{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 2551,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.40051139052957296,
      "epoch": 0.0039204155640497895,
      "grad_norm": 30.875,
      "learning_rate": 7.058823529411766e-07,
      "loss": 1.316241455078125,
      "mean_token_accuracy": 0.7522591009736062,
      "num_tokens": 130734.0,
      "step": 10
    },
    {
      "entropy": 0.39800101406872274,
      "epoch": 0.007840831128099579,
      "grad_norm": 28.875,
      "learning_rate": 1.4901960784313726e-06,
      "loss": 1.2585247993469237,
      "mean_token_accuracy": 0.7558266721665859,
      "num_tokens": 267547.0,
      "step": 20
    },
    {
      "entropy": 0.4350921854376793,
      "epoch": 0.011761246692149369,
      "grad_norm": 19.25,
      "learning_rate": 2.274509803921569e-06,
      "loss": 1.2427661895751954,
      "mean_token_accuracy": 0.7490861490368843,
      "num_tokens": 399201.0,
      "step": 30
    },
    {
      "entropy": 0.4768921874463558,
      "epoch": 0.015681662256199158,
      "grad_norm": 12.4375,
      "learning_rate": 3.058823529411765e-06,
      "loss": 1.1601675987243651,
      "mean_token_accuracy": 0.7548606485128403,
      "num_tokens": 527876.0,
      "step": 40
    },
    {
      "entropy": 0.5394699670374393,
      "epoch": 0.019602077820248948,
      "grad_norm": 10.125,
      "learning_rate": 3.843137254901962e-06,
      "loss": 1.0834686279296875,
      "mean_token_accuracy": 0.7578018061816693,
      "num_tokens": 658209.0,
      "step": 50
    },
    {
      "entropy": 0.5797790199518204,
      "epoch": 0.023522493384298737,
      "grad_norm": 6.28125,
      "learning_rate": 4.627450980392157e-06,
      "loss": 0.9067214965820313,
      "mean_token_accuracy": 0.7733228243887424,
      "num_tokens": 791743.0,
      "step": 60
    },
    {
      "entropy": 0.6502218697220087,
      "epoch": 0.027442908948348527,
      "grad_norm": 5.0,
      "learning_rate": 5.411764705882353e-06,
      "loss": 0.8242039680480957,
      "mean_token_accuracy": 0.7773927368223668,
      "num_tokens": 922196.0,
      "step": 70
    },
    {
      "entropy": 0.6219119645655156,
      "epoch": 0.031363324512398316,
      "grad_norm": 2.953125,
      "learning_rate": 6.19607843137255e-06,
      "loss": 0.7194723129272461,
      "mean_token_accuracy": 0.7994848638772964,
      "num_tokens": 1053869.0,
      "step": 80
    },
    {
      "entropy": 0.6228452675044537,
      "epoch": 0.035283740076448106,
      "grad_norm": 2.796875,
      "learning_rate": 6.9803921568627454e-06,
      "loss": 0.6950749397277832,
      "mean_token_accuracy": 0.8034804306924344,
      "num_tokens": 1179999.0,
      "step": 90
    },
    {
      "entropy": 0.6416549410670995,
      "epoch": 0.039204155640497895,
      "grad_norm": 2.890625,
      "learning_rate": 7.764705882352941e-06,
      "loss": 0.6787714958190918,
      "mean_token_accuracy": 0.8052642524242402,
      "num_tokens": 1313342.0,
      "step": 100
    },
    {
      "entropy": 0.6458045944571496,
      "epoch": 0.043124571204547685,
      "grad_norm": 2.078125,
      "learning_rate": 8.549019607843138e-06,
      "loss": 0.6732481956481934,
      "mean_token_accuracy": 0.803642563521862,
      "num_tokens": 1444756.0,
      "step": 110
    },
    {
      "entropy": 0.634775460511446,
      "epoch": 0.047044986768597474,
      "grad_norm": 1.9375,
      "learning_rate": 9.333333333333334e-06,
      "loss": 0.6695634841918945,
      "mean_token_accuracy": 0.8045761771500111,
      "num_tokens": 1577527.0,
      "step": 120
    },
    {
      "entropy": 0.6200754787772894,
      "epoch": 0.050965402332647264,
      "grad_norm": 1.9765625,
      "learning_rate": 1.011764705882353e-05,
      "loss": 0.6515310764312744,
      "mean_token_accuracy": 0.8107773900032044,
      "num_tokens": 1712989.0,
      "step": 130
    },
    {
      "entropy": 0.6283363737165928,
      "epoch": 0.05488581789669705,
      "grad_norm": 1.859375,
      "learning_rate": 1.0901960784313726e-05,
      "loss": 0.6463139057159424,
      "mean_token_accuracy": 0.8106771603226661,
      "num_tokens": 1842532.0,
      "step": 140
    },
    {
      "entropy": 0.6225632183253765,
      "epoch": 0.058806233460746836,
      "grad_norm": 1.90625,
      "learning_rate": 1.1686274509803922e-05,
      "loss": 0.6356926918029785,
      "mean_token_accuracy": 0.8131030358374118,
      "num_tokens": 1973776.0,
      "step": 150
    },
    {
      "entropy": 0.5999462876468897,
      "epoch": 0.06272664902479663,
      "grad_norm": 1.9375,
      "learning_rate": 1.2470588235294119e-05,
      "loss": 0.6088289737701416,
      "mean_token_accuracy": 0.8192408412694931,
      "num_tokens": 2106467.0,
      "step": 160
    },
    {
      "entropy": 0.6009281285107135,
      "epoch": 0.06664706458884642,
      "grad_norm": 1.96875,
      "learning_rate": 1.3254901960784314e-05,
      "loss": 0.6100459575653077,
      "mean_token_accuracy": 0.8174788504838943,
      "num_tokens": 2243425.0,
      "step": 170
    },
    {
      "entropy": 0.6414772845804692,
      "epoch": 0.07056748015289621,
      "grad_norm": 2.0625,
      "learning_rate": 1.403921568627451e-05,
      "loss": 0.6445958614349365,
      "mean_token_accuracy": 0.8082262150943279,
      "num_tokens": 2376987.0,
      "step": 180
    },
    {
      "entropy": 0.6355918630957603,
      "epoch": 0.074487895716946,
      "grad_norm": 1.921875,
      "learning_rate": 1.4823529411764707e-05,
      "loss": 0.6472668170928955,
      "mean_token_accuracy": 0.8071325562894345,
      "num_tokens": 2510548.0,
      "step": 190
    },
    {
      "entropy": 0.6236784052103758,
      "epoch": 0.07840831128099579,
      "grad_norm": 1.875,
      "learning_rate": 1.5607843137254904e-05,
      "loss": 0.6356116771697998,
      "mean_token_accuracy": 0.8135113798081874,
      "num_tokens": 2643661.0,
      "step": 200
    },
    {
      "entropy": 0.6072400402277708,
      "epoch": 0.08232872684504558,
      "grad_norm": 1.84375,
      "learning_rate": 1.63921568627451e-05,
      "loss": 0.6156791687011719,
      "mean_token_accuracy": 0.8167861931025981,
      "num_tokens": 2774529.0,
      "step": 210
    },
    {
      "entropy": 0.5829322092235089,
      "epoch": 0.08624914240909537,
      "grad_norm": 1.8203125,
      "learning_rate": 1.7176470588235293e-05,
      "loss": 0.5909051418304443,
      "mean_token_accuracy": 0.8222801245748996,
      "num_tokens": 2908058.0,
      "step": 220
    },
    {
      "entropy": 0.6099963016808033,
      "epoch": 0.09016955797314516,
      "grad_norm": 1.6875,
      "learning_rate": 1.796078431372549e-05,
      "loss": 0.6119081974029541,
      "mean_token_accuracy": 0.8179662935435772,
      "num_tokens": 3040096.0,
      "step": 230
    },
    {
      "entropy": 0.6330130327492952,
      "epoch": 0.09408997353719495,
      "grad_norm": 1.96875,
      "learning_rate": 1.8745098039215686e-05,
      "loss": 0.6453900814056397,
      "mean_token_accuracy": 0.8088094264268875,
      "num_tokens": 3171172.0,
      "step": 240
    },
    {
      "entropy": 0.6220353674143553,
      "epoch": 0.09801038910124474,
      "grad_norm": 1.9609375,
      "learning_rate": 1.9529411764705885e-05,
      "loss": 0.6278114318847656,
      "mean_token_accuracy": 0.8124852173030377,
      "num_tokens": 3304081.0,
      "step": 250
    },
    {
      "entropy": 0.6234287895262242,
      "epoch": 0.10193080466529453,
      "grad_norm": 1.8359375,
      "learning_rate": 1.996515679442509e-05,
      "loss": 0.6276582717895508,
      "mean_token_accuracy": 0.8146361976861953,
      "num_tokens": 3438063.0,
      "step": 260
    },
    {
      "entropy": 0.6037415701895952,
      "epoch": 0.10585122022934432,
      "grad_norm": 1.8125,
      "learning_rate": 1.9878048780487806e-05,
      "loss": 0.6147459506988525,
      "mean_token_accuracy": 0.8160658605396748,
      "num_tokens": 3572684.0,
      "step": 270
    },
    {
      "entropy": 0.6154889557510614,
      "epoch": 0.1097716357933941,
      "grad_norm": 1.828125,
      "learning_rate": 1.9790940766550523e-05,
      "loss": 0.620861291885376,
      "mean_token_accuracy": 0.8154690660536289,
      "num_tokens": 3706498.0,
      "step": 280
    },
    {
      "entropy": 0.5835129704326392,
      "epoch": 0.11369205135744388,
      "grad_norm": 1.828125,
      "learning_rate": 1.970383275261324e-05,
      "loss": 0.5846771240234375,
      "mean_token_accuracy": 0.8243817768990993,
      "num_tokens": 3837023.0,
      "step": 290
    },
    {
      "entropy": 0.5867879837751389,
      "epoch": 0.11761246692149367,
      "grad_norm": 1.828125,
      "learning_rate": 1.961672473867596e-05,
      "loss": 0.5936897277832032,
      "mean_token_accuracy": 0.8227211274206638,
      "num_tokens": 3966412.0,
      "step": 300
    },
    {
      "entropy": 0.6125645771622658,
      "epoch": 0.12153288248554346,
      "grad_norm": 1.6953125,
      "learning_rate": 1.9529616724738677e-05,
      "loss": 0.6198267459869384,
      "mean_token_accuracy": 0.8152951590716839,
      "num_tokens": 4096711.0,
      "step": 310
    },
    {
      "entropy": 0.610775976255536,
      "epoch": 0.12545329804959326,
      "grad_norm": 1.9296875,
      "learning_rate": 1.9442508710801397e-05,
      "loss": 0.6171989917755127,
      "mean_token_accuracy": 0.815869303047657,
      "num_tokens": 4228066.0,
      "step": 320
    },
    {
      "entropy": 0.6040709633380175,
      "epoch": 0.12937371361364305,
      "grad_norm": 1.9609375,
      "learning_rate": 1.9355400696864114e-05,
      "loss": 0.6103555202484131,
      "mean_token_accuracy": 0.8194533243775368,
      "num_tokens": 4357664.0,
      "step": 330
    },
    {
      "entropy": 0.6143804207444191,
      "epoch": 0.13329412917769284,
      "grad_norm": 1.8828125,
      "learning_rate": 1.926829268292683e-05,
      "loss": 0.6221739768981933,
      "mean_token_accuracy": 0.8153439976274968,
      "num_tokens": 4487688.0,
      "step": 340
    },
    {
      "entropy": 0.6031508389860392,
      "epoch": 0.13721454474174263,
      "grad_norm": 1.84375,
      "learning_rate": 1.9181184668989547e-05,
      "loss": 0.6023824214935303,
      "mean_token_accuracy": 0.8190992563962937,
      "num_tokens": 4619172.0,
      "step": 350
    },
    {
      "entropy": 0.6009325047954917,
      "epoch": 0.14113496030579242,
      "grad_norm": 1.8125,
      "learning_rate": 1.9094076655052267e-05,
      "loss": 0.6117077827453613,
      "mean_token_accuracy": 0.8167035676538944,
      "num_tokens": 4749078.0,
      "step": 360
    },
    {
      "entropy": 0.6013668723404407,
      "epoch": 0.1450553758698422,
      "grad_norm": 1.6875,
      "learning_rate": 1.9006968641114984e-05,
      "loss": 0.5986335277557373,
      "mean_token_accuracy": 0.818621464818716,
      "num_tokens": 4881903.0,
      "step": 370
    },
    {
      "entropy": 0.5910887397825718,
      "epoch": 0.148975791433892,
      "grad_norm": 1.7578125,
      "learning_rate": 1.89198606271777e-05,
      "loss": 0.5908828735351562,
      "mean_token_accuracy": 0.8213229507207871,
      "num_tokens": 5015453.0,
      "step": 380
    },
    {
      "entropy": 0.5584426306188106,
      "epoch": 0.1528962069979418,
      "grad_norm": 1.7265625,
      "learning_rate": 1.8832752613240418e-05,
      "loss": 0.5678494930267334,
      "mean_token_accuracy": 0.8277932897210121,
      "num_tokens": 5147133.0,
      "step": 390
    },
    {
      "entropy": 0.5676757667213679,
      "epoch": 0.15681662256199158,
      "grad_norm": 1.6640625,
      "learning_rate": 1.8745644599303138e-05,
      "loss": 0.5685997009277344,
      "mean_token_accuracy": 0.8280104413628578,
      "num_tokens": 5278668.0,
      "step": 400
    },
    {
      "entropy": 0.5922414932399989,
      "epoch": 0.16073703812604137,
      "grad_norm": 1.765625,
      "learning_rate": 1.8658536585365855e-05,
      "loss": 0.5975120067596436,
      "mean_token_accuracy": 0.8212826780974865,
      "num_tokens": 5411982.0,
      "step": 410
    },
    {
      "entropy": 0.6230888426303863,
      "epoch": 0.16465745369009116,
      "grad_norm": 1.84375,
      "learning_rate": 1.8571428571428575e-05,
      "loss": 0.6278081417083741,
      "mean_token_accuracy": 0.8109267845749855,
      "num_tokens": 5546647.0,
      "step": 420
    },
    {
      "entropy": 0.5828145634382963,
      "epoch": 0.16857786925414095,
      "grad_norm": 1.765625,
      "learning_rate": 1.8484320557491292e-05,
      "loss": 0.5854227542877197,
      "mean_token_accuracy": 0.8228457011282444,
      "num_tokens": 5679428.0,
      "step": 430
    },
    {
      "entropy": 0.5902278333902359,
      "epoch": 0.17249828481819074,
      "grad_norm": 1.75,
      "learning_rate": 1.839721254355401e-05,
      "loss": 0.5893653869628906,
      "mean_token_accuracy": 0.8216007687151432,
      "num_tokens": 5811806.0,
      "step": 440
    },
    {
      "entropy": 0.5815000183880329,
      "epoch": 0.17641870038224053,
      "grad_norm": 1.828125,
      "learning_rate": 1.8310104529616726e-05,
      "loss": 0.5896779060363769,
      "mean_token_accuracy": 0.8225190363824367,
      "num_tokens": 5945059.0,
      "step": 450
    },
    {
      "entropy": 0.6150704674422741,
      "epoch": 0.18033911594629032,
      "grad_norm": 1.828125,
      "learning_rate": 1.8222996515679442e-05,
      "loss": 0.6127326965332032,
      "mean_token_accuracy": 0.8161274991929531,
      "num_tokens": 6075109.0,
      "step": 460
    },
    {
      "entropy": 0.5994422011077404,
      "epoch": 0.1842595315103401,
      "grad_norm": 1.578125,
      "learning_rate": 1.8135888501742163e-05,
      "loss": 0.6054869174957276,
      "mean_token_accuracy": 0.8187170140445232,
      "num_tokens": 6206390.0,
      "step": 470
    },
    {
      "entropy": 0.5805528864264489,
      "epoch": 0.1881799470743899,
      "grad_norm": 1.875,
      "learning_rate": 1.804878048780488e-05,
      "loss": 0.5855868339538575,
      "mean_token_accuracy": 0.8244414009153843,
      "num_tokens": 6343008.0,
      "step": 480
    },
    {
      "entropy": 0.591498001664877,
      "epoch": 0.1921003626384397,
      "grad_norm": 1.7578125,
      "learning_rate": 1.7961672473867596e-05,
      "loss": 0.5893340587615967,
      "mean_token_accuracy": 0.8216389425098896,
      "num_tokens": 6474582.0,
      "step": 490
    },
    {
      "entropy": 0.626478412002325,
      "epoch": 0.19602077820248948,
      "grad_norm": 1.921875,
      "learning_rate": 1.7874564459930313e-05,
      "loss": 0.6263772487640381,
      "mean_token_accuracy": 0.8120981454849243,
      "num_tokens": 6607442.0,
      "step": 500
    },
    {
      "entropy": 0.5900808859616518,
      "epoch": 0.19994119376653927,
      "grad_norm": 1.5546875,
      "learning_rate": 1.7787456445993033e-05,
      "loss": 0.6009780883789062,
      "mean_token_accuracy": 0.819120641052723,
      "num_tokens": 6734350.0,
      "step": 510
    },
    {
      "entropy": 0.600034212693572,
      "epoch": 0.20386160933058906,
      "grad_norm": 1.78125,
      "learning_rate": 1.770034843205575e-05,
      "loss": 0.5948196887969971,
      "mean_token_accuracy": 0.8216341696679592,
      "num_tokens": 6869288.0,
      "step": 520
    },
    {
      "entropy": 0.5796573795378208,
      "epoch": 0.20778202489463884,
      "grad_norm": 1.9296875,
      "learning_rate": 1.761324041811847e-05,
      "loss": 0.5877546787261962,
      "mean_token_accuracy": 0.8229651033878327,
      "num_tokens": 7002603.0,
      "step": 530
    },
    {
      "entropy": 0.5981475539505482,
      "epoch": 0.21170244045868863,
      "grad_norm": 1.734375,
      "learning_rate": 1.7526132404181187e-05,
      "loss": 0.5986037254333496,
      "mean_token_accuracy": 0.8196288175880909,
      "num_tokens": 7135386.0,
      "step": 540
    },
    {
      "entropy": 0.5765955623239278,
      "epoch": 0.21562285602273842,
      "grad_norm": 1.6484375,
      "learning_rate": 1.7439024390243904e-05,
      "loss": 0.5820430278778076,
      "mean_token_accuracy": 0.8261640995740891,
      "num_tokens": 7269113.0,
      "step": 550
    },
    {
      "entropy": 0.5835400529205799,
      "epoch": 0.2195432715867882,
      "grad_norm": 1.96875,
      "learning_rate": 1.735191637630662e-05,
      "loss": 0.5847792148590087,
      "mean_token_accuracy": 0.8227986313402653,
      "num_tokens": 7397199.0,
      "step": 560
    },
    {
      "entropy": 0.5926254410296679,
      "epoch": 0.22346368715083798,
      "grad_norm": 1.6640625,
      "learning_rate": 1.7264808362369338e-05,
      "loss": 0.5946429252624512,
      "mean_token_accuracy": 0.8213822573423386,
      "num_tokens": 7528883.0,
      "step": 570
    },
    {
      "entropy": 0.5802333429455757,
      "epoch": 0.22738410271488776,
      "grad_norm": 1.8203125,
      "learning_rate": 1.7177700348432058e-05,
      "loss": 0.5808559417724609,
      "mean_token_accuracy": 0.8252004392445087,
      "num_tokens": 7661460.0,
      "step": 580
    },
    {
      "entropy": 0.5792893666774035,
      "epoch": 0.23130451827893755,
      "grad_norm": 1.78125,
      "learning_rate": 1.7090592334494775e-05,
      "loss": 0.5844778060913086,
      "mean_token_accuracy": 0.8232058539986611,
      "num_tokens": 7792654.0,
      "step": 590
    },
    {
      "entropy": 0.5875658553093672,
      "epoch": 0.23522493384298734,
      "grad_norm": 1.6328125,
      "learning_rate": 1.700348432055749e-05,
      "loss": 0.5934580802917481,
      "mean_token_accuracy": 0.8228402160108089,
      "num_tokens": 7929712.0,
      "step": 600
    },
    {
      "entropy": 0.6018010344356298,
      "epoch": 0.23914534940703713,
      "grad_norm": 1.8125,
      "learning_rate": 1.691637630662021e-05,
      "loss": 0.6018301010131836,
      "mean_token_accuracy": 0.8196375787258148,
      "num_tokens": 8062790.0,
      "step": 610
    },
    {
      "entropy": 0.5656840804964304,
      "epoch": 0.24306576497108692,
      "grad_norm": 1.65625,
      "learning_rate": 1.682926829268293e-05,
      "loss": 0.5681605339050293,
      "mean_token_accuracy": 0.829612398892641,
      "num_tokens": 8195902.0,
      "step": 620
    },
    {
      "entropy": 0.590726625546813,
      "epoch": 0.2469861805351367,
      "grad_norm": 1.7265625,
      "learning_rate": 1.6742160278745645e-05,
      "loss": 0.5950279712677002,
      "mean_token_accuracy": 0.8214106187224388,
      "num_tokens": 8326346.0,
      "step": 630
    },
    {
      "entropy": 0.5998687874525785,
      "epoch": 0.25090659609918653,
      "grad_norm": 1.734375,
      "learning_rate": 1.6655052264808366e-05,
      "loss": 0.5983600616455078,
      "mean_token_accuracy": 0.8189966283738613,
      "num_tokens": 8460029.0,
      "step": 640
    },
    {
      "entropy": 0.5715964876115323,
      "epoch": 0.2548270116632363,
      "grad_norm": 1.8359375,
      "learning_rate": 1.6567944250871082e-05,
      "loss": 0.5704405307769775,
      "mean_token_accuracy": 0.8268976680934429,
      "num_tokens": 8593771.0,
      "step": 650
    },
    {
      "entropy": 0.5904807798564434,
      "epoch": 0.2587474272272861,
      "grad_norm": 1.7265625,
      "learning_rate": 1.64808362369338e-05,
      "loss": 0.5930946350097657,
      "mean_token_accuracy": 0.8214319236576557,
      "num_tokens": 8727138.0,
      "step": 660
    },
    {
      "entropy": 0.587051372230053,
      "epoch": 0.26266784279133587,
      "grad_norm": 1.9296875,
      "learning_rate": 1.6393728222996516e-05,
      "loss": 0.5872443675994873,
      "mean_token_accuracy": 0.8230736367404461,
      "num_tokens": 8859295.0,
      "step": 670
    },
    {
      "entropy": 0.5858452804386616,
      "epoch": 0.2665882583553857,
      "grad_norm": 1.6640625,
      "learning_rate": 1.6306620209059233e-05,
      "loss": 0.5850666522979736,
      "mean_token_accuracy": 0.824101684987545,
      "num_tokens": 8990881.0,
      "step": 680
    },
    {
      "entropy": 0.5871569372713565,
      "epoch": 0.27050867391943545,
      "grad_norm": 1.640625,
      "learning_rate": 1.6219512195121953e-05,
      "loss": 0.594059944152832,
      "mean_token_accuracy": 0.8213366828858852,
      "num_tokens": 9123499.0,
      "step": 690
    },
    {
      "entropy": 0.5705580234527587,
      "epoch": 0.27442908948348527,
      "grad_norm": 1.828125,
      "learning_rate": 1.613240418118467e-05,
      "loss": 0.578093433380127,
      "mean_token_accuracy": 0.8268906474113464,
      "num_tokens": 9257724.0,
      "step": 700
    },
    {
      "entropy": 0.5944200098514557,
      "epoch": 0.27834950504753503,
      "grad_norm": 1.8671875,
      "learning_rate": 1.604529616724739e-05,
      "loss": 0.5930802345275878,
      "mean_token_accuracy": 0.8196114718914032,
      "num_tokens": 9393701.0,
      "step": 710
    },
    {
      "entropy": 0.5565787792205811,
      "epoch": 0.28226992061158485,
      "grad_norm": 1.7265625,
      "learning_rate": 1.5958188153310107e-05,
      "loss": 0.556602954864502,
      "mean_token_accuracy": 0.8305394992232322,
      "num_tokens": 9525747.0,
      "step": 720
    },
    {
      "entropy": 0.5937954898923635,
      "epoch": 0.2861903361756346,
      "grad_norm": 1.78125,
      "learning_rate": 1.5871080139372824e-05,
      "loss": 0.6040976524353028,
      "mean_token_accuracy": 0.819646991044283,
      "num_tokens": 9656394.0,
      "step": 730
    },
    {
      "entropy": 0.5655720047652721,
      "epoch": 0.2901107517396844,
      "grad_norm": 1.515625,
      "learning_rate": 1.578397212543554e-05,
      "loss": 0.5677313804626465,
      "mean_token_accuracy": 0.8287642747163773,
      "num_tokens": 9786846.0,
      "step": 740
    },
    {
      "entropy": 0.5690872885286808,
      "epoch": 0.2940311673037342,
      "grad_norm": 1.59375,
      "learning_rate": 1.569686411149826e-05,
      "loss": 0.577194356918335,
      "mean_token_accuracy": 0.8268992677330971,
      "num_tokens": 9920679.0,
      "step": 750
    },
    {
      "entropy": 0.5751310177147388,
      "epoch": 0.297951582867784,
      "grad_norm": 1.609375,
      "learning_rate": 1.5609756097560978e-05,
      "loss": 0.5761964797973633,
      "mean_token_accuracy": 0.8257504008710385,
      "num_tokens": 10052844.0,
      "step": 760
    },
    {
      "entropy": 0.5573539979755878,
      "epoch": 0.30187199843183377,
      "grad_norm": 1.71875,
      "learning_rate": 1.5522648083623694e-05,
      "loss": 0.5600615978240967,
      "mean_token_accuracy": 0.8300629191100597,
      "num_tokens": 10183471.0,
      "step": 770
    },
    {
      "entropy": 0.5975829754024744,
      "epoch": 0.3057924139958836,
      "grad_norm": 1.703125,
      "learning_rate": 1.543554006968641e-05,
      "loss": 0.5992860317230224,
      "mean_token_accuracy": 0.8206060849130153,
      "num_tokens": 10313392.0,
      "step": 780
    },
    {
      "entropy": 0.5773690041154623,
      "epoch": 0.30971282955993334,
      "grad_norm": 1.75,
      "learning_rate": 1.5348432055749128e-05,
      "loss": 0.5855609893798828,
      "mean_token_accuracy": 0.8234818607568741,
      "num_tokens": 10448285.0,
      "step": 790
    },
    {
      "entropy": 0.5769198387861252,
      "epoch": 0.31363324512398316,
      "grad_norm": 1.6484375,
      "learning_rate": 1.5261324041811848e-05,
      "loss": 0.5776193618774415,
      "mean_token_accuracy": 0.8255582615733147,
      "num_tokens": 10584236.0,
      "step": 800
    },
    {
      "entropy": 0.5881247483193874,
      "epoch": 0.3175536606880329,
      "grad_norm": 1.7734375,
      "learning_rate": 1.5174216027874567e-05,
      "loss": 0.593212080001831,
      "mean_token_accuracy": 0.8212029553949833,
      "num_tokens": 10716890.0,
      "step": 810
    },
    {
      "entropy": 0.5945972129702568,
      "epoch": 0.32147407625208274,
      "grad_norm": 1.6171875,
      "learning_rate": 1.5087108013937284e-05,
      "loss": 0.5877634525299072,
      "mean_token_accuracy": 0.8238732725381851,
      "num_tokens": 10852266.0,
      "step": 820
    },
    {
      "entropy": 0.5811539199203253,
      "epoch": 0.3253944918161325,
      "grad_norm": 1.796875,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.5907205104827881,
      "mean_token_accuracy": 0.8204894207417965,
      "num_tokens": 10985262.0,
      "step": 830
    },
    {
      "entropy": 0.5849005732685327,
      "epoch": 0.3293149073801823,
      "grad_norm": 1.5625,
      "learning_rate": 1.4912891986062719e-05,
      "loss": 0.5833800792694092,
      "mean_token_accuracy": 0.8241490855813026,
      "num_tokens": 11112681.0,
      "step": 840
    },
    {
      "entropy": 0.5829326704144477,
      "epoch": 0.3332353229442321,
      "grad_norm": 1.78125,
      "learning_rate": 1.4825783972125436e-05,
      "loss": 0.5943526268005371,
      "mean_token_accuracy": 0.8208344720304013,
      "num_tokens": 11240261.0,
      "step": 850
    },
    {
      "entropy": 0.6181501217186451,
      "epoch": 0.3371557385082819,
      "grad_norm": 1.7109375,
      "learning_rate": 1.4738675958188156e-05,
      "loss": 0.6180446147918701,
      "mean_token_accuracy": 0.8148257359862328,
      "num_tokens": 11370906.0,
      "step": 860
    },
    {
      "entropy": 0.5804374283179641,
      "epoch": 0.34107615407233166,
      "grad_norm": 1.65625,
      "learning_rate": 1.4651567944250873e-05,
      "loss": 0.5808406829833984,
      "mean_token_accuracy": 0.8237656883895397,
      "num_tokens": 11502338.0,
      "step": 870
    },
    {
      "entropy": 0.5756821602582931,
      "epoch": 0.3449965696363815,
      "grad_norm": 1.703125,
      "learning_rate": 1.456445993031359e-05,
      "loss": 0.5798084735870361,
      "mean_token_accuracy": 0.8261652231216431,
      "num_tokens": 11632165.0,
      "step": 880
    },
    {
      "entropy": 0.6068328361958265,
      "epoch": 0.34891698520043124,
      "grad_norm": 1.671875,
      "learning_rate": 1.4477351916376308e-05,
      "loss": 0.6065092563629151,
      "mean_token_accuracy": 0.8196589723229408,
      "num_tokens": 11763511.0,
      "step": 890
    },
    {
      "entropy": 0.5891778867691755,
      "epoch": 0.35283740076448106,
      "grad_norm": 1.609375,
      "learning_rate": 1.4390243902439025e-05,
      "loss": 0.592728567123413,
      "mean_token_accuracy": 0.8230046413838863,
      "num_tokens": 11895820.0,
      "step": 900
    },
    {
      "entropy": 0.6052472297102213,
      "epoch": 0.3567578163285308,
      "grad_norm": 1.7265625,
      "learning_rate": 1.4303135888501742e-05,
      "loss": 0.6157879829406738,
      "mean_token_accuracy": 0.8158482946455479,
      "num_tokens": 12025859.0,
      "step": 910
    },
    {
      "entropy": 0.5880816575139761,
      "epoch": 0.36067823189258064,
      "grad_norm": 1.828125,
      "learning_rate": 1.4216027874564462e-05,
      "loss": 0.5871949672698975,
      "mean_token_accuracy": 0.8238230250775814,
      "num_tokens": 12154614.0,
      "step": 920
    },
    {
      "entropy": 0.5858425028622151,
      "epoch": 0.3645986474566304,
      "grad_norm": 1.65625,
      "learning_rate": 1.4128919860627179e-05,
      "loss": 0.5850693225860596,
      "mean_token_accuracy": 0.8225606501102447,
      "num_tokens": 12289271.0,
      "step": 930
    },
    {
      "entropy": 0.6006566017866135,
      "epoch": 0.3685190630206802,
      "grad_norm": 1.765625,
      "learning_rate": 1.4041811846689897e-05,
      "loss": 0.6035278797149658,
      "mean_token_accuracy": 0.8171994499862194,
      "num_tokens": 12419706.0,
      "step": 940
    },
    {
      "entropy": 0.5940211407840252,
      "epoch": 0.37243947858473,
      "grad_norm": 1.9375,
      "learning_rate": 1.3954703832752614e-05,
      "loss": 0.5872697830200195,
      "mean_token_accuracy": 0.8215577825903893,
      "num_tokens": 12551888.0,
      "step": 950
    },
    {
      "entropy": 0.5769729625433684,
      "epoch": 0.3763598941487798,
      "grad_norm": 1.78125,
      "learning_rate": 1.3867595818815331e-05,
      "loss": 0.5793991088867188,
      "mean_token_accuracy": 0.8250952236354351,
      "num_tokens": 12685739.0,
      "step": 960
    },
    {
      "entropy": 0.5762162335216999,
      "epoch": 0.38028030971282956,
      "grad_norm": 1.640625,
      "learning_rate": 1.378048780487805e-05,
      "loss": 0.5773720741271973,
      "mean_token_accuracy": 0.8255156740546227,
      "num_tokens": 12817700.0,
      "step": 970
    },
    {
      "entropy": 0.5669316282495857,
      "epoch": 0.3842007252768794,
      "grad_norm": 1.78125,
      "learning_rate": 1.3693379790940768e-05,
      "loss": 0.5728450775146484,
      "mean_token_accuracy": 0.8277040965855121,
      "num_tokens": 12949817.0,
      "step": 980
    },
    {
      "entropy": 0.5594307694584131,
      "epoch": 0.38812114084092914,
      "grad_norm": 1.796875,
      "learning_rate": 1.3606271777003486e-05,
      "loss": 0.5640110015869141,
      "mean_token_accuracy": 0.8282136105000972,
      "num_tokens": 13081117.0,
      "step": 990
    },
    {
      "entropy": 0.5791688058525324,
      "epoch": 0.39204155640497895,
      "grad_norm": 1.7578125,
      "learning_rate": 1.3519163763066203e-05,
      "loss": 0.5831320285797119,
      "mean_token_accuracy": 0.8250723823904991,
      "num_tokens": 13210883.0,
      "step": 1000
    },
    {
      "entropy": 0.5651333026587964,
      "epoch": 0.3959619719690287,
      "grad_norm": 1.5859375,
      "learning_rate": 1.343205574912892e-05,
      "loss": 0.5608913898468018,
      "mean_token_accuracy": 0.8293542221188546,
      "num_tokens": 13341155.0,
      "step": 1010
    },
    {
      "entropy": 0.5587001539766788,
      "epoch": 0.39988238753307853,
      "grad_norm": 1.59375,
      "learning_rate": 1.3344947735191639e-05,
      "loss": 0.5611968040466309,
      "mean_token_accuracy": 0.8305698707699776,
      "num_tokens": 13474505.0,
      "step": 1020
    },
    {
      "entropy": 0.5640604499727487,
      "epoch": 0.4038028030971283,
      "grad_norm": 1.796875,
      "learning_rate": 1.3257839721254357e-05,
      "loss": 0.565390157699585,
      "mean_token_accuracy": 0.8282499298453331,
      "num_tokens": 13607139.0,
      "step": 1030
    },
    {
      "entropy": 0.5637519735842943,
      "epoch": 0.4077232186611781,
      "grad_norm": 1.6484375,
      "learning_rate": 1.3170731707317076e-05,
      "loss": 0.5695384502410888,
      "mean_token_accuracy": 0.8287955388426781,
      "num_tokens": 13738326.0,
      "step": 1040
    },
    {
      "entropy": 0.580963845923543,
      "epoch": 0.4116436342252279,
      "grad_norm": 1.765625,
      "learning_rate": 1.3083623693379792e-05,
      "loss": 0.5769538402557373,
      "mean_token_accuracy": 0.8235732421278954,
      "num_tokens": 13868452.0,
      "step": 1050
    },
    {
      "entropy": 0.5678158435970545,
      "epoch": 0.4155640497892777,
      "grad_norm": 1.8125,
      "learning_rate": 1.299651567944251e-05,
      "loss": 0.5710372924804688,
      "mean_token_accuracy": 0.8272775359451771,
      "num_tokens": 14002979.0,
      "step": 1060
    },
    {
      "entropy": 0.5668239049613476,
      "epoch": 0.41948446535332745,
      "grad_norm": 1.6328125,
      "learning_rate": 1.2909407665505226e-05,
      "loss": 0.5655869483947754,
      "mean_token_accuracy": 0.8278685718774795,
      "num_tokens": 14138341.0,
      "step": 1070
    },
    {
      "entropy": 0.5467250619083643,
      "epoch": 0.42340488091737727,
      "grad_norm": 1.75,
      "learning_rate": 1.2822299651567945e-05,
      "loss": 0.5479381561279297,
      "mean_token_accuracy": 0.8330328531563282,
      "num_tokens": 14266623.0,
      "step": 1080
    },
    {
      "entropy": 0.574169309064746,
      "epoch": 0.42732529648142703,
      "grad_norm": 1.578125,
      "learning_rate": 1.2735191637630663e-05,
      "loss": 0.580225658416748,
      "mean_token_accuracy": 0.8270016670227051,
      "num_tokens": 14397016.0,
      "step": 1090
    },
    {
      "entropy": 0.5842641271650791,
      "epoch": 0.43124571204547685,
      "grad_norm": 1.6171875,
      "learning_rate": 1.2648083623693382e-05,
      "loss": 0.5883337020874023,
      "mean_token_accuracy": 0.8239847645163536,
      "num_tokens": 14530946.0,
      "step": 1100
    },
    {
      "entropy": 0.5795733086764813,
      "epoch": 0.4351661276095266,
      "grad_norm": 1.71875,
      "learning_rate": 1.2560975609756098e-05,
      "loss": 0.5835409164428711,
      "mean_token_accuracy": 0.8237806595861912,
      "num_tokens": 14668493.0,
      "step": 1110
    },
    {
      "entropy": 0.5661398351192475,
      "epoch": 0.4390865431735764,
      "grad_norm": 1.8203125,
      "learning_rate": 1.2473867595818815e-05,
      "loss": 0.5688445091247558,
      "mean_token_accuracy": 0.8283659070730209,
      "num_tokens": 14798023.0,
      "step": 1120
    },
    {
      "entropy": 0.5716561190783978,
      "epoch": 0.4430069587376262,
      "grad_norm": 1.734375,
      "learning_rate": 1.2386759581881534e-05,
      "loss": 0.5737256526947021,
      "mean_token_accuracy": 0.8261031933128834,
      "num_tokens": 14927026.0,
      "step": 1130
    },
    {
      "entropy": 0.5815631907433272,
      "epoch": 0.44692737430167595,
      "grad_norm": 1.796875,
      "learning_rate": 1.2299651567944252e-05,
      "loss": 0.5841750144958496,
      "mean_token_accuracy": 0.8246684789657592,
      "num_tokens": 15059756.0,
      "step": 1140
    },
    {
      "entropy": 0.5747593600302935,
      "epoch": 0.45084778986572577,
      "grad_norm": 1.5546875,
      "learning_rate": 1.2212543554006971e-05,
      "loss": 0.5716405391693116,
      "mean_token_accuracy": 0.8256537966430187,
      "num_tokens": 15192135.0,
      "step": 1150
    },
    {
      "entropy": 0.5864708483219147,
      "epoch": 0.45476820542977553,
      "grad_norm": 1.875,
      "learning_rate": 1.2125435540069688e-05,
      "loss": 0.5913478851318359,
      "mean_token_accuracy": 0.8215075552463531,
      "num_tokens": 15326137.0,
      "step": 1160
    },
    {
      "entropy": 0.5958302663639188,
      "epoch": 0.45868862099382535,
      "grad_norm": 1.6796875,
      "learning_rate": 1.2038327526132404e-05,
      "loss": 0.5945512294769287,
      "mean_token_accuracy": 0.8203706957399846,
      "num_tokens": 15455625.0,
      "step": 1170
    },
    {
      "entropy": 0.5705973919481039,
      "epoch": 0.4626090365578751,
      "grad_norm": 1.8203125,
      "learning_rate": 1.1951219512195123e-05,
      "loss": 0.574383544921875,
      "mean_token_accuracy": 0.8264761112630368,
      "num_tokens": 15589324.0,
      "step": 1180
    },
    {
      "entropy": 0.5563835114240646,
      "epoch": 0.4665294521219249,
      "grad_norm": 1.640625,
      "learning_rate": 1.186411149825784e-05,
      "loss": 0.5601149559020996,
      "mean_token_accuracy": 0.8305692337453365,
      "num_tokens": 15725979.0,
      "step": 1190
    },
    {
      "entropy": 0.5868437562137843,
      "epoch": 0.4704498676859747,
      "grad_norm": 1.7109375,
      "learning_rate": 1.177700348432056e-05,
      "loss": 0.583247709274292,
      "mean_token_accuracy": 0.8226650767028332,
      "num_tokens": 15857159.0,
      "step": 1200
    },
    {
      "entropy": 0.5893880043178796,
      "epoch": 0.4743702832500245,
      "grad_norm": 1.671875,
      "learning_rate": 1.1689895470383277e-05,
      "loss": 0.6006449222564697,
      "mean_token_accuracy": 0.8204010248184204,
      "num_tokens": 15992142.0,
      "step": 1210
    },
    {
      "entropy": 0.5774647582322359,
      "epoch": 0.47829069881407427,
      "grad_norm": 1.734375,
      "learning_rate": 1.1602787456445994e-05,
      "loss": 0.5778764247894287,
      "mean_token_accuracy": 0.8252091869711876,
      "num_tokens": 16125065.0,
      "step": 1220
    },
    {
      "entropy": 0.5712914764881134,
      "epoch": 0.4822111143781241,
      "grad_norm": 1.828125,
      "learning_rate": 1.1515679442508712e-05,
      "loss": 0.569354248046875,
      "mean_token_accuracy": 0.8279658198356629,
      "num_tokens": 16254336.0,
      "step": 1230
    },
    {
      "entropy": 0.5910291790962219,
      "epoch": 0.48613152994217385,
      "grad_norm": 1.65625,
      "learning_rate": 1.1428571428571429e-05,
      "loss": 0.5976007461547852,
      "mean_token_accuracy": 0.8213621146976948,
      "num_tokens": 16387933.0,
      "step": 1240
    },
    {
      "entropy": 0.5892755780369043,
      "epoch": 0.49005194550622366,
      "grad_norm": 1.890625,
      "learning_rate": 1.1341463414634146e-05,
      "loss": 0.5938167572021484,
      "mean_token_accuracy": 0.8218666344881058,
      "num_tokens": 16519028.0,
      "step": 1250
    },
    {
      "entropy": 0.6019801579415798,
      "epoch": 0.4939723610702734,
      "grad_norm": 2.109375,
      "learning_rate": 1.1254355400696866e-05,
      "loss": 0.601860523223877,
      "mean_token_accuracy": 0.8193305231630802,
      "num_tokens": 16648075.0,
      "step": 1260
    },
    {
      "entropy": 0.577896298468113,
      "epoch": 0.49789277663432324,
      "grad_norm": 1.5859375,
      "learning_rate": 1.1167247386759583e-05,
      "loss": 0.5790606498718261,
      "mean_token_accuracy": 0.8256713755428791,
      "num_tokens": 16784827.0,
      "step": 1270
    },
    {
      "entropy": 0.5828588411211968,
      "epoch": 0.5018131921983731,
      "grad_norm": 1.8671875,
      "learning_rate": 1.1080139372822301e-05,
      "loss": 0.5897656440734863,
      "mean_token_accuracy": 0.8219093471765518,
      "num_tokens": 16914566.0,
      "step": 1280
    },
    {
      "entropy": 0.5996304292231798,
      "epoch": 0.5057336077624228,
      "grad_norm": 1.625,
      "learning_rate": 1.0993031358885018e-05,
      "loss": 0.6056152820587158,
      "mean_token_accuracy": 0.8185230061411858,
      "num_tokens": 17047396.0,
      "step": 1290
    },
    {
      "entropy": 0.5672778323292732,
      "epoch": 0.5096540233264726,
      "grad_norm": 1.7109375,
      "learning_rate": 1.0905923344947735e-05,
      "loss": 0.5606307983398438,
      "mean_token_accuracy": 0.8289205953478813,
      "num_tokens": 17173863.0,
      "step": 1300
    },
    {
      "entropy": 0.574262236058712,
      "epoch": 0.5135744388905223,
      "grad_norm": 1.546875,
      "learning_rate": 1.0818815331010455e-05,
      "loss": 0.5738330364227295,
      "mean_token_accuracy": 0.8259080529212952,
      "num_tokens": 17305917.0,
      "step": 1310
    },
    {
      "entropy": 0.5749511100351811,
      "epoch": 0.5174948544545722,
      "grad_norm": 1.671875,
      "learning_rate": 1.0731707317073172e-05,
      "loss": 0.5709137916564941,
      "mean_token_accuracy": 0.8254921354353428,
      "num_tokens": 17436774.0,
      "step": 1320
    },
    {
      "entropy": 0.595110259950161,
      "epoch": 0.521415270018622,
      "grad_norm": 1.6484375,
      "learning_rate": 1.0644599303135889e-05,
      "loss": 0.5969788551330566,
      "mean_token_accuracy": 0.8186600834131241,
      "num_tokens": 17569314.0,
      "step": 1330
    },
    {
      "entropy": 0.5713474582880735,
      "epoch": 0.5253356855826717,
      "grad_norm": 1.7421875,
      "learning_rate": 1.0557491289198607e-05,
      "loss": 0.5697742462158203,
      "mean_token_accuracy": 0.8268165580928326,
      "num_tokens": 17703511.0,
      "step": 1340
    },
    {
      "entropy": 0.5673128481954336,
      "epoch": 0.5292561011467215,
      "grad_norm": 1.8828125,
      "learning_rate": 1.0470383275261324e-05,
      "loss": 0.5756454944610596,
      "mean_token_accuracy": 0.8265850961208343,
      "num_tokens": 17837601.0,
      "step": 1350
    },
    {
      "entropy": 0.5455819856375456,
      "epoch": 0.5331765167107714,
      "grad_norm": 1.5390625,
      "learning_rate": 1.0383275261324041e-05,
      "loss": 0.5512715339660644,
      "mean_token_accuracy": 0.8321440435945988,
      "num_tokens": 17968783.0,
      "step": 1360
    },
    {
      "entropy": 0.5659995820373297,
      "epoch": 0.5370969322748211,
      "grad_norm": 1.890625,
      "learning_rate": 1.0296167247386761e-05,
      "loss": 0.5727234363555909,
      "mean_token_accuracy": 0.8270003162324429,
      "num_tokens": 18100467.0,
      "step": 1370
    },
    {
      "entropy": 0.5558266244828701,
      "epoch": 0.5410173478388709,
      "grad_norm": 1.71875,
      "learning_rate": 1.0209059233449478e-05,
      "loss": 0.5511263847351074,
      "mean_token_accuracy": 0.8310491070151329,
      "num_tokens": 18236883.0,
      "step": 1380
    },
    {
      "entropy": 0.5848776459693908,
      "epoch": 0.5449377634029207,
      "grad_norm": 1.78125,
      "learning_rate": 1.0121951219512197e-05,
      "loss": 0.5840703964233398,
      "mean_token_accuracy": 0.8221600718796254,
      "num_tokens": 18366506.0,
      "step": 1390
    },
    {
      "entropy": 0.5772740695625543,
      "epoch": 0.5488581789669705,
      "grad_norm": 1.6171875,
      "learning_rate": 1.0034843205574913e-05,
      "loss": 0.5818367004394531,
      "mean_token_accuracy": 0.8251958817243576,
      "num_tokens": 18499076.0,
      "step": 1400
    },
    {
      "entropy": 0.5877921745181084,
      "epoch": 0.5527785945310203,
      "grad_norm": 1.71875,
      "learning_rate": 9.947735191637632e-06,
      "loss": 0.5859257221221924,
      "mean_token_accuracy": 0.8229536414146423,
      "num_tokens": 18627838.0,
      "step": 1410
    },
    {
      "entropy": 0.5713149573653936,
      "epoch": 0.5566990100950701,
      "grad_norm": 1.8046875,
      "learning_rate": 9.860627177700349e-06,
      "loss": 0.5759402751922608,
      "mean_token_accuracy": 0.8259151518344879,
      "num_tokens": 18757141.0,
      "step": 1420
    },
    {
      "entropy": 0.5910889457911253,
      "epoch": 0.5606194256591198,
      "grad_norm": 1.7421875,
      "learning_rate": 9.773519163763067e-06,
      "loss": 0.5957850933074951,
      "mean_token_accuracy": 0.8201118834316731,
      "num_tokens": 18888358.0,
      "step": 1430
    },
    {
      "entropy": 0.5661411985754967,
      "epoch": 0.5645398412231697,
      "grad_norm": 1.6015625,
      "learning_rate": 9.686411149825786e-06,
      "loss": 0.5634143829345704,
      "mean_token_accuracy": 0.8272467255592346,
      "num_tokens": 19020119.0,
      "step": 1440
    },
    {
      "entropy": 0.5582704734057188,
      "epoch": 0.5684602567872195,
      "grad_norm": 1.6484375,
      "learning_rate": 9.599303135888503e-06,
      "loss": 0.5596834659576416,
      "mean_token_accuracy": 0.831222715228796,
      "num_tokens": 19152625.0,
      "step": 1450
    },
    {
      "entropy": 0.5848473913967609,
      "epoch": 0.5723806723512692,
      "grad_norm": 1.5625,
      "learning_rate": 9.51219512195122e-06,
      "loss": 0.5914738655090332,
      "mean_token_accuracy": 0.821570199728012,
      "num_tokens": 19282646.0,
      "step": 1460
    },
    {
      "entropy": 0.5610104382038117,
      "epoch": 0.576301087915319,
      "grad_norm": 1.625,
      "learning_rate": 9.425087108013938e-06,
      "loss": 0.5633234500885009,
      "mean_token_accuracy": 0.8292761743068695,
      "num_tokens": 19418710.0,
      "step": 1470
    },
    {
      "entropy": 0.5768874280154705,
      "epoch": 0.5802215034793688,
      "grad_norm": 1.6875,
      "learning_rate": 9.337979094076656e-06,
      "loss": 0.5723718166351318,
      "mean_token_accuracy": 0.827151071280241,
      "num_tokens": 19548708.0,
      "step": 1480
    },
    {
      "entropy": 0.5698943838477135,
      "epoch": 0.5841419190434186,
      "grad_norm": 1.8359375,
      "learning_rate": 9.250871080139373e-06,
      "loss": 0.570045804977417,
      "mean_token_accuracy": 0.8272446699440479,
      "num_tokens": 19684081.0,
      "step": 1490
    },
    {
      "entropy": 0.5802167691290379,
      "epoch": 0.5880623346074684,
      "grad_norm": 1.65625,
      "learning_rate": 9.163763066202092e-06,
      "loss": 0.5790982246398926,
      "mean_token_accuracy": 0.8250758126378059,
      "num_tokens": 19817358.0,
      "step": 1500
    },
    {
      "entropy": 0.5764854367822408,
      "epoch": 0.5919827501715181,
      "grad_norm": 1.6640625,
      "learning_rate": 9.076655052264809e-06,
      "loss": 0.5897452354431152,
      "mean_token_accuracy": 0.8235625132918358,
      "num_tokens": 19947516.0,
      "step": 1510
    },
    {
      "entropy": 0.568101118132472,
      "epoch": 0.595903165735568,
      "grad_norm": 1.7734375,
      "learning_rate": 8.989547038327527e-06,
      "loss": 0.5667146205902099,
      "mean_token_accuracy": 0.8266186200082302,
      "num_tokens": 20080882.0,
      "step": 1520
    },
    {
      "entropy": 0.5726208575069904,
      "epoch": 0.5998235812996178,
      "grad_norm": 1.6171875,
      "learning_rate": 8.902439024390244e-06,
      "loss": 0.5737858772277832,
      "mean_token_accuracy": 0.8270730949938297,
      "num_tokens": 20208040.0,
      "step": 1530
    },
    {
      "entropy": 0.5912529457360506,
      "epoch": 0.6037439968636675,
      "grad_norm": 1.671875,
      "learning_rate": 8.815331010452962e-06,
      "loss": 0.5949665069580078,
      "mean_token_accuracy": 0.821034874767065,
      "num_tokens": 20342523.0,
      "step": 1540
    },
    {
      "entropy": 0.5770879618823528,
      "epoch": 0.6076644124277173,
      "grad_norm": 1.609375,
      "learning_rate": 8.728222996515681e-06,
      "loss": 0.5765514373779297,
      "mean_token_accuracy": 0.8256359219551086,
      "num_tokens": 20475073.0,
      "step": 1550
    },
    {
      "entropy": 0.5801304239779711,
      "epoch": 0.6115848279917672,
      "grad_norm": 1.8671875,
      "learning_rate": 8.641114982578398e-06,
      "loss": 0.5835923194885254,
      "mean_token_accuracy": 0.8234978429973125,
      "num_tokens": 20607544.0,
      "step": 1560
    },
    {
      "entropy": 0.5704087316989899,
      "epoch": 0.6155052435558169,
      "grad_norm": 1.8203125,
      "learning_rate": 8.554006968641115e-06,
      "loss": 0.5746907234191895,
      "mean_token_accuracy": 0.826793709397316,
      "num_tokens": 20740547.0,
      "step": 1570
    },
    {
      "entropy": 0.5753558877855539,
      "epoch": 0.6194256591198667,
      "grad_norm": 1.7890625,
      "learning_rate": 8.466898954703833e-06,
      "loss": 0.5788108825683593,
      "mean_token_accuracy": 0.8253151409327983,
      "num_tokens": 20872233.0,
      "step": 1580
    },
    {
      "entropy": 0.58878487162292,
      "epoch": 0.6233460746839165,
      "grad_norm": 1.640625,
      "learning_rate": 8.379790940766552e-06,
      "loss": 0.5929996490478515,
      "mean_token_accuracy": 0.8213138461112977,
      "num_tokens": 21004889.0,
      "step": 1590
    },
    {
      "entropy": 0.5900854453444481,
      "epoch": 0.6272664902479663,
      "grad_norm": 1.8125,
      "learning_rate": 8.292682926829268e-06,
      "loss": 0.5887657642364502,
      "mean_token_accuracy": 0.8229175060987473,
      "num_tokens": 21136824.0,
      "step": 1600
    },
    {
      "entropy": 0.5752101615071297,
      "epoch": 0.6311869058120161,
      "grad_norm": 1.65625,
      "learning_rate": 8.205574912891987e-06,
      "loss": 0.5777340412139893,
      "mean_token_accuracy": 0.8268123485147953,
      "num_tokens": 21270819.0,
      "step": 1610
    },
    {
      "entropy": 0.5686212468892335,
      "epoch": 0.6351073213760658,
      "grad_norm": 1.625,
      "learning_rate": 8.118466898954704e-06,
      "loss": 0.5700377941131591,
      "mean_token_accuracy": 0.8275717034935951,
      "num_tokens": 21403507.0,
      "step": 1620
    },
    {
      "entropy": 0.588944623246789,
      "epoch": 0.6390277369401156,
      "grad_norm": 1.7109375,
      "learning_rate": 8.031358885017422e-06,
      "loss": 0.5895246028900146,
      "mean_token_accuracy": 0.8222635351121426,
      "num_tokens": 21539521.0,
      "step": 1630
    },
    {
      "entropy": 0.5585341576486826,
      "epoch": 0.6429481525041655,
      "grad_norm": 1.6875,
      "learning_rate": 7.94425087108014e-06,
      "loss": 0.5642005443572998,
      "mean_token_accuracy": 0.8300835333764554,
      "num_tokens": 21672827.0,
      "step": 1640
    },
    {
      "entropy": 0.601011986285448,
      "epoch": 0.6468685680682152,
      "grad_norm": 1.703125,
      "learning_rate": 7.857142857142858e-06,
      "loss": 0.6064223766326904,
      "mean_token_accuracy": 0.8166609443724155,
      "num_tokens": 21803538.0,
      "step": 1650
    },
    {
      "entropy": 0.5758274313062429,
      "epoch": 0.650788983632265,
      "grad_norm": 1.5625,
      "learning_rate": 7.770034843205574e-06,
      "loss": 0.5752018928527832,
      "mean_token_accuracy": 0.8261359445750713,
      "num_tokens": 21937475.0,
      "step": 1660
    },
    {
      "entropy": 0.5863972809165716,
      "epoch": 0.6547093991963148,
      "grad_norm": 1.9140625,
      "learning_rate": 7.682926829268293e-06,
      "loss": 0.5884019374847412,
      "mean_token_accuracy": 0.8222149938344956,
      "num_tokens": 22076290.0,
      "step": 1670
    },
    {
      "entropy": 0.5837476711720229,
      "epoch": 0.6586298147603646,
      "grad_norm": 1.625,
      "learning_rate": 7.595818815331011e-06,
      "loss": 0.5863306522369385,
      "mean_token_accuracy": 0.8223751485347748,
      "num_tokens": 22210848.0,
      "step": 1680
    },
    {
      "entropy": 0.5637796241790056,
      "epoch": 0.6625502303244144,
      "grad_norm": 1.6953125,
      "learning_rate": 7.508710801393729e-06,
      "loss": 0.5643290996551513,
      "mean_token_accuracy": 0.8271290838718415,
      "num_tokens": 22342981.0,
      "step": 1690
    },
    {
      "entropy": 0.6036053825169801,
      "epoch": 0.6664706458884642,
      "grad_norm": 1.7109375,
      "learning_rate": 7.421602787456447e-06,
      "loss": 0.6125275135040283,
      "mean_token_accuracy": 0.8163446709513664,
      "num_tokens": 22473724.0,
      "step": 1700
    },
    {
      "entropy": 0.5783885445445776,
      "epoch": 0.6703910614525139,
      "grad_norm": 1.609375,
      "learning_rate": 7.334494773519164e-06,
      "loss": 0.5844399452209472,
      "mean_token_accuracy": 0.8247374482452869,
      "num_tokens": 22603403.0,
      "step": 1710
    },
    {
      "entropy": 0.5773308865725995,
      "epoch": 0.6743114770165638,
      "grad_norm": 1.546875,
      "learning_rate": 7.247386759581882e-06,
      "loss": 0.5766704082489014,
      "mean_token_accuracy": 0.8254517287015914,
      "num_tokens": 22737181.0,
      "step": 1720
    },
    {
      "entropy": 0.5680316600948572,
      "epoch": 0.6782318925806136,
      "grad_norm": 1.75,
      "learning_rate": 7.1602787456446e-06,
      "loss": 0.5672269821166992,
      "mean_token_accuracy": 0.8279304966330528,
      "num_tokens": 22869837.0,
      "step": 1730
    },
    {
      "entropy": 0.5941996563225984,
      "epoch": 0.6821523081446633,
      "grad_norm": 1.7578125,
      "learning_rate": 7.0731707317073175e-06,
      "loss": 0.5911007404327393,
      "mean_token_accuracy": 0.8221167460083961,
      "num_tokens": 23003947.0,
      "step": 1740
    },
    {
      "entropy": 0.5591667950153351,
      "epoch": 0.6860727237087131,
      "grad_norm": 1.71875,
      "learning_rate": 6.986062717770036e-06,
      "loss": 0.5607988834381104,
      "mean_token_accuracy": 0.8297918803989888,
      "num_tokens": 23134580.0,
      "step": 1750
    },
    {
      "entropy": 0.5676139583811164,
      "epoch": 0.689993139272763,
      "grad_norm": 1.6875,
      "learning_rate": 6.898954703832753e-06,
      "loss": 0.5650452613830567,
      "mean_token_accuracy": 0.8275837257504464,
      "num_tokens": 23264614.0,
      "step": 1760
    },
    {
      "entropy": 0.5673956066370011,
      "epoch": 0.6939135548368127,
      "grad_norm": 1.7265625,
      "learning_rate": 6.8118466898954705e-06,
      "loss": 0.5670190334320069,
      "mean_token_accuracy": 0.8293615274131299,
      "num_tokens": 23396741.0,
      "step": 1770
    },
    {
      "entropy": 0.5742564305663109,
      "epoch": 0.6978339704008625,
      "grad_norm": 1.6875,
      "learning_rate": 6.724738675958189e-06,
      "loss": 0.5740202903747559,
      "mean_token_accuracy": 0.8251322150230408,
      "num_tokens": 23527918.0,
      "step": 1780
    },
    {
      "entropy": 0.5919697143137455,
      "epoch": 0.7017543859649122,
      "grad_norm": 1.8046875,
      "learning_rate": 6.637630662020907e-06,
      "loss": 0.5907740116119384,
      "mean_token_accuracy": 0.8220984056591988,
      "num_tokens": 23657724.0,
      "step": 1790
    },
    {
      "entropy": 0.5903078105300665,
      "epoch": 0.7056748015289621,
      "grad_norm": 1.7109375,
      "learning_rate": 6.5505226480836235e-06,
      "loss": 0.5937789440155029,
      "mean_token_accuracy": 0.8217935189604759,
      "num_tokens": 23786805.0,
      "step": 1800
    },
    {
      "entropy": 0.5733942896127701,
      "epoch": 0.7095952170930119,
      "grad_norm": 1.6171875,
      "learning_rate": 6.463414634146342e-06,
      "loss": 0.5705701828002929,
      "mean_token_accuracy": 0.8282980337738991,
      "num_tokens": 23923037.0,
      "step": 1810
    },
    {
      "entropy": 0.5554268516600132,
      "epoch": 0.7135156326570616,
      "grad_norm": 1.7890625,
      "learning_rate": 6.37630662020906e-06,
      "loss": 0.5606462955474854,
      "mean_token_accuracy": 0.8297950372099876,
      "num_tokens": 24053827.0,
      "step": 1820
    },
    {
      "entropy": 0.6039480961859226,
      "epoch": 0.7174360482211114,
      "grad_norm": 1.703125,
      "learning_rate": 6.289198606271778e-06,
      "loss": 0.6047250747680664,
      "mean_token_accuracy": 0.8179130852222443,
      "num_tokens": 24186062.0,
      "step": 1830
    },
    {
      "entropy": 0.570480278134346,
      "epoch": 0.7213564637851613,
      "grad_norm": 1.6796875,
      "learning_rate": 6.202090592334495e-06,
      "loss": 0.5673944473266601,
      "mean_token_accuracy": 0.8284930318593979,
      "num_tokens": 24315993.0,
      "step": 1840
    },
    {
      "entropy": 0.5931850384920836,
      "epoch": 0.725276879349211,
      "grad_norm": 1.78125,
      "learning_rate": 6.114982578397213e-06,
      "loss": 0.5964553833007813,
      "mean_token_accuracy": 0.8212155938148499,
      "num_tokens": 24446858.0,
      "step": 1850
    },
    {
      "entropy": 0.6047229062765836,
      "epoch": 0.7291972949132608,
      "grad_norm": 1.828125,
      "learning_rate": 6.027874564459931e-06,
      "loss": 0.6088790416717529,
      "mean_token_accuracy": 0.817912295460701,
      "num_tokens": 24580558.0,
      "step": 1860
    },
    {
      "entropy": 0.5825779817998409,
      "epoch": 0.7331177104773106,
      "grad_norm": 1.859375,
      "learning_rate": 5.940766550522649e-06,
      "loss": 0.5805138111114502,
      "mean_token_accuracy": 0.8254112429916859,
      "num_tokens": 24718156.0,
      "step": 1870
    },
    {
      "entropy": 0.5747309617698193,
      "epoch": 0.7370381260413604,
      "grad_norm": 1.6953125,
      "learning_rate": 5.853658536585366e-06,
      "loss": 0.5785604953765869,
      "mean_token_accuracy": 0.8248820044100285,
      "num_tokens": 24853874.0,
      "step": 1880
    },
    {
      "entropy": 0.601536936685443,
      "epoch": 0.7409585416054102,
      "grad_norm": 1.8671875,
      "learning_rate": 5.766550522648084e-06,
      "loss": 0.6086882591247559,
      "mean_token_accuracy": 0.8180929891765117,
      "num_tokens": 24983767.0,
      "step": 1890
    },
    {
      "entropy": 0.5778564881533385,
      "epoch": 0.74487895716946,
      "grad_norm": 1.6484375,
      "learning_rate": 5.679442508710802e-06,
      "loss": 0.5777944087982178,
      "mean_token_accuracy": 0.8248727723956109,
      "num_tokens": 25112481.0,
      "step": 1900
    },
    {
      "entropy": 0.5561993703246116,
      "epoch": 0.7487993727335097,
      "grad_norm": 1.5859375,
      "learning_rate": 5.59233449477352e-06,
      "loss": 0.5545031547546386,
      "mean_token_accuracy": 0.8313787803053856,
      "num_tokens": 25244621.0,
      "step": 1910
    },
    {
      "entropy": 0.5606953276321291,
      "epoch": 0.7527197882975596,
      "grad_norm": 1.640625,
      "learning_rate": 5.505226480836237e-06,
      "loss": 0.5607188224792481,
      "mean_token_accuracy": 0.8297309316694736,
      "num_tokens": 25376819.0,
      "step": 1920
    },
    {
      "entropy": 0.5783195950090885,
      "epoch": 0.7566402038616094,
      "grad_norm": 1.78125,
      "learning_rate": 5.418118466898955e-06,
      "loss": 0.57740797996521,
      "mean_token_accuracy": 0.8247730396687984,
      "num_tokens": 25509601.0,
      "step": 1930
    },
    {
      "entropy": 0.5782700140029192,
      "epoch": 0.7605606194256591,
      "grad_norm": 1.6796875,
      "learning_rate": 5.331010452961673e-06,
      "loss": 0.5782852649688721,
      "mean_token_accuracy": 0.824606055766344,
      "num_tokens": 25643050.0,
      "step": 1940
    },
    {
      "entropy": 0.561905774474144,
      "epoch": 0.7644810349897089,
      "grad_norm": 1.6171875,
      "learning_rate": 5.243902439024391e-06,
      "loss": 0.5643265724182129,
      "mean_token_accuracy": 0.8287928692996502,
      "num_tokens": 25769796.0,
      "step": 1950
    },
    {
      "entropy": 0.5690379086881876,
      "epoch": 0.7684014505537587,
      "grad_norm": 1.6796875,
      "learning_rate": 5.156794425087108e-06,
      "loss": 0.5731242179870606,
      "mean_token_accuracy": 0.8275946989655495,
      "num_tokens": 25904632.0,
      "step": 1960
    },
    {
      "entropy": 0.5504785589873791,
      "epoch": 0.7723218661178085,
      "grad_norm": 1.53125,
      "learning_rate": 5.0696864111498264e-06,
      "loss": 0.5516434192657471,
      "mean_token_accuracy": 0.8325082875788212,
      "num_tokens": 26035719.0,
      "step": 1970
    },
    {
      "entropy": 0.5725247742608189,
      "epoch": 0.7762422816818583,
      "grad_norm": 1.734375,
      "learning_rate": 4.982578397212544e-06,
      "loss": 0.5785993576049805,
      "mean_token_accuracy": 0.8263410687446594,
      "num_tokens": 26167472.0,
      "step": 1980
    },
    {
      "entropy": 0.5720619566738605,
      "epoch": 0.780162697245908,
      "grad_norm": 1.5390625,
      "learning_rate": 4.895470383275262e-06,
      "loss": 0.5713248252868652,
      "mean_token_accuracy": 0.8277824930846691,
      "num_tokens": 26301163.0,
      "step": 1990
    },
    {
      "entropy": 0.5604069098830223,
      "epoch": 0.7840831128099579,
      "grad_norm": 1.6875,
      "learning_rate": 4.8083623693379794e-06,
      "loss": 0.5647254467010498,
      "mean_token_accuracy": 0.8302777230739593,
      "num_tokens": 26430868.0,
      "step": 2000
    },
    {
      "entropy": 0.5837776899337769,
      "epoch": 0.7880035283740077,
      "grad_norm": 1.7734375,
      "learning_rate": 4.721254355400697e-06,
      "loss": 0.5858076095581055,
      "mean_token_accuracy": 0.8225198201835155,
      "num_tokens": 26560775.0,
      "step": 2010
    },
    {
      "entropy": 0.5733296349644661,
      "epoch": 0.7919239439380574,
      "grad_norm": 1.625,
      "learning_rate": 4.634146341463416e-06,
      "loss": 0.5758630275726319,
      "mean_token_accuracy": 0.8253522992134095,
      "num_tokens": 26693234.0,
      "step": 2020
    },
    {
      "entropy": 0.557684974372387,
      "epoch": 0.7958443595021072,
      "grad_norm": 1.7109375,
      "learning_rate": 4.5470383275261325e-06,
      "loss": 0.5551248073577881,
      "mean_token_accuracy": 0.8318465322256088,
      "num_tokens": 26828863.0,
      "step": 2030
    },
    {
      "entropy": 0.5628693040460349,
      "epoch": 0.7997647750661571,
      "grad_norm": 1.75,
      "learning_rate": 4.45993031358885e-06,
      "loss": 0.571585750579834,
      "mean_token_accuracy": 0.8270311810076236,
      "num_tokens": 26960662.0,
      "step": 2040
    },
    {
      "entropy": 0.5897116485983134,
      "epoch": 0.8036851906302068,
      "grad_norm": 1.7421875,
      "learning_rate": 4.372822299651569e-06,
      "loss": 0.5966588497161865,
      "mean_token_accuracy": 0.8230571210384369,
      "num_tokens": 27091359.0,
      "step": 2050
    },
    {
      "entropy": 0.5869602940976619,
      "epoch": 0.8076056061942566,
      "grad_norm": 1.7109375,
      "learning_rate": 4.2857142857142855e-06,
      "loss": 0.588222885131836,
      "mean_token_accuracy": 0.8202385693788529,
      "num_tokens": 27225435.0,
      "step": 2060
    },
    {
      "entropy": 0.6028570268303156,
      "epoch": 0.8115260217583063,
      "grad_norm": 1.6328125,
      "learning_rate": 4.198606271777004e-06,
      "loss": 0.606553316116333,
      "mean_token_accuracy": 0.8188532285392285,
      "num_tokens": 27358034.0,
      "step": 2070
    },
    {
      "entropy": 0.5583208829164505,
      "epoch": 0.8154464373223562,
      "grad_norm": 1.78125,
      "learning_rate": 4.111498257839722e-06,
      "loss": 0.5582189083099365,
      "mean_token_accuracy": 0.8306586474180222,
      "num_tokens": 27490146.0,
      "step": 2080
    },
    {
      "entropy": 0.5879612069576978,
      "epoch": 0.819366852886406,
      "grad_norm": 1.8359375,
      "learning_rate": 4.024390243902439e-06,
      "loss": 0.5868856906890869,
      "mean_token_accuracy": 0.821238712221384,
      "num_tokens": 27621710.0,
      "step": 2090
    },
    {
      "entropy": 0.5645837895572186,
      "epoch": 0.8232872684504557,
      "grad_norm": 1.6640625,
      "learning_rate": 3.937282229965157e-06,
      "loss": 0.5639263153076172,
      "mean_token_accuracy": 0.8281869366765022,
      "num_tokens": 27758436.0,
      "step": 2100
    },
    {
      "entropy": 0.5650495100766421,
      "epoch": 0.8272076840145055,
      "grad_norm": 1.71875,
      "learning_rate": 3.850174216027875e-06,
      "loss": 0.5659061908721924,
      "mean_token_accuracy": 0.8303346544504165,
      "num_tokens": 27886659.0,
      "step": 2110
    },
    {
      "entropy": 0.5669478211551905,
      "epoch": 0.8311280995785554,
      "grad_norm": 1.875,
      "learning_rate": 3.7630662020905927e-06,
      "loss": 0.5676323890686035,
      "mean_token_accuracy": 0.8264915093779563,
      "num_tokens": 28020869.0,
      "step": 2120
    },
    {
      "entropy": 0.5540701054036618,
      "epoch": 0.8350485151426051,
      "grad_norm": 1.71875,
      "learning_rate": 3.67595818815331e-06,
      "loss": 0.5579257011413574,
      "mean_token_accuracy": 0.8317541219294071,
      "num_tokens": 28152491.0,
      "step": 2130
    },
    {
      "entropy": 0.5800126396119595,
      "epoch": 0.8389689307066549,
      "grad_norm": 1.8125,
      "learning_rate": 3.588850174216028e-06,
      "loss": 0.5853462219238281,
      "mean_token_accuracy": 0.8237727522850037,
      "num_tokens": 28284698.0,
      "step": 2140
    },
    {
      "entropy": 0.6118997160345316,
      "epoch": 0.8428893462707047,
      "grad_norm": 1.6796875,
      "learning_rate": 3.501742160278746e-06,
      "loss": 0.6129735469818115,
      "mean_token_accuracy": 0.8170387588441372,
      "num_tokens": 28413374.0,
      "step": 2150
    },
    {
      "entropy": 0.5741523541510105,
      "epoch": 0.8468097618347545,
      "grad_norm": 1.703125,
      "learning_rate": 3.414634146341464e-06,
      "loss": 0.5773540019989014,
      "mean_token_accuracy": 0.8257443487644196,
      "num_tokens": 28545962.0,
      "step": 2160
    },
    {
      "entropy": 0.5787751715630293,
      "epoch": 0.8507301773988043,
      "grad_norm": 1.6796875,
      "learning_rate": 3.3275261324041815e-06,
      "loss": 0.5790078163146972,
      "mean_token_accuracy": 0.825350683927536,
      "num_tokens": 28680445.0,
      "step": 2170
    },
    {
      "entropy": 0.5716471575200558,
      "epoch": 0.8546505929628541,
      "grad_norm": 1.6953125,
      "learning_rate": 3.240418118466899e-06,
      "loss": 0.5730648994445801,
      "mean_token_accuracy": 0.8268438093364239,
      "num_tokens": 28811694.0,
      "step": 2180
    },
    {
      "entropy": 0.5720145717263222,
      "epoch": 0.8585710085269038,
      "grad_norm": 1.84375,
      "learning_rate": 3.1533101045296173e-06,
      "loss": 0.5759499549865723,
      "mean_token_accuracy": 0.8254265673458576,
      "num_tokens": 28945890.0,
      "step": 2190
    },
    {
      "entropy": 0.5667183842509985,
      "epoch": 0.8624914240909537,
      "grad_norm": 1.796875,
      "learning_rate": 3.0662020905923345e-06,
      "loss": 0.567453670501709,
      "mean_token_accuracy": 0.8277363143861294,
      "num_tokens": 29075294.0,
      "step": 2200
    },
    {
      "entropy": 0.583325557038188,
      "epoch": 0.8664118396550035,
      "grad_norm": 1.7109375,
      "learning_rate": 2.9790940766550526e-06,
      "loss": 0.5870438098907471,
      "mean_token_accuracy": 0.8234334781765937,
      "num_tokens": 29207413.0,
      "step": 2210
    },
    {
      "entropy": 0.5614024806767702,
      "epoch": 0.8703322552190532,
      "grad_norm": 1.65625,
      "learning_rate": 2.8919860627177703e-06,
      "loss": 0.5655649185180665,
      "mean_token_accuracy": 0.8296878322958946,
      "num_tokens": 29340755.0,
      "step": 2220
    },
    {
      "entropy": 0.5752765364944935,
      "epoch": 0.874252670783103,
      "grad_norm": 1.65625,
      "learning_rate": 2.8048780487804884e-06,
      "loss": 0.5743560314178466,
      "mean_token_accuracy": 0.8251941919326782,
      "num_tokens": 29474374.0,
      "step": 2230
    },
    {
      "entropy": 0.5772768836468458,
      "epoch": 0.8781730863471529,
      "grad_norm": 1.5234375,
      "learning_rate": 2.7177700348432056e-06,
      "loss": 0.5725958347320557,
      "mean_token_accuracy": 0.826750586181879,
      "num_tokens": 29607009.0,
      "step": 2240
    },
    {
      "entropy": 0.5837304752320052,
      "epoch": 0.8820935019112026,
      "grad_norm": 1.875,
      "learning_rate": 2.6306620209059237e-06,
      "loss": 0.588666296005249,
      "mean_token_accuracy": 0.8247058235108853,
      "num_tokens": 29738318.0,
      "step": 2250
    },
    {
      "entropy": 0.5540932007133961,
      "epoch": 0.8860139174752524,
      "grad_norm": 1.8359375,
      "learning_rate": 2.5435540069686414e-06,
      "loss": 0.5564557552337647,
      "mean_token_accuracy": 0.8303055338561535,
      "num_tokens": 29871530.0,
      "step": 2260
    },
    {
      "entropy": 0.5647958315908909,
      "epoch": 0.8899343330393021,
      "grad_norm": 1.90625,
      "learning_rate": 2.456445993031359e-06,
      "loss": 0.5653763771057129,
      "mean_token_accuracy": 0.8286120660603047,
      "num_tokens": 30004318.0,
      "step": 2270
    },
    {
      "entropy": 0.5644253201782703,
      "epoch": 0.8938547486033519,
      "grad_norm": 1.734375,
      "learning_rate": 2.3693379790940767e-06,
      "loss": 0.5634883880615235,
      "mean_token_accuracy": 0.8275229759514332,
      "num_tokens": 30136805.0,
      "step": 2280
    },
    {
      "entropy": 0.5744159776717425,
      "epoch": 0.8977751641674018,
      "grad_norm": 1.7421875,
      "learning_rate": 2.282229965156795e-06,
      "loss": 0.5716060161590576,
      "mean_token_accuracy": 0.82558439001441,
      "num_tokens": 30267921.0,
      "step": 2290
    },
    {
      "entropy": 0.5650353617966175,
      "epoch": 0.9016955797314515,
      "grad_norm": 1.6953125,
      "learning_rate": 2.1951219512195125e-06,
      "loss": 0.5693279266357422,
      "mean_token_accuracy": 0.8278473079204559,
      "num_tokens": 30402696.0,
      "step": 2300
    },
    {
      "entropy": 0.5965178959071636,
      "epoch": 0.9056159952955013,
      "grad_norm": 1.71875,
      "learning_rate": 2.10801393728223e-06,
      "loss": 0.6028908729553223,
      "mean_token_accuracy": 0.8189698673784733,
      "num_tokens": 30536391.0,
      "step": 2310
    },
    {
      "entropy": 0.601612939313054,
      "epoch": 0.9095364108595511,
      "grad_norm": 1.796875,
      "learning_rate": 2.020905923344948e-06,
      "loss": 0.6042677402496338,
      "mean_token_accuracy": 0.8197429470717907,
      "num_tokens": 30669243.0,
      "step": 2320
    },
    {
      "entropy": 0.5706041093915701,
      "epoch": 0.9134568264236009,
      "grad_norm": 1.8984375,
      "learning_rate": 1.9337979094076655e-06,
      "loss": 0.5693963527679443,
      "mean_token_accuracy": 0.8271980971097946,
      "num_tokens": 30800574.0,
      "step": 2330
    },
    {
      "entropy": 0.5640899043530225,
      "epoch": 0.9173772419876507,
      "grad_norm": 1.6484375,
      "learning_rate": 1.8466898954703836e-06,
      "loss": 0.5705442905426026,
      "mean_token_accuracy": 0.8264196418225765,
      "num_tokens": 30933674.0,
      "step": 2340
    },
    {
      "entropy": 0.5657230116426945,
      "epoch": 0.9212976575517005,
      "grad_norm": 1.609375,
      "learning_rate": 1.7595818815331012e-06,
      "loss": 0.570911979675293,
      "mean_token_accuracy": 0.8267294079065323,
      "num_tokens": 31065102.0,
      "step": 2350
    },
    {
      "entropy": 0.562781137228012,
      "epoch": 0.9252180731157502,
      "grad_norm": 1.796875,
      "learning_rate": 1.6724738675958191e-06,
      "loss": 0.5703943729400635,
      "mean_token_accuracy": 0.8268945284187794,
      "num_tokens": 31195291.0,
      "step": 2360
    },
    {
      "entropy": 0.5673436559736729,
      "epoch": 0.9291384886798001,
      "grad_norm": 1.609375,
      "learning_rate": 1.5853658536585368e-06,
      "loss": 0.5722959995269775,
      "mean_token_accuracy": 0.8286142982542515,
      "num_tokens": 31327802.0,
      "step": 2370
    },
    {
      "entropy": 0.5538178488612175,
      "epoch": 0.9330589042438499,
      "grad_norm": 1.859375,
      "learning_rate": 1.4982578397212545e-06,
      "loss": 0.5541855812072753,
      "mean_token_accuracy": 0.8326897613704205,
      "num_tokens": 31454999.0,
      "step": 2380
    },
    {
      "entropy": 0.5489677041769028,
      "epoch": 0.9369793198078996,
      "grad_norm": 1.5625,
      "learning_rate": 1.4111498257839723e-06,
      "loss": 0.5547789573669434,
      "mean_token_accuracy": 0.8312770992517471,
      "num_tokens": 31592105.0,
      "step": 2390
    },
    {
      "entropy": 0.5815275024622679,
      "epoch": 0.9408997353719494,
      "grad_norm": 1.8671875,
      "learning_rate": 1.32404181184669e-06,
      "loss": 0.589658784866333,
      "mean_token_accuracy": 0.8223450765013695,
      "num_tokens": 31720785.0,
      "step": 2400
    },
    {
      "entropy": 0.5828426022082567,
      "epoch": 0.9448201509359992,
      "grad_norm": 1.71875,
      "learning_rate": 1.2369337979094077e-06,
      "loss": 0.5806317806243897,
      "mean_token_accuracy": 0.8260537378489972,
      "num_tokens": 31852738.0,
      "step": 2410
    },
    {
      "entropy": 0.5780322037637233,
      "epoch": 0.948740566500049,
      "grad_norm": 1.671875,
      "learning_rate": 1.1498257839721255e-06,
      "loss": 0.573900556564331,
      "mean_token_accuracy": 0.8251914605498314,
      "num_tokens": 31985785.0,
      "step": 2420
    },
    {
      "entropy": 0.5561736188828945,
      "epoch": 0.9526609820640988,
      "grad_norm": 1.734375,
      "learning_rate": 1.0627177700348432e-06,
      "loss": 0.5585554599761963,
      "mean_token_accuracy": 0.8310537829995155,
      "num_tokens": 32116876.0,
      "step": 2430
    },
    {
      "entropy": 0.5960801500827074,
      "epoch": 0.9565813976281485,
      "grad_norm": 1.8515625,
      "learning_rate": 9.75609756097561e-07,
      "loss": 0.6023911476135254,
      "mean_token_accuracy": 0.8196251414716244,
      "num_tokens": 32251738.0,
      "step": 2440
    },
    {
      "entropy": 0.5717530060559511,
      "epoch": 0.9605018131921984,
      "grad_norm": 1.4609375,
      "learning_rate": 8.885017421602789e-07,
      "loss": 0.5710067749023438,
      "mean_token_accuracy": 0.8258462838828564,
      "num_tokens": 32386153.0,
      "step": 2450
    },
    {
      "entropy": 0.5853737752884627,
      "epoch": 0.9644222287562482,
      "grad_norm": 1.875,
      "learning_rate": 8.013937282229965e-07,
      "loss": 0.59298734664917,
      "mean_token_accuracy": 0.8215095445513725,
      "num_tokens": 32521937.0,
      "step": 2460
    },
    {
      "entropy": 0.5701922554522753,
      "epoch": 0.9683426443202979,
      "grad_norm": 1.7578125,
      "learning_rate": 7.142857142857143e-07,
      "loss": 0.5723202228546143,
      "mean_token_accuracy": 0.8279052451252937,
      "num_tokens": 32653667.0,
      "step": 2470
    },
    {
      "entropy": 0.5777422614395619,
      "epoch": 0.9722630598843477,
      "grad_norm": 1.6875,
      "learning_rate": 6.271777003484321e-07,
      "loss": 0.581749153137207,
      "mean_token_accuracy": 0.8253626257181168,
      "num_tokens": 32787099.0,
      "step": 2480
    },
    {
      "entropy": 0.5758198279887438,
      "epoch": 0.9761834754483976,
      "grad_norm": 1.6015625,
      "learning_rate": 5.400696864111499e-07,
      "loss": 0.5754932403564453,
      "mean_token_accuracy": 0.8263446964323521,
      "num_tokens": 32919330.0,
      "step": 2490
    },
    {
      "entropy": 0.5611678268760443,
      "epoch": 0.9801038910124473,
      "grad_norm": 1.703125,
      "learning_rate": 4.5296167247386764e-07,
      "loss": 0.5594530582427979,
      "mean_token_accuracy": 0.8305567562580108,
      "num_tokens": 33052864.0,
      "step": 2500
    },
    {
      "entropy": 0.5629599597305059,
      "epoch": 0.9840243065764971,
      "grad_norm": 1.7578125,
      "learning_rate": 3.6585365853658536e-07,
      "loss": 0.5671555042266846,
      "mean_token_accuracy": 0.8285158857703209,
      "num_tokens": 33187097.0,
      "step": 2510
    },
    {
      "entropy": 0.5846845716238022,
      "epoch": 0.9879447221405468,
      "grad_norm": 1.640625,
      "learning_rate": 2.7874564459930313e-07,
      "loss": 0.5863873481750488,
      "mean_token_accuracy": 0.8224268153309822,
      "num_tokens": 33324336.0,
      "step": 2520
    },
    {
      "entropy": 0.5745269916951656,
      "epoch": 0.9918651377045967,
      "grad_norm": 1.8046875,
      "learning_rate": 1.916376306620209e-07,
      "loss": 0.5790841579437256,
      "mean_token_accuracy": 0.8244781382381916,
      "num_tokens": 33457098.0,
      "step": 2530
    },
    {
      "entropy": 0.5654200822114944,
      "epoch": 0.9957855532686465,
      "grad_norm": 1.6640625,
      "learning_rate": 1.045296167247387e-07,
      "loss": 0.5639691829681397,
      "mean_token_accuracy": 0.8280748799443245,
      "num_tokens": 33590704.0,
      "step": 2540
    },
    {
      "entropy": 0.5788687650114298,
      "epoch": 0.9997059688326962,
      "grad_norm": 1.7265625,
      "learning_rate": 1.7421602787456446e-08,
      "loss": 0.5817639350891113,
      "mean_token_accuracy": 0.8252955496311187,
      "num_tokens": 33724217.0,
      "step": 2550
    }
  ],
  "logging_steps": 10,
  "max_steps": 2551,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.2075681983043174e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}