
      "learning_rate": 1.557240029692705e-05,
      "loss": 0.5288,
      "mean_token_accuracy": 0.8413975268602372,
      "num_tokens": 87418899.0,
      "step": 1140
    },
    {
      "entropy": 0.8063643991947174,
      "epoch": 1.67907995618839,
      "grad_norm": 0.6214607357978821,
      "learning_rate": 1.5493626520332758e-05,
      "loss": 0.5046,
      "mean_token_accuracy": 0.8487621054053307,
      "num_tokens": 88197755.0,
      "step": 1150
    },
    {
      "entropy": 0.811661048233509,
      "epoch": 1.6936838262139466,
      "grad_norm": 0.587363600730896,
      "learning_rate": 1.5414361432856475e-05,
      "loss": 0.5232,
      "mean_token_accuracy": 0.8434179335832596,
      "num_tokens": 88960882.0,
      "step": 1160
    },
    {
      "entropy": 0.824562780559063,
      "epoch": 1.7082876962395035,
      "grad_norm": 0.5840896964073181,
      "learning_rate": 1.533461212340384e-05,
      "loss": 0.5295,
      "mean_token_accuracy": 0.8425872087478637,
      "num_tokens": 89730659.0,
      "step": 1170
    },
    {
      "entropy": 0.8084111362695694,
      "epoch": 1.7228915662650603,
      "grad_norm": 0.6445785760879517,
      "learning_rate": 1.5254385724185872e-05,
      "loss": 0.5189,
      "mean_token_accuracy": 0.8443948805332184,
      "num_tokens": 90519374.0,
      "step": 1180
    },
    {
      "entropy": 0.802851003408432,
      "epoch": 1.737495436290617,
      "grad_norm": 0.5873962044715881,
      "learning_rate": 1.5173689410081091e-05,
      "loss": 0.5215,
      "mean_token_accuracy": 0.8436427339911461,
      "num_tokens": 91280437.0,
      "step": 1190
    },
    {
      "entropy": 0.8080698132514954,
      "epoch": 1.7520993063161738,
      "grad_norm": 0.594432532787323,
      "learning_rate": 1.5092530397993877e-05,
      "loss": 0.519,
      "mean_token_accuracy": 0.8449363052845001,
      "num_tokens": 92051212.0,
      "step": 1200
    },
    {
      "entropy": 0.8046671271324157,
      "epoch": 1.7667031763417307,
      "grad_norm": 0.6073573231697083,
      "learning_rate": 1.5010915946209013e-05,
      "loss": 0.5207,
      "mean_token_accuracy": 0.8437707021832466,
      "num_tokens": 92824656.0,
      "step": 1210
    },
    {
      "entropy": 0.799671696126461,
      "epoch": 1.7813070463672873,
      "grad_norm": 0.5851039886474609,
      "learning_rate": 1.492885335374258e-05,
      "loss": 0.5171,
      "mean_token_accuracy": 0.845556665956974,
      "num_tokens": 93601248.0,
      "step": 1220
    },
    {
      "entropy": 0.7995420083403587,
      "epoch": 1.795910916392844,
      "grad_norm": 0.6009591221809387,
      "learning_rate": 1.4846349959689166e-05,
      "loss": 0.5064,
      "mean_token_accuracy": 0.8478008210659027,
      "num_tokens": 94367653.0,
      "step": 1230
    },
    {
      "entropy": 0.8032964497804642,
      "epoch": 1.810514786418401,
      "grad_norm": 0.6265828013420105,
      "learning_rate": 1.4763413142565524e-05,
      "loss": 0.5126,
      "mean_token_accuracy": 0.8462978065013885,
      "num_tokens": 95134532.0,
      "step": 1240
    },
    {
      "entropy": 0.8157682687044143,
      "epoch": 1.8251186564439577,
      "grad_norm": 0.6303804516792297,
      "learning_rate": 1.468005031965068e-05,
      "loss": 0.5285,
      "mean_token_accuracy": 0.8411597847938538,
      "num_tokens": 95888124.0,
      "step": 1250
    },
    {
      "entropy": 0.8242416352033615,
      "epoch": 1.8397225264695143,
      "grad_norm": 0.6679196953773499,
      "learning_rate": 1.4596268946322587e-05,
      "loss": 0.5277,
      "mean_token_accuracy": 0.8423770368099213,
      "num_tokens": 96640919.0,
      "step": 1260
    },
    {
      "entropy": 0.8027110368013382,
      "epoch": 1.8543263964950711,
      "grad_norm": 0.6514571309089661,
      "learning_rate": 1.4512076515391375e-05,
      "loss": 0.5144,
      "mean_token_accuracy": 0.845685575902462,
      "num_tokens": 97402349.0,
      "step": 1270
    },
    {
      "entropy": 0.8184534996747971,
      "epoch": 1.868930266520628,
      "grad_norm": 0.6248390078544617,
      "learning_rate": 1.4427480556429237e-05,
      "loss": 0.5232,
      "mean_token_accuracy": 0.8441895946860314,
      "num_tokens": 98195965.0,
      "step": 1280
    },
    {
      "entropy": 0.8160968586802483,
      "epoch": 1.8835341365461846,
      "grad_norm": 0.6021704077720642,
      "learning_rate": 1.4342488635097044e-05,
      "loss": 0.5149,
      "mean_token_accuracy": 0.8458479061722756,
      "num_tokens": 98968374.0,
      "step": 1290
    },
    {
      "entropy": 0.8257978558540344,
      "epoch": 1.8981380065717415,
      "grad_norm": 0.5873125195503235,
      "learning_rate": 1.425710835246773e-05,
      "loss": 0.5265,
      "mean_token_accuracy": 0.8433559283614158,
      "num_tokens": 99729725.0,
      "step": 1300
    },
    {
      "entropy": 0.8362466841936111,
      "epoch": 1.9127418765972983,
      "grad_norm": 0.6376807689666748,
      "learning_rate": 1.4171347344346494e-05,
      "loss": 0.5343,
      "mean_token_accuracy": 0.8405352383852005,
      "num_tokens": 100511457.0,
      "step": 1310
    },
    {
      "entropy": 0.806542806327343,
      "epoch": 1.927345746622855,
      "grad_norm": 0.6307621002197266,
      "learning_rate": 1.4085213280587916e-05,
      "loss": 0.508,
      "mean_token_accuracy": 0.8474292501807212,
      "num_tokens": 101294648.0,
      "step": 1320
    },
    {
      "entropy": 0.828097864985466,
      "epoch": 1.9419496166484118,
      "grad_norm": 0.6401568055152893,
      "learning_rate": 1.3998713864410029e-05,
      "loss": 0.523,
      "mean_token_accuracy": 0.8436363622546196,
      "num_tokens": 102057660.0,
      "step": 1330
    },
    {
      "entropy": 0.8092041403055191,
      "epoch": 1.9565534866739687,
      "grad_norm": 0.5988907217979431,
      "learning_rate": 1.3911856831705372e-05,
      "loss": 0.5199,
      "mean_token_accuracy": 0.8438414841890335,
      "num_tokens": 102826372.0,
      "step": 1340
    },
    {
      "entropy": 0.8254723891615867,
      "epoch": 1.9711573566995253,
      "grad_norm": 0.6260574460029602,
      "learning_rate": 1.3824649950349173e-05,
      "loss": 0.5307,
      "mean_token_accuracy": 0.8420073762536049,
      "num_tokens": 103585161.0,
      "step": 1350
    },
    {
      "entropy": 0.8164636805653572,
      "epoch": 1.9857612267250822,
      "grad_norm": 0.6160163879394531,
      "learning_rate": 1.373710101950464e-05,
      "loss": 0.5213,
      "mean_token_accuracy": 0.8451445356011391,
      "num_tokens": 104380569.0,
      "step": 1360
    },
    {
      "entropy": 0.8088174370618967,
      "epoch": 2.0,
      "grad_norm": 0.6965954303741455,
      "learning_rate": 1.3649217868925435e-05,
      "loss": 0.5143,
      "mean_token_accuracy": 0.8466584361516513,
      "num_tokens": 105148277.0,
      "step": 1370
    }
  ],
  "logging_steps": 10,
  "max_steps": 3425,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.54747959995138e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}