{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.977645305514158,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.816821652650833,
      "epoch": 0.029806259314456036,
      "grad_norm": 0.2774021029472351,
      "learning_rate": 1.7821782178217823e-05,
      "loss": 3.5324947357177736,
      "mean_token_accuracy": 0.372644878923893,
      "num_tokens": 34736.0,
      "step": 10
    },
    {
      "entropy": 2.968262720108032,
      "epoch": 0.05961251862891207,
      "grad_norm": 0.21174120903015137,
      "learning_rate": 3.762376237623763e-05,
      "loss": 3.449769973754883,
      "mean_token_accuracy": 0.3707723684608936,
      "num_tokens": 69389.0,
      "step": 20
    },
    {
      "entropy": 3.3959435164928435,
      "epoch": 0.08941877794336811,
      "grad_norm": 0.23471760749816895,
      "learning_rate": 5.742574257425742e-05,
      "loss": 3.2028858184814455,
      "mean_token_accuracy": 0.3946275718510151,
      "num_tokens": 105158.0,
      "step": 30
    },
    {
      "entropy": 2.953880912065506,
      "epoch": 0.11922503725782414,
      "grad_norm": 0.27773070335388184,
      "learning_rate": 7.722772277227723e-05,
      "loss": 2.948764419555664,
      "mean_token_accuracy": 0.4256064437329769,
      "num_tokens": 140056.0,
      "step": 40
    },
    {
      "entropy": 2.8933359563350676,
      "epoch": 0.14903129657228018,
      "grad_norm": 0.22581885755062103,
      "learning_rate": 9.702970297029703e-05,
      "loss": 2.9272085189819337,
      "mean_token_accuracy": 0.4304308548569679,
      "num_tokens": 175246.0,
      "step": 50
    },
    {
      "entropy": 2.904828208684921,
      "epoch": 0.17883755588673622,
      "grad_norm": 0.26275673508644104,
      "learning_rate": 0.00011683168316831683,
      "loss": 2.8511804580688476,
      "mean_token_accuracy": 0.4422623157501221,
      "num_tokens": 209396.0,
      "step": 60
    },
    {
      "entropy": 2.831114035844803,
      "epoch": 0.20864381520119224,
      "grad_norm": 0.26085418462753296,
      "learning_rate": 0.00013663366336633665,
      "loss": 2.807806205749512,
      "mean_token_accuracy": 0.44567784368991853,
      "num_tokens": 243099.0,
      "step": 70
    },
    {
      "entropy": 2.802034729719162,
      "epoch": 0.23845007451564829,
      "grad_norm": 0.27744612097740173,
      "learning_rate": 0.00015643564356435644,
      "loss": 2.797773742675781,
      "mean_token_accuracy": 0.4452659018337727,
      "num_tokens": 277092.0,
      "step": 80
    },
    {
      "entropy": 2.68398916721344,
      "epoch": 0.26825633383010433,
      "grad_norm": 0.2549440860748291,
      "learning_rate": 0.00017623762376237624,
      "loss": 2.667187309265137,
      "mean_token_accuracy": 0.4650408893823624,
      "num_tokens": 309973.0,
      "step": 90
    },
    {
      "entropy": 2.735769587755203,
      "epoch": 0.29806259314456035,
      "grad_norm": 0.2550068497657776,
      "learning_rate": 0.00019603960396039606,
      "loss": 2.7133861541748048,
      "mean_token_accuracy": 0.45147006064653394,
      "num_tokens": 343245.0,
      "step": 100
    },
    {
      "entropy": 2.7108645856380464,
      "epoch": 0.32786885245901637,
      "grad_norm": 0.2297118753194809,
      "learning_rate": 0.00019996161095037213,
      "loss": 2.703385925292969,
      "mean_token_accuracy": 0.4572730727493763,
      "num_tokens": 377504.0,
      "step": 110
    },
    {
      "entropy": 2.689351201057434,
      "epoch": 0.35767511177347244,
      "grad_norm": 0.22158177196979523,
      "learning_rate": 0.00019980570595006243,
      "loss": 2.661332702636719,
      "mean_token_accuracy": 0.46000284850597384,
      "num_tokens": 412358.0,
      "step": 120
    },
    {
      "entropy": 2.7110758304595945,
      "epoch": 0.38748137108792846,
      "grad_norm": 0.21041730046272278,
      "learning_rate": 0.00019953007256545487,
      "loss": 2.7031484603881837,
      "mean_token_accuracy": 0.4535141125321388,
      "num_tokens": 446263.0,
      "step": 130
    },
    {
      "entropy": 2.6860289573669434,
      "epoch": 0.4172876304023845,
      "grad_norm": 0.25008049607276917,
      "learning_rate": 0.0001991350414503882,
      "loss": 2.6748600006103516,
      "mean_token_accuracy": 0.4574957236647606,
      "num_tokens": 480416.0,
      "step": 140
    },
    {
      "entropy": 2.665454810857773,
      "epoch": 0.44709388971684055,
      "grad_norm": 0.22758014500141144,
      "learning_rate": 0.00019862108648996456,
      "loss": 2.6443389892578124,
      "mean_token_accuracy": 0.46647369116544724,
      "num_tokens": 516270.0,
      "step": 150
    },
    {
      "entropy": 2.6447577595710756,
      "epoch": 0.47690014903129657,
      "grad_norm": 0.22190025448799133,
      "learning_rate": 0.0001979888242320699,
      "loss": 2.62310791015625,
      "mean_token_accuracy": 0.46614872962236403,
      "num_tokens": 550064.0,
      "step": 160
    },
    {
      "entropy": 2.614084929227829,
      "epoch": 0.5067064083457526,
      "grad_norm": 0.21979136765003204,
      "learning_rate": 0.00019723901314775422,
      "loss": 2.6258529663085937,
      "mean_token_accuracy": 0.4651683732867241,
      "num_tokens": 586118.0,
      "step": 170
    },
    {
      "entropy": 2.596314066648483,
      "epoch": 0.5365126676602087,
      "grad_norm": 0.1965494453907013,
      "learning_rate": 0.00019637255272135864,
      "loss": 2.5659564971923827,
      "mean_token_accuracy": 0.4735427416861057,
      "num_tokens": 622428.0,
      "step": 180
    },
    {
      "entropy": 2.5857157707214355,
      "epoch": 0.5663189269746647,
      "grad_norm": 0.22657693922519684,
      "learning_rate": 0.00019539048237148077,
      "loss": 2.5758071899414063,
      "mean_token_accuracy": 0.4692229345440865,
      "num_tokens": 656709.0,
      "step": 190
    },
    {
      "entropy": 2.669593596458435,
      "epoch": 0.5961251862891207,
      "grad_norm": 0.20716820657253265,
      "learning_rate": 0.0001942939802040729,
      "loss": 2.6469730377197265,
      "mean_token_accuracy": 0.4617524668574333,
      "num_tokens": 690997.0,
      "step": 200
    },
    {
      "entropy": 2.5580246448516846,
      "epoch": 0.6259314456035767,
      "grad_norm": 0.19625356793403625,
      "learning_rate": 0.00019308436159916906,
      "loss": 2.561942291259766,
      "mean_token_accuracy": 0.4759944908320904,
      "num_tokens": 724913.0,
      "step": 210
    },
    {
      "entropy": 2.6400313675403595,
      "epoch": 0.6557377049180327,
      "grad_norm": 0.19937431812286377,
      "learning_rate": 0.00019176307763293563,
      "loss": 2.6140087127685545,
      "mean_token_accuracy": 0.4678444378077984,
      "num_tokens": 760650.0,
      "step": 220
    },
    {
      "entropy": 2.583085960149765,
      "epoch": 0.6855439642324889,
      "grad_norm": 0.1995794028043747,
      "learning_rate": 0.0001903317133369395,
      "loss": 2.5737834930419923,
      "mean_token_accuracy": 0.4739532507956028,
      "num_tokens": 793116.0,
      "step": 230
    },
    {
      "entropy": 2.5687488198280333,
      "epoch": 0.7153502235469449,
      "grad_norm": 0.21024885773658752,
      "learning_rate": 0.00018879198579672068,
      "loss": 2.5532905578613283,
      "mean_token_accuracy": 0.474745012819767,
      "num_tokens": 827475.0,
      "step": 240
    },
    {
      "entropy": 2.6337378442287447,
      "epoch": 0.7451564828614009,
      "grad_norm": 0.2064039707183838,
      "learning_rate": 0.00018714574209195153,
      "loss": 2.609242630004883,
      "mean_token_accuracy": 0.46810367628932,
      "num_tokens": 862585.0,
      "step": 250
    },
    {
      "entropy": 2.5550214648246765,
      "epoch": 0.7749627421758569,
      "grad_norm": 0.19126427173614502,
      "learning_rate": 0.00018539495708065303,
      "loss": 2.5519735336303713,
      "mean_token_accuracy": 0.4743352234363556,
      "num_tokens": 897650.0,
      "step": 260
    },
    {
      "entropy": 2.663702791929245,
      "epoch": 0.8047690014903129,
      "grad_norm": 0.2193765789270401,
      "learning_rate": 0.00018354173103012612,
      "loss": 2.616712188720703,
      "mean_token_accuracy": 0.46795518398284913,
      "num_tokens": 931572.0,
      "step": 270
    },
    {
      "entropy": 2.536553019285202,
      "epoch": 0.834575260804769,
      "grad_norm": 0.21028311550617218,
      "learning_rate": 0.0001815882870974404,
      "loss": 2.542274284362793,
      "mean_token_accuracy": 0.47298133969306944,
      "num_tokens": 965499.0,
      "step": 280
    },
    {
      "entropy": 2.4884628772735597,
      "epoch": 0.8643815201192251,
      "grad_norm": 0.1993735283613205,
      "learning_rate": 0.0001795369686625024,
      "loss": 2.4780502319335938,
      "mean_token_accuracy": 0.48176979944109916,
      "num_tokens": 1000080.0,
      "step": 290
    },
    {
      "entropy": 2.5663754105567933,
      "epoch": 0.8941877794336811,
      "grad_norm": 0.20134030282497406,
      "learning_rate": 0.00017739023651690264,
      "loss": 2.5468618392944338,
      "mean_token_accuracy": 0.4690897837281227,
      "num_tokens": 1033073.0,
      "step": 300
    },
    {
      "entropy": 2.593527537584305,
      "epoch": 0.9239940387481371,
      "grad_norm": 0.2191304713487625,
      "learning_rate": 0.00017515066591191432,
      "loss": 2.5609457015991213,
      "mean_token_accuracy": 0.47070475965738295,
      "num_tokens": 1065983.0,
      "step": 310
    },
    {
      "entropy": 2.5465128242969515,
      "epoch": 0.9538002980625931,
      "grad_norm": 0.2097494900226593,
      "learning_rate": 0.00017282094346918395,
      "loss": 2.5461181640625,
      "mean_token_accuracy": 0.47892046719789505,
      "num_tokens": 1099493.0,
      "step": 320
    },
    {
      "entropy": 2.5847067058086397,
      "epoch": 0.9836065573770492,
      "grad_norm": 0.21845956146717072,
      "learning_rate": 0.00017040386395782094,
      "loss": 2.5489078521728517,
      "mean_token_accuracy": 0.47342579662799833,
      "num_tokens": 1133498.0,
      "step": 330
    },
    {
      "entropy": 2.5168235364713167,
      "epoch": 1.0119225037257824,
      "grad_norm": 0.16953276097774506,
      "learning_rate": 0.00016790232694175163,
      "loss": 2.4647663116455076,
      "mean_token_accuracy": 0.4858273769679822,
      "num_tokens": 1165528.0,
      "step": 340
    },
    {
      "entropy": 2.412318927049637,
      "epoch": 1.0417287630402385,
      "grad_norm": 0.21309202909469604,
      "learning_rate": 0.00016531933330136,
      "loss": 2.4225946426391602,
      "mean_token_accuracy": 0.4951951213181019,
      "num_tokens": 1200350.0,
      "step": 350
    },
    {
      "entropy": 2.47369105219841,
      "epoch": 1.0715350223546944,
      "grad_norm": 0.2284303605556488,
      "learning_rate": 0.0001626579816335877,
      "loss": 2.43377685546875,
      "mean_token_accuracy": 0.4937798798084259,
      "num_tokens": 1234588.0,
      "step": 360
    },
    {
      "entropy": 2.506095367670059,
      "epoch": 1.1013412816691506,
      "grad_norm": 0.23142993450164795,
      "learning_rate": 0.0001599214645348118,
      "loss": 2.470874786376953,
      "mean_token_accuracy": 0.48065008595585823,
      "num_tokens": 1269639.0,
      "step": 370
    },
    {
      "entropy": 2.4424583077430726,
      "epoch": 1.1311475409836065,
      "grad_norm": 0.22852115333080292,
      "learning_rate": 0.0001571130647709596,
      "loss": 2.429966163635254,
      "mean_token_accuracy": 0.48865539133548735,
      "num_tokens": 1303891.0,
      "step": 380
    },
    {
      "entropy": 2.4099853932857513,
      "epoch": 1.1609538002980626,
      "grad_norm": 0.2168155312538147,
      "learning_rate": 0.00015423615133945454,
      "loss": 2.4010440826416017,
      "mean_token_accuracy": 0.49221119433641436,
      "num_tokens": 1338574.0,
      "step": 390
    },
    {
      "entropy": 2.45645033121109,
      "epoch": 1.1907600596125185,
      "grad_norm": 0.24261075258255005,
      "learning_rate": 0.0001512941754277176,
      "loss": 2.423468589782715,
      "mean_token_accuracy": 0.48970257937908174,
      "num_tokens": 1372137.0,
      "step": 400
    },
    {
      "entropy": 2.458910769224167,
      "epoch": 1.2205663189269746,
      "grad_norm": 0.22965072095394135,
      "learning_rate": 0.00014829066627307247,
      "loss": 2.433644104003906,
      "mean_token_accuracy": 0.4894383355975151,
      "num_tokens": 1406334.0,
      "step": 410
    },
    {
      "entropy": 2.3751603603363036,
      "epoch": 1.2503725782414308,
      "grad_norm": 0.23441405594348907,
      "learning_rate": 0.0001452292269290208,
      "loss": 2.359486770629883,
      "mean_token_accuracy": 0.4992436982691288,
      "num_tokens": 1440890.0,
      "step": 420
    },
    {
      "entropy": 2.4156671941280363,
      "epoch": 1.2801788375558867,
      "grad_norm": 0.22591960430145264,
      "learning_rate": 0.00014211352994296654,
      "loss": 2.3734575271606446,
      "mean_token_accuracy": 0.49520150795578954,
      "num_tokens": 1475175.0,
      "step": 430
    },
    {
      "entropy": 2.432661032676697,
      "epoch": 1.3099850968703428,
      "grad_norm": 0.22666800022125244,
      "learning_rate": 0.00013894731295057447,
      "loss": 2.4230409622192384,
      "mean_token_accuracy": 0.48941941633820535,
      "num_tokens": 1509313.0,
      "step": 440
    },
    {
      "entropy": 2.437654364109039,
      "epoch": 1.339791356184799,
      "grad_norm": 0.22223567962646484,
      "learning_rate": 0.00013573437419204764,
      "loss": 2.428188705444336,
      "mean_token_accuracy": 0.49134416282176974,
      "num_tokens": 1543845.0,
      "step": 450
    },
    {
      "entropy": 2.494117498397827,
      "epoch": 1.3695976154992549,
      "grad_norm": 0.2305568903684616,
      "learning_rate": 0.00013247856795570295,
      "loss": 2.470295524597168,
      "mean_token_accuracy": 0.4862105637788773,
      "num_tokens": 1578471.0,
      "step": 460
    },
    {
      "entropy": 2.3662930488586427,
      "epoch": 1.3994038748137108,
      "grad_norm": 0.22210069000720978,
      "learning_rate": 0.00012918379995431064,
      "loss": 2.361283874511719,
      "mean_token_accuracy": 0.5016951121389865,
      "num_tokens": 1613175.0,
      "step": 470
    },
    {
      "entropy": 2.4781057953834535,
      "epoch": 1.4292101341281669,
      "grad_norm": 0.24383410811424255,
      "learning_rate": 0.00012585402263974382,
      "loss": 2.447409248352051,
      "mean_token_accuracy": 0.4863020680844784,
      "num_tokens": 1647739.0,
      "step": 480
    },
    {
      "entropy": 2.3419785887002944,
      "epoch": 1.459016393442623,
      "grad_norm": 0.23496992886066437,
      "learning_rate": 0.0001224932304615599,
      "loss": 2.311822509765625,
      "mean_token_accuracy": 0.5073645271360874,
      "num_tokens": 1682164.0,
      "step": 490
    },
    {
      "entropy": 2.443506282567978,
      "epoch": 1.488822652757079,
      "grad_norm": 0.2239231914281845,
      "learning_rate": 0.0001191054550752,
      "loss": 2.4512943267822265,
      "mean_token_accuracy": 0.48844458982348443,
      "num_tokens": 1717962.0,
      "step": 500
    },
    {
      "entropy": 2.499474382400513,
      "epoch": 1.518628912071535,
      "grad_norm": 0.22167044878005981,
      "learning_rate": 0.00011569476050555636,
      "loss": 2.4503448486328123,
      "mean_token_accuracy": 0.4871962830424309,
      "num_tokens": 1752763.0,
      "step": 510
    },
    {
      "entropy": 2.314634236693382,
      "epoch": 1.5484351713859912,
      "grad_norm": 0.2162325233221054,
      "learning_rate": 0.00011226523827170876,
      "loss": 2.3087753295898437,
      "mean_token_accuracy": 0.5063634611666202,
      "num_tokens": 1785480.0,
      "step": 520
    },
    {
      "entropy": 2.4157556533813476,
      "epoch": 1.578241430700447,
      "grad_norm": 0.21877898275852203,
      "learning_rate": 0.0001088210024786781,
      "loss": 2.397838592529297,
      "mean_token_accuracy": 0.4927471660077572,
      "num_tokens": 1820722.0,
      "step": 530
    },
    {
      "entropy": 2.41140421628952,
      "epoch": 1.608047690014903,
      "grad_norm": 0.22834156453609467,
      "learning_rate": 0.00010536618488208618,
      "loss": 2.364009666442871,
      "mean_token_accuracy": 0.4945222452282906,
      "num_tokens": 1853974.0,
      "step": 540
    },
    {
      "entropy": 2.3557627379894255,
      "epoch": 1.6378539493293591,
      "grad_norm": 0.21658001840114594,
      "learning_rate": 0.000101904929931641,
      "loss": 2.3448652267456054,
      "mean_token_accuracy": 0.5083381831645966,
      "num_tokens": 1888594.0,
      "step": 550
    },
    {
      "entropy": 2.410889434814453,
      "epoch": 1.6676602086438153,
      "grad_norm": 0.23936094343662262,
      "learning_rate": 9.844138979939437e-05,
      "loss": 2.386252593994141,
      "mean_token_accuracy": 0.4946880467236042,
      "num_tokens": 1921108.0,
      "step": 560
    },
    {
      "entropy": 2.3835754096508026,
      "epoch": 1.6974664679582712,
      "grad_norm": 0.21563449501991272,
      "learning_rate": 9.497971939873566e-05,
      "loss": 2.3662214279174805,
      "mean_token_accuracy": 0.49903169125318525,
      "num_tokens": 1954828.0,
      "step": 570
    },
    {
      "entropy": 2.408819016814232,
      "epoch": 1.7272727272727273,
      "grad_norm": 0.22929024696350098,
      "learning_rate": 9.152407140009684e-05,
      "loss": 2.3863504409790037,
      "mean_token_accuracy": 0.4925551608204842,
      "num_tokens": 1989058.0,
      "step": 580
    },
    {
      "entropy": 2.4016968578100206,
      "epoch": 1.7570789865871834,
      "grad_norm": 0.2191687524318695,
      "learning_rate": 8.807859124934838e-05,
      "loss": 2.359476661682129,
      "mean_token_accuracy": 0.499463415145874,
      "num_tokens": 2024403.0,
      "step": 590
    },
    {
      "entropy": 2.440983992815018,
      "epoch": 1.7868852459016393,
      "grad_norm": 0.23283228278160095,
      "learning_rate": 8.464741219486176e-05,
      "loss": 2.4185159683227537,
      "mean_token_accuracy": 0.4903191089630127,
      "num_tokens": 2058237.0,
      "step": 600
    },
    {
      "entropy": 2.4129498898983,
      "epoch": 1.8166915052160952,
      "grad_norm": 0.2509646415710449,
      "learning_rate": 8.123465032920416e-05,
      "loss": 2.3897493362426756,
      "mean_token_accuracy": 0.49621374905109406,
      "num_tokens": 2092574.0,
      "step": 610
    },
    {
      "entropy": 2.414195305109024,
      "epoch": 1.8464977645305514,
      "grad_norm": 0.21443071961402893,
      "learning_rate": 7.784439965141382e-05,
      "loss": 2.406075859069824,
      "mean_token_accuracy": 0.49434508606791494,
      "num_tokens": 2127183.0,
      "step": 620
    },
    {
      "entropy": 2.4428183257579805,
      "epoch": 1.8763040238450075,
      "grad_norm": 0.21860645711421967,
      "learning_rate": 7.448072715577885e-05,
      "loss": 2.432174491882324,
      "mean_token_accuracy": 0.48932347595691683,
      "num_tokens": 2163382.0,
      "step": 630
    },
    {
      "entropy": 2.391178274154663,
      "epoch": 1.9061102831594634,
      "grad_norm": 0.2208397090435028,
      "learning_rate": 7.114766795301137e-05,
      "loss": 2.365252685546875,
      "mean_token_accuracy": 0.5012997359037399,
      "num_tokens": 2199533.0,
      "step": 640
    },
    {
      "entropy": 2.3402610540390016,
      "epoch": 1.9359165424739195,
      "grad_norm": 0.2488349825143814,
      "learning_rate": 6.784922042966968e-05,
      "loss": 2.3241010665893556,
      "mean_token_accuracy": 0.5052258335053921,
      "num_tokens": 2231453.0,
      "step": 650
    },
    {
      "entropy": 2.3716456323862074,
      "epoch": 1.9657228017883757,
      "grad_norm": 0.23293083906173706,
      "learning_rate": 6.458934145163539e-05,
      "loss": 2.346171188354492,
      "mean_token_accuracy": 0.49953882917761805,
      "num_tokens": 2264687.0,
      "step": 660
    },
    {
      "entropy": 2.368867373466492,
      "epoch": 1.9955290611028316,
      "grad_norm": 0.22324605286121368,
      "learning_rate": 6.137194161739916e-05,
      "loss": 2.3555429458618162,
      "mean_token_accuracy": 0.5024733081459999,
      "num_tokens": 2299593.0,
      "step": 670
    },
    {
      "entropy": 2.3251365015381262,
      "epoch": 2.023845007451565,
      "grad_norm": 0.20680658519268036,
      "learning_rate": 5.820088056684954e-05,
      "loss": 2.2231800079345705,
      "mean_token_accuracy": 0.5198341866857127,
      "num_tokens": 2332348.0,
      "step": 680
    },
    {
      "entropy": 2.2193444639444353,
      "epoch": 2.0536512667660207,
      "grad_norm": 0.2577013373374939,
      "learning_rate": 5.507996235119258e-05,
      "loss": 2.1863868713378904,
      "mean_token_accuracy": 0.5221044242382049,
      "num_tokens": 2366585.0,
      "step": 690
    },
    {
      "entropy": 2.2398266583681106,
      "epoch": 2.083457526080477,
      "grad_norm": 0.2560054659843445,
      "learning_rate": 5.2012930869556455e-05,
      "loss": 2.221183967590332,
      "mean_token_accuracy": 0.5183231100440026,
      "num_tokens": 2399727.0,
      "step": 700
    },
    {
      "entropy": 2.3025130838155747,
      "epoch": 2.113263785394933,
      "grad_norm": 0.2595839202404022,
      "learning_rate": 4.900346537775513e-05,
      "loss": 2.2654613494873046,
      "mean_token_accuracy": 0.5095799826085567,
      "num_tokens": 2434902.0,
      "step": 710
    },
    {
      "entropy": 2.2881407916545866,
      "epoch": 2.143070044709389,
      "grad_norm": 0.25487998127937317,
      "learning_rate": 4.6055176074599384e-05,
      "loss": 2.268834686279297,
      "mean_token_accuracy": 0.5120450228452682,
      "num_tokens": 2469293.0,
      "step": 720
    },
    {
      "entropy": 2.2822261214256288,
      "epoch": 2.172876304023845,
      "grad_norm": 0.2520729601383209,
      "learning_rate": 4.317159977104962e-05,
      "loss": 2.2329111099243164,
      "mean_token_accuracy": 0.5223084904253483,
      "num_tokens": 2504450.0,
      "step": 730
    },
    {
      "entropy": 2.2816298216581345,
      "epoch": 2.202682563338301,
      "grad_norm": 0.2571701407432556,
      "learning_rate": 4.035619564740555e-05,
      "loss": 2.244193267822266,
      "mean_token_accuracy": 0.5198718920350075,
      "num_tokens": 2539794.0,
      "step": 740
    },
    {
      "entropy": 2.3046247869729997,
      "epoch": 2.232488822652757,
      "grad_norm": 0.269325315952301,
      "learning_rate": 3.7612341103622986e-05,
      "loss": 2.2744611740112304,
      "mean_token_accuracy": 0.5107228621840477,
      "num_tokens": 2573431.0,
      "step": 750
    },
    {
      "entropy": 2.2834656864404677,
      "epoch": 2.262295081967213,
      "grad_norm": 0.24246659874916077,
      "learning_rate": 3.494332770773558e-05,
      "loss": 2.2619428634643555,
      "mean_token_accuracy": 0.5113484673202038,
      "num_tokens": 2608407.0,
      "step": 760
    },
    {
      "entropy": 2.254980501532555,
      "epoch": 2.2921013412816693,
      "grad_norm": 0.21794407069683075,
      "learning_rate": 3.235235724724152e-05,
      "loss": 2.1981428146362303,
      "mean_token_accuracy": 0.5231673911213874,
      "num_tokens": 2643407.0,
      "step": 770
    },
    {
      "entropy": 2.2178608655929564,
      "epoch": 2.321907600596125,
      "grad_norm": 0.27703768014907837,
      "learning_rate": 2.984253788819242e-05,
      "loss": 2.1888092041015623,
      "mean_token_accuracy": 0.5267924278974533,
      "num_tokens": 2676109.0,
      "step": 780
    },
    {
      "entropy": 2.2608650267124175,
      "epoch": 2.351713859910581,
      "grad_norm": 0.2602367401123047,
      "learning_rate": 2.7416880446592088e-05,
      "loss": 2.2434055328369142,
      "mean_token_accuracy": 0.5198688216507434,
      "num_tokens": 2710017.0,
      "step": 790
    },
    {
      "entropy": 2.2819056689739226,
      "epoch": 2.381520119225037,
      "grad_norm": 0.28437793254852295,
      "learning_rate": 2.507829477657737e-05,
      "loss": 2.2739871978759765,
      "mean_token_accuracy": 0.5141860462725163,
      "num_tokens": 2744602.0,
      "step": 800
    },
    {
      "entropy": 2.2509973287582397,
      "epoch": 2.4113263785394934,
      "grad_norm": 0.26067861914634705,
      "learning_rate": 2.2829586279714698e-05,
      "loss": 2.2272113800048827,
      "mean_token_accuracy": 0.5193138346076012,
      "num_tokens": 2778900.0,
      "step": 810
    },
    {
      "entropy": 2.2565877825021743,
      "epoch": 2.4411326378539493,
      "grad_norm": 0.2630312144756317,
      "learning_rate": 2.067345253959938e-05,
      "loss": 2.200933074951172,
      "mean_token_accuracy": 0.5254437021911145,
      "num_tokens": 2813698.0,
      "step": 820
    },
    {
      "entropy": 2.281704491376877,
      "epoch": 2.470938897168405,
      "grad_norm": 0.2686123251914978,
      "learning_rate": 1.8612480085794802e-05,
      "loss": 2.2527828216552734,
      "mean_token_accuracy": 0.5161294609308242,
      "num_tokens": 2847215.0,
      "step": 830
    },
    {
      "entropy": 2.2780510157346727,
      "epoch": 2.5007451564828616,
      "grad_norm": 0.2607590854167938,
      "learning_rate": 1.6649141290993753e-05,
      "loss": 2.2597888946533202,
      "mean_token_accuracy": 0.5173437915742397,
      "num_tokens": 2881469.0,
      "step": 840
    },
    {
      "entropy": 2.261881074309349,
      "epoch": 2.5305514157973175,
      "grad_norm": 0.2662050724029541,
      "learning_rate": 1.4785791405123994e-05,
      "loss": 2.2056772232055666,
      "mean_token_accuracy": 0.5222107157111168,
      "num_tokens": 2915844.0,
      "step": 850
    },
    {
      "entropy": 2.2895468354225157,
      "epoch": 2.5603576751117734,
      "grad_norm": 0.26619791984558105,
      "learning_rate": 1.3024665729956053e-05,
      "loss": 2.2531227111816405,
      "mean_token_accuracy": 0.5150591425597668,
      "num_tokens": 2950580.0,
      "step": 860
    },
    {
      "entropy": 2.269976469874382,
      "epoch": 2.5901639344262293,
      "grad_norm": 0.2605002820491791,
      "learning_rate": 1.1367876937602473e-05,
      "loss": 2.227630615234375,
      "mean_token_accuracy": 0.52215935587883,
      "num_tokens": 2985115.0,
      "step": 870
    },
    {
      "entropy": 2.296455204486847,
      "epoch": 2.6199701937406856,
      "grad_norm": 0.23708687722682953,
      "learning_rate": 9.817412536125448e-06,
      "loss": 2.2528875350952147,
      "mean_token_accuracy": 0.5143393971025944,
      "num_tokens": 3019573.0,
      "step": 880
    },
    {
      "entropy": 2.288746526837349,
      "epoch": 2.6497764530551415,
      "grad_norm": 0.246008038520813,
      "learning_rate": 8.375132485293158e-06,
      "loss": 2.258831787109375,
      "mean_token_accuracy": 0.5151149004697799,
      "num_tokens": 3054980.0,
      "step": 890
    },
    {
      "entropy": 2.2559049487113954,
      "epoch": 2.679582712369598,
      "grad_norm": 0.2760912775993347,
      "learning_rate": 7.042766965344782e-06,
      "loss": 2.2129356384277346,
      "mean_token_accuracy": 0.5217158675193787,
      "num_tokens": 3087334.0,
      "step": 900
    },
    {
      "entropy": 2.244227999448776,
      "epoch": 2.709388971684054,
      "grad_norm": 0.2549956142902374,
      "learning_rate": 5.821914301440956e-06,
      "loss": 2.2193668365478514,
      "mean_token_accuracy": 0.5202659279108047,
      "num_tokens": 3122143.0,
      "step": 910
    },
    {
      "entropy": 2.2112534821033476,
      "epoch": 2.7391952309985097,
      "grad_norm": 0.2654170095920563,
      "learning_rate": 4.71403904628962e-06,
      "loss": 2.1555202484130858,
      "mean_token_accuracy": 0.5308095216751099,
      "num_tokens": 3155863.0,
      "step": 920
    },
    {
      "entropy": 2.2209642380476,
      "epoch": 2.7690014903129656,
      "grad_norm": 0.26565906405448914,
      "learning_rate": 3.7204702232470967e-06,
      "loss": 2.191381645202637,
      "mean_token_accuracy": 0.5267351999878883,
      "num_tokens": 3190963.0,
      "step": 930
    },
    {
      "entropy": 2.2214305847883224,
      "epoch": 2.7988077496274215,
      "grad_norm": 0.2883058488368988,
      "learning_rate": 2.842399732002277e-06,
      "loss": 2.1862348556518554,
      "mean_token_accuracy": 0.5217300400137901,
      "num_tokens": 3223632.0,
      "step": 940
    },
    {
      "entropy": 2.2640267431735994,
      "epoch": 2.828614008941878,
      "grad_norm": 0.2695910930633545,
      "learning_rate": 2.0808809187563118e-06,
      "loss": 2.2413665771484377,
      "mean_token_accuracy": 0.5154105827212334,
      "num_tokens": 3257254.0,
      "step": 950
    },
    {
      "entropy": 2.2266465932130814,
      "epoch": 2.8584202682563338,
      "grad_norm": 0.2571880519390106,
      "learning_rate": 1.4368273126131427e-06,
      "loss": 2.2180950164794924,
      "mean_token_accuracy": 0.5234906762838364,
      "num_tokens": 3291063.0,
      "step": 960
    },
    {
      "entropy": 2.2674947023391723,
      "epoch": 2.88822652757079,
      "grad_norm": 0.25067442655563354,
      "learning_rate": 9.110115296965483e-07,
      "loss": 2.24371452331543,
      "mean_token_accuracy": 0.5170701034367085,
      "num_tokens": 3327288.0,
      "step": 970
    },
    {
      "entropy": 2.2928719133138658,
      "epoch": 2.918032786885246,
      "grad_norm": 0.270214706659317,
      "learning_rate": 5.040643463086303e-07,
      "loss": 2.264175796508789,
      "mean_token_accuracy": 0.5135901235044003,
      "num_tokens": 3362002.0,
      "step": 980
    },
    {
      "entropy": 2.235772040486336,
      "epoch": 2.947839046199702,
      "grad_norm": 0.24047863483428955,
      "learning_rate": 2.1647394224129093e-07,
      "loss": 2.2019769668579103,
      "mean_token_accuracy": 0.5209534049034119,
      "num_tokens": 3395413.0,
      "step": 990
    },
    {
      "entropy": 2.24095900952816,
      "epoch": 2.977645305514158,
      "grad_norm": 0.25486046075820923,
      "learning_rate": 4.858531514864151e-08,
      "loss": 2.183963966369629,
      "mean_token_accuracy": 0.5257976442575455,
      "num_tokens": 3430011.0,
      "step": 1000
    }
  ],
  "logging_steps": 10,
  "max_steps": 1008,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.094608146578739e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}