{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9997090485888857,
  "eval_steps": 100,
  "global_step": 859,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005819028222286878,
      "grad_norm": 4.729677200317383,
      "learning_rate": 1.1627906976744188e-06,
      "loss": 0.8687,
      "mean_token_accuracy": 0.8001353883684503,
      "step": 5
    },
    {
      "epoch": 0.011638056444573755,
      "grad_norm": 3.4830121994018555,
      "learning_rate": 2.3255813953488376e-06,
      "loss": 0.8575,
      "mean_token_accuracy": 0.7992099680535122,
      "step": 10
    },
    {
      "epoch": 0.017457084666860634,
      "grad_norm": 1.5148690938949585,
      "learning_rate": 3.4883720930232564e-06,
      "loss": 0.7863,
      "mean_token_accuracy": 0.8034960847656469,
      "step": 15
    },
    {
      "epoch": 0.02327611288914751,
      "grad_norm": 1.5541874170303345,
      "learning_rate": 4.651162790697675e-06,
      "loss": 0.7052,
      "mean_token_accuracy": 0.813427774762903,
      "step": 20
    },
    {
      "epoch": 0.02909514111143439,
      "grad_norm": 1.0066415071487427,
      "learning_rate": 5.8139534883720935e-06,
      "loss": 0.6611,
      "mean_token_accuracy": 0.8186242834655136,
      "step": 25
    },
    {
      "epoch": 0.03491416933372127,
      "grad_norm": 0.7388224601745605,
      "learning_rate": 6.976744186046513e-06,
      "loss": 0.6075,
      "mean_token_accuracy": 0.8315073685463057,
      "step": 30
    },
    {
      "epoch": 0.04073319755600815,
      "grad_norm": 0.5410726070404053,
      "learning_rate": 8.139534883720931e-06,
      "loss": 0.5962,
      "mean_token_accuracy": 0.832317557757456,
      "step": 35
    },
    {
      "epoch": 0.04655222577829502,
      "grad_norm": 0.53746497631073,
      "learning_rate": 9.30232558139535e-06,
      "loss": 0.5733,
      "mean_token_accuracy": 0.8373400698056097,
      "step": 40
    },
    {
      "epoch": 0.0523712540005819,
      "grad_norm": 0.512347400188446,
      "learning_rate": 1.046511627906977e-05,
      "loss": 0.5627,
      "mean_token_accuracy": 0.8384384618000557,
      "step": 45
    },
    {
      "epoch": 0.05819028222286878,
      "grad_norm": 0.5112928748130798,
      "learning_rate": 1.1627906976744187e-05,
      "loss": 0.5752,
      "mean_token_accuracy": 0.836557222748716,
      "step": 50
    },
    {
      "epoch": 0.06400931044515566,
      "grad_norm": 0.4823078513145447,
      "learning_rate": 1.2790697674418606e-05,
      "loss": 0.5546,
      "mean_token_accuracy": 0.8406492042617113,
      "step": 55
    },
    {
      "epoch": 0.06982833866744254,
      "grad_norm": 0.48803916573524475,
      "learning_rate": 1.3953488372093025e-05,
      "loss": 0.5552,
      "mean_token_accuracy": 0.8404829421801073,
      "step": 60
    },
    {
      "epoch": 0.07564736688972941,
      "grad_norm": 0.5077307224273682,
      "learning_rate": 1.5116279069767443e-05,
      "loss": 0.5537,
      "mean_token_accuracy": 0.8408575111390093,
      "step": 65
    },
    {
      "epoch": 0.0814663951120163,
      "grad_norm": 0.5025640726089478,
      "learning_rate": 1.6279069767441862e-05,
      "loss": 0.5324,
      "mean_token_accuracy": 0.8454815162857129,
      "step": 70
    },
    {
      "epoch": 0.08728542333430317,
      "grad_norm": 0.4802047610282898,
      "learning_rate": 1.744186046511628e-05,
      "loss": 0.5298,
      "mean_token_accuracy": 0.8467988429180437,
      "step": 75
    },
    {
      "epoch": 0.09310445155659004,
      "grad_norm": 0.495412141084671,
      "learning_rate": 1.86046511627907e-05,
      "loss": 0.5332,
      "mean_token_accuracy": 0.847131608146347,
      "step": 80
    },
    {
      "epoch": 0.09892347977887693,
      "grad_norm": 0.4867161512374878,
      "learning_rate": 1.9767441860465116e-05,
      "loss": 0.5315,
      "mean_token_accuracy": 0.8456074647350855,
      "step": 85
    },
    {
      "epoch": 0.1047425080011638,
      "grad_norm": 0.5120564699172974,
      "learning_rate": 1.999867863992634e-05,
      "loss": 0.5419,
      "mean_token_accuracy": 0.8431668032243257,
      "step": 90
    },
    {
      "epoch": 0.11056153622345069,
      "grad_norm": 0.5223878026008606,
      "learning_rate": 1.9993311213101313e-05,
      "loss": 0.5379,
      "mean_token_accuracy": 0.8446533035042771,
      "step": 95
    },
    {
      "epoch": 0.11638056444573756,
      "grad_norm": 0.4827958643436432,
      "learning_rate": 1.998381734913258e-05,
      "loss": 0.528,
      "mean_token_accuracy": 0.8463738754778392,
      "step": 100
    },
    {
      "epoch": 0.12219959266802444,
      "grad_norm": 0.49347934126853943,
      "learning_rate": 1.997020096822557e-05,
      "loss": 0.524,
      "mean_token_accuracy": 0.8469525801009745,
      "step": 105
    },
    {
      "epoch": 0.12801862089031132,
      "grad_norm": 0.4837639033794403,
      "learning_rate": 1.9952467692855043e-05,
      "loss": 0.5411,
      "mean_token_accuracy": 0.8443662513099179,
      "step": 110
    },
    {
      "epoch": 0.13383764911259818,
      "grad_norm": 0.44964534044265747,
      "learning_rate": 1.993062484544341e-05,
      "loss": 0.5122,
      "mean_token_accuracy": 0.8500057242518922,
      "step": 115
    },
    {
      "epoch": 0.13965667733488507,
      "grad_norm": 0.4763283133506775,
      "learning_rate": 1.990468144533722e-05,
      "loss": 0.5467,
      "mean_token_accuracy": 0.842521201317496,
      "step": 120
    },
    {
      "epoch": 0.14547570555717196,
      "grad_norm": 0.4805982708930969,
      "learning_rate": 1.9874648205082847e-05,
      "loss": 0.5267,
      "mean_token_accuracy": 0.8467973214319761,
      "step": 125
    },
    {
      "epoch": 0.15129473377945882,
      "grad_norm": 0.47071903944015503,
      "learning_rate": 1.9840537526003085e-05,
      "loss": 0.534,
      "mean_token_accuracy": 0.8436821750241809,
      "step": 130
    },
    {
      "epoch": 0.1571137620017457,
      "grad_norm": 0.49743327498435974,
      "learning_rate": 1.9802363493076392e-05,
      "loss": 0.5175,
      "mean_token_accuracy": 0.8488135271716413,
      "step": 135
    },
    {
      "epoch": 0.1629327902240326,
      "grad_norm": 0.46099644899368286,
      "learning_rate": 1.9760141869120917e-05,
      "loss": 0.5188,
      "mean_token_accuracy": 0.8494274636540204,
      "step": 140
    },
    {
      "epoch": 0.16875181844631945,
      "grad_norm": 0.43989109992980957,
      "learning_rate": 1.971389008828573e-05,
      "loss": 0.51,
      "mean_token_accuracy": 0.8514118628602676,
      "step": 145
    },
    {
      "epoch": 0.17457084666860634,
      "grad_norm": 0.44857239723205566,
      "learning_rate": 1.9663627248851903e-05,
      "loss": 0.505,
      "mean_token_accuracy": 0.8521678886513273,
      "step": 150
    },
    {
      "epoch": 0.18038987489089323,
      "grad_norm": 0.4434855580329895,
      "learning_rate": 1.9609374105346458e-05,
      "loss": 0.5348,
      "mean_token_accuracy": 0.8445110785073409,
      "step": 155
    },
    {
      "epoch": 0.18620890311318009,
      "grad_norm": 0.45866096019744873,
      "learning_rate": 1.9551153059972397e-05,
      "loss": 0.5387,
      "mean_token_accuracy": 0.8436002910675257,
      "step": 160
    },
    {
      "epoch": 0.19202793133546697,
      "grad_norm": 0.4627978801727295,
      "learning_rate": 1.948898815335841e-05,
      "loss": 0.5164,
      "mean_token_accuracy": 0.848416589531312,
      "step": 165
    },
    {
      "epoch": 0.19784695955775386,
      "grad_norm": 0.4659474790096283,
      "learning_rate": 1.9422905054631996e-05,
      "loss": 0.5129,
      "mean_token_accuracy": 0.8507672546269303,
      "step": 170
    },
    {
      "epoch": 0.20366598778004075,
      "grad_norm": 0.4798222482204437,
      "learning_rate": 1.9352931050820175e-05,
      "loss": 0.5173,
      "mean_token_accuracy": 0.8474816378041764,
      "step": 175
    },
    {
      "epoch": 0.2094850160023276,
      "grad_norm": 0.5056359171867371,
      "learning_rate": 1.9279095035582153e-05,
      "loss": 0.5128,
      "mean_token_accuracy": 0.8497545293493058,
      "step": 180
    },
    {
      "epoch": 0.2153040442246145,
      "grad_norm": 0.41810187697410583,
      "learning_rate": 1.9201427497278518e-05,
      "loss": 0.5105,
      "mean_token_accuracy": 0.8495568220795893,
      "step": 185
    },
    {
      "epoch": 0.22112307244690138,
      "grad_norm": 0.45786747336387634,
      "learning_rate": 1.911996050638202e-05,
      "loss": 0.508,
      "mean_token_accuracy": 0.8502620839044074,
      "step": 190
    },
    {
      "epoch": 0.22694210066918824,
      "grad_norm": 0.4433153569698334,
      "learning_rate": 1.9034727702235023e-05,
      "loss": 0.5251,
      "mean_token_accuracy": 0.8470169206232894,
      "step": 195
    },
    {
      "epoch": 0.23276112889147513,
      "grad_norm": 0.43002447485923767,
      "learning_rate": 1.8945764279159144e-05,
      "loss": 0.51,
      "mean_token_accuracy": 0.8515899255383441,
      "step": 200
    },
    {
      "epoch": 0.23858015711376201,
      "grad_norm": 0.43477922677993774,
      "learning_rate": 1.8853106971922833e-05,
      "loss": 0.5162,
      "mean_token_accuracy": 0.8489151326449708,
      "step": 205
    },
    {
      "epoch": 0.24439918533604887,
      "grad_norm": 0.4386807978153229,
      "learning_rate": 1.8756794040572834e-05,
      "loss": 0.528,
      "mean_token_accuracy": 0.8479445785479058,
      "step": 210
    },
    {
      "epoch": 0.25021821355833573,
      "grad_norm": 0.43842801451683044,
      "learning_rate": 1.8656865254635877e-05,
      "loss": 0.5007,
      "mean_token_accuracy": 0.852635679878258,
      "step": 215
    },
    {
      "epoch": 0.25603724178062265,
      "grad_norm": 0.46626922488212585,
      "learning_rate": 1.8553361876697025e-05,
      "loss": 0.5274,
      "mean_token_accuracy": 0.8479028685155747,
      "step": 220
    },
    {
      "epoch": 0.2618562700029095,
      "grad_norm": 0.4394637644290924,
      "learning_rate": 1.8446326645361542e-05,
      "loss": 0.5113,
      "mean_token_accuracy": 0.8496161991247855,
      "step": 225
    },
    {
      "epoch": 0.26767529822519637,
      "grad_norm": 0.46903783082962036,
      "learning_rate": 1.8335803757607274e-05,
      "loss": 0.5042,
      "mean_token_accuracy": 0.851652269825092,
      "step": 230
    },
    {
      "epoch": 0.2734943264474833,
      "grad_norm": 0.4322178363800049,
      "learning_rate": 1.822183885053483e-05,
      "loss": 0.5254,
      "mean_token_accuracy": 0.8445282817492037,
      "step": 235
    },
    {
      "epoch": 0.27931335466977014,
      "grad_norm": 0.4331161677837372,
      "learning_rate": 1.8104478982523117e-05,
      "loss": 0.5043,
      "mean_token_accuracy": 0.8515775238870337,
      "step": 240
    },
    {
      "epoch": 0.285132382892057,
      "grad_norm": 0.4396701753139496,
      "learning_rate": 1.7983772613798006e-05,
      "loss": 0.5003,
      "mean_token_accuracy": 0.8519097787097343,
      "step": 245
    },
    {
      "epoch": 0.2909514111143439,
      "grad_norm": 0.4399448335170746,
      "learning_rate": 1.7859769586422122e-05,
      "loss": 0.495,
      "mean_token_accuracy": 0.8534961898688037,
      "step": 250
    },
    {
      "epoch": 0.2967704393366308,
      "grad_norm": 0.4380975067615509,
      "learning_rate": 1.773252110371409e-05,
      "loss": 0.5117,
      "mean_token_accuracy": 0.849924917367521,
      "step": 255
    },
    {
      "epoch": 0.30258946755891764,
      "grad_norm": 0.4397619962692261,
      "learning_rate": 1.7602079709105644e-05,
      "loss": 0.4925,
      "mean_token_accuracy": 0.8539092350587474,
      "step": 260
    },
    {
      "epoch": 0.30840849578120455,
      "grad_norm": 0.4179295003414154,
      "learning_rate": 1.7468499264445405e-05,
      "loss": 0.4958,
      "mean_token_accuracy": 0.8536148051263105,
      "step": 265
    },
    {
      "epoch": 0.3142275240034914,
      "grad_norm": 0.4244683086872101,
      "learning_rate": 1.733183492775825e-05,
      "loss": 0.5052,
      "mean_token_accuracy": 0.8512700493372496,
      "step": 270
    },
    {
      "epoch": 0.32004655222577827,
      "grad_norm": 0.44008669257164,
      "learning_rate": 1.7192143130469466e-05,
      "loss": 0.5043,
      "mean_token_accuracy": 0.8509524474133208,
      "step": 275
    },
    {
      "epoch": 0.3258655804480652,
      "grad_norm": 0.42162829637527466,
      "learning_rate": 1.7049481554103107e-05,
      "loss": 0.5036,
      "mean_token_accuracy": 0.8519439016703739,
      "step": 280
    },
    {
      "epoch": 0.33168460867035204,
      "grad_norm": 0.44394752383232117,
      "learning_rate": 1.690390910646411e-05,
      "loss": 0.4962,
      "mean_token_accuracy": 0.8529925945929515,
      "step": 285
    },
    {
      "epoch": 0.3375036368926389,
      "grad_norm": 0.43867236375808716,
      "learning_rate": 1.6755485897314122e-05,
      "loss": 0.4986,
      "mean_token_accuracy": 0.8537370793513223,
      "step": 290
    },
    {
      "epoch": 0.3433226651149258,
      "grad_norm": 0.4371063709259033,
      "learning_rate": 1.6604273213550957e-05,
      "loss": 0.4994,
      "mean_token_accuracy": 0.853144893346005,
      "step": 295
    },
    {
      "epoch": 0.3491416933372127,
      "grad_norm": 0.4125002920627594,
      "learning_rate": 1.6450333493902046e-05,
      "loss": 0.5145,
      "mean_token_accuracy": 0.8495829518753062,
      "step": 300
    },
    {
      "epoch": 0.35496072155949954,
      "grad_norm": 0.422319233417511,
      "learning_rate": 1.6293730303142218e-05,
      "loss": 0.5098,
      "mean_token_accuracy": 0.8512014355945091,
      "step": 305
    },
    {
      "epoch": 0.36077974978178645,
      "grad_norm": 0.4452224671840668,
      "learning_rate": 1.6134528305846537e-05,
      "loss": 0.4868,
      "mean_token_accuracy": 0.856624218400263,
      "step": 310
    },
    {
      "epoch": 0.3665987780040733,
      "grad_norm": 0.4240838885307312,
      "learning_rate": 1.5972793239689038e-05,
      "loss": 0.4885,
      "mean_token_accuracy": 0.8560715988512444,
      "step": 315
    },
    {
      "epoch": 0.37241780622636017,
      "grad_norm": 0.4351670742034912,
      "learning_rate": 1.5808591888298314e-05,
      "loss": 0.4949,
      "mean_token_accuracy": 0.8541411151521825,
      "step": 320
    },
    {
      "epoch": 0.3782368344486471,
      "grad_norm": 0.41906318068504333,
      "learning_rate": 1.5641992053681213e-05,
      "loss": 0.5087,
      "mean_token_accuracy": 0.850579562045576,
      "step": 325
    },
    {
      "epoch": 0.38405586267093395,
      "grad_norm": 0.42045196890830994,
      "learning_rate": 1.5473062528226082e-05,
      "loss": 0.4853,
      "mean_token_accuracy": 0.8547947922524161,
      "step": 330
    },
    {
      "epoch": 0.3898748908932208,
      "grad_norm": 0.43161919713020325,
      "learning_rate": 1.530187306629697e-05,
      "loss": 0.5078,
      "mean_token_accuracy": 0.851508238713485,
      "step": 335
    },
    {
      "epoch": 0.3956939191155077,
      "grad_norm": 0.4057161509990692,
      "learning_rate": 1.5128494355430698e-05,
      "loss": 0.4848,
      "mean_token_accuracy": 0.8560923491214802,
      "step": 340
    },
    {
      "epoch": 0.4015129473377946,
      "grad_norm": 0.43893197178840637,
      "learning_rate": 1.4952997987148554e-05,
      "loss": 0.5114,
      "mean_token_accuracy": 0.849725545842043,
      "step": 345
    },
    {
      "epoch": 0.4073319755600815,
      "grad_norm": 0.432678759098053,
      "learning_rate": 1.4775456427394732e-05,
      "loss": 0.5008,
      "mean_token_accuracy": 0.8532302583194186,
      "step": 350
    },
    {
      "epoch": 0.41315100378236835,
      "grad_norm": 0.40871691703796387,
      "learning_rate": 1.4595942986613696e-05,
      "loss": 0.4984,
      "mean_token_accuracy": 0.8532044769095325,
      "step": 355
    },
    {
      "epoch": 0.4189700320046552,
      "grad_norm": 0.43299028277397156,
      "learning_rate": 1.4414531789478841e-05,
      "loss": 0.5092,
      "mean_token_accuracy": 0.8504414872190281,
      "step": 360
    },
    {
      "epoch": 0.42478906022694213,
      "grad_norm": 0.41469907760620117,
      "learning_rate": 1.423129774428495e-05,
      "loss": 0.496,
      "mean_token_accuracy": 0.8539978662641772,
      "step": 365
    },
    {
      "epoch": 0.430608088449229,
      "grad_norm": 0.44283366203308105,
      "learning_rate": 1.4046316512017044e-05,
      "loss": 0.4887,
      "mean_token_accuracy": 0.8548189689797085,
      "step": 370
    },
    {
      "epoch": 0.43642711667151585,
      "grad_norm": 0.43634092807769775,
      "learning_rate": 1.3859664475108483e-05,
      "loss": 0.4949,
      "mean_token_accuracy": 0.8529859121625627,
      "step": 375
    },
    {
      "epoch": 0.44224614489380276,
      "grad_norm": 0.41822701692581177,
      "learning_rate": 1.3671418705901116e-05,
      "loss": 0.5025,
      "mean_token_accuracy": 0.8515156968159479,
      "step": 380
    },
    {
      "epoch": 0.4480651731160896,
      "grad_norm": 0.42008936405181885,
      "learning_rate": 1.3481656934820588e-05,
      "loss": 0.4957,
      "mean_token_accuracy": 0.8534913304740183,
      "step": 385
    },
    {
      "epoch": 0.4538842013383765,
      "grad_norm": 0.4045666754245758,
      "learning_rate": 1.32904575182799e-05,
      "loss": 0.4837,
      "mean_token_accuracy": 0.8559156045618721,
      "step": 390
    },
    {
      "epoch": 0.4597032295606634,
      "grad_norm": 0.4344455301761627,
      "learning_rate": 1.309789940632448e-05,
      "loss": 0.4863,
      "mean_token_accuracy": 0.8565342245889443,
      "step": 395
    },
    {
      "epoch": 0.46552225778295026,
      "grad_norm": 0.4040561616420746,
      "learning_rate": 1.2904062110032145e-05,
      "loss": 0.4888,
      "mean_token_accuracy": 0.8559227694031748,
      "step": 400
    },
    {
      "epoch": 0.4713412860052371,
      "grad_norm": 0.41261550784111023,
      "learning_rate": 1.270902566868139e-05,
      "loss": 0.4999,
      "mean_token_accuracy": 0.8528450147039587,
      "step": 405
    },
    {
      "epoch": 0.47716031422752403,
      "grad_norm": 0.41591835021972656,
      "learning_rate": 1.2512870616701571e-05,
      "loss": 0.4858,
      "mean_token_accuracy": 0.8563491876605844,
      "step": 410
    },
    {
      "epoch": 0.4829793424498109,
      "grad_norm": 0.4135805368423462,
      "learning_rate": 1.2315677950418645e-05,
      "loss": 0.491,
      "mean_token_accuracy": 0.8547786818046786,
      "step": 415
    },
    {
      "epoch": 0.48879837067209775,
      "grad_norm": 0.42353108525276184,
      "learning_rate": 1.2117529094610177e-05,
      "loss": 0.4889,
      "mean_token_accuracy": 0.8551664301669952,
      "step": 420
    },
    {
      "epoch": 0.49461739889438466,
      "grad_norm": 0.4162733554840088,
      "learning_rate": 1.1918505868883414e-05,
      "loss": 0.4838,
      "mean_token_accuracy": 0.856896010365834,
      "step": 425
    },
    {
      "epoch": 0.5004364271166715,
      "grad_norm": 0.41785794496536255,
      "learning_rate": 1.1718690453890356e-05,
      "loss": 0.4917,
      "mean_token_accuracy": 0.8560505977625705,
      "step": 430
    },
    {
      "epoch": 0.5062554553389584,
      "grad_norm": 0.44993114471435547,
      "learning_rate": 1.1518165357393716e-05,
      "loss": 0.5001,
      "mean_token_accuracy": 0.852639218024868,
      "step": 435
    },
    {
      "epoch": 0.5120744835612453,
      "grad_norm": 0.41796672344207764,
      "learning_rate": 1.1317013380197832e-05,
      "loss": 0.4823,
      "mean_token_accuracy": 0.8560310313992504,
      "step": 440
    },
    {
      "epoch": 0.5178935117835322,
      "grad_norm": 0.44025880098342896,
      "learning_rate": 1.111531758195856e-05,
      "loss": 0.48,
      "mean_token_accuracy": 0.8575471706652287,
      "step": 445
    },
    {
      "epoch": 0.523712540005819,
      "grad_norm": 0.4278118312358856,
      "learning_rate": 1.091316124688629e-05,
      "loss": 0.4971,
      "mean_token_accuracy": 0.852513281695311,
      "step": 450
    },
    {
      "epoch": 0.5295315682281059,
      "grad_norm": 0.42229601740837097,
      "learning_rate": 1.0710627849356246e-05,
      "loss": 0.4773,
      "mean_token_accuracy": 0.8587941342459156,
      "step": 455
    },
    {
      "epoch": 0.5353505964503927,
      "grad_norm": 0.4171890318393707,
      "learning_rate": 1.0507801019440235e-05,
      "loss": 0.4776,
      "mean_token_accuracy": 0.8578564343102109,
      "step": 460
    },
    {
      "epoch": 0.5411696246726797,
      "grad_norm": 0.41802269220352173,
      "learning_rate": 1.0304764508374152e-05,
      "loss": 0.4837,
      "mean_token_accuracy": 0.8558379914367457,
      "step": 465
    },
    {
      "epoch": 0.5469886528949666,
      "grad_norm": 0.43975040316581726,
      "learning_rate": 1.0101602153975398e-05,
      "loss": 0.4925,
      "mean_token_accuracy": 0.8547882250775738,
      "step": 470
    },
    {
      "epoch": 0.5528076811172534,
      "grad_norm": 0.3969772160053253,
      "learning_rate": 9.898397846024604e-06,
      "loss": 0.4725,
      "mean_token_accuracy": 0.8595373324490415,
      "step": 475
    },
    {
      "epoch": 0.5586267093395403,
      "grad_norm": 0.41007471084594727,
      "learning_rate": 9.69523549162585e-06,
      "loss": 0.4851,
      "mean_token_accuracy": 0.855284327692498,
      "step": 480
    },
    {
      "epoch": 0.5644457375618271,
      "grad_norm": 0.4255843758583069,
      "learning_rate": 9.492198980559766e-06,
      "loss": 0.4905,
      "mean_token_accuracy": 0.8558610098037495,
      "step": 485
    },
    {
      "epoch": 0.570264765784114,
      "grad_norm": 0.41996172070503235,
      "learning_rate": 9.289372150643759e-06,
      "loss": 0.4882,
      "mean_token_accuracy": 0.8572012931662899,
      "step": 490
    },
    {
      "epoch": 0.576083794006401,
      "grad_norm": 0.4162617623806,
      "learning_rate": 9.08683875311371e-06,
      "loss": 0.4932,
      "mean_token_accuracy": 0.8533011128757575,
      "step": 495
    },
    {
      "epoch": 0.5819028222286878,
      "grad_norm": 0.4218461811542511,
      "learning_rate": 8.884682418041443e-06,
      "loss": 0.4795,
      "mean_token_accuracy": 0.8570308869685629,
      "step": 500
    },
    {
      "epoch": 0.5877218504509747,
      "grad_norm": 0.41105833649635315,
      "learning_rate": 8.682986619802171e-06,
      "loss": 0.489,
      "mean_token_accuracy": 0.8549116723561353,
      "step": 505
    },
    {
      "epoch": 0.5935408786732616,
      "grad_norm": 0.3960190713405609,
      "learning_rate": 8.481834642606287e-06,
      "loss": 0.4804,
      "mean_token_accuracy": 0.8588179330924269,
      "step": 510
    },
    {
      "epoch": 0.5993599068955484,
      "grad_norm": 0.4174754321575165,
      "learning_rate": 8.281309546109649e-06,
      "loss": 0.4904,
      "mean_token_accuracy": 0.8553032555444215,
      "step": 515
    },
    {
      "epoch": 0.6051789351178353,
      "grad_norm": 0.41682904958724976,
      "learning_rate": 8.081494131116588e-06,
      "loss": 0.4797,
      "mean_token_accuracy": 0.85778211568845,
      "step": 520
    },
    {
      "epoch": 0.6109979633401222,
      "grad_norm": 0.39849600195884705,
      "learning_rate": 7.882470905389827e-06,
      "loss": 0.4983,
      "mean_token_accuracy": 0.8528337732738089,
      "step": 525
    },
    {
      "epoch": 0.6168169915624091,
      "grad_norm": 0.4062061309814453,
      "learning_rate": 7.684322049581359e-06,
      "loss": 0.4918,
      "mean_token_accuracy": 0.8539959811143045,
      "step": 530
    },
    {
      "epoch": 0.622636019784696,
      "grad_norm": 0.38275521993637085,
      "learning_rate": 7.487129383298433e-06,
      "loss": 0.4741,
      "mean_token_accuracy": 0.8592997100935363,
      "step": 535
    },
    {
      "epoch": 0.6284550480069828,
      "grad_norm": 0.3841171860694885,
      "learning_rate": 7.29097433131861e-06,
      "loss": 0.4721,
      "mean_token_accuracy": 0.859289038573644,
      "step": 540
    },
    {
      "epoch": 0.6342740762292697,
      "grad_norm": 0.41029033064842224,
      "learning_rate": 7.095937889967854e-06,
      "loss": 0.4688,
      "mean_token_accuracy": 0.8600678009977978,
      "step": 545
    },
    {
      "epoch": 0.6400931044515565,
      "grad_norm": 0.4171801209449768,
      "learning_rate": 6.90210059367552e-06,
      "loss": 0.4811,
      "mean_token_accuracy": 0.8564380862600244,
      "step": 550
    },
    {
      "epoch": 0.6459121326738435,
      "grad_norm": 0.41676321625709534,
      "learning_rate": 6.7095424817201035e-06,
      "loss": 0.4775,
      "mean_token_accuracy": 0.8579857639838829,
      "step": 555
    },
    {
      "epoch": 0.6517311608961304,
      "grad_norm": 0.40989112854003906,
      "learning_rate": 6.518343065179414e-06,
      "loss": 0.4916,
      "mean_token_accuracy": 0.8554774335695117,
      "step": 560
    },
    {
      "epoch": 0.6575501891184172,
      "grad_norm": 0.41312697529792786,
      "learning_rate": 6.328581294098887e-06,
      "loss": 0.4841,
      "mean_token_accuracy": 0.8565219937605406,
      "step": 565
    },
    {
      "epoch": 0.6633692173407041,
      "grad_norm": 0.40967249870300293,
      "learning_rate": 6.140335524891518e-06,
      "loss": 0.4806,
      "mean_token_accuracy": 0.8576536743376204,
      "step": 570
    },
    {
      "epoch": 0.669188245562991,
      "grad_norm": 0.39644986391067505,
      "learning_rate": 5.953683487982958e-06,
      "loss": 0.4846,
      "mean_token_accuracy": 0.8560946241771303,
      "step": 575
    },
    {
      "epoch": 0.6750072737852778,
      "grad_norm": 0.40402865409851074,
      "learning_rate": 5.768702255715053e-06,
      "loss": 0.4871,
      "mean_token_accuracy": 0.8555429979464371,
      "step": 580
    },
    {
      "epoch": 0.6808263020075648,
      "grad_norm": 0.4249478578567505,
      "learning_rate": 5.58546821052116e-06,
      "loss": 0.4776,
      "mean_token_accuracy": 0.8574913247699447,
      "step": 585
    },
    {
      "epoch": 0.6866453302298516,
      "grad_norm": 0.41155800223350525,
      "learning_rate": 5.404057013386306e-06,
      "loss": 0.4686,
      "mean_token_accuracy": 0.8604994771907524,
      "step": 590
    },
    {
      "epoch": 0.6924643584521385,
      "grad_norm": 0.4111210107803345,
      "learning_rate": 5.224543572605272e-06,
      "loss": 0.4888,
      "mean_token_accuracy": 0.8553543295438057,
      "step": 595
    },
    {
      "epoch": 0.6982833866744254,
      "grad_norm": 0.40666860342025757,
      "learning_rate": 5.047002012851447e-06,
      "loss": 0.4915,
      "mean_token_accuracy": 0.8555927078584643,
      "step": 600
    },
    {
      "epoch": 0.7041024148967122,
      "grad_norm": 0.41017675399780273,
      "learning_rate": 4.871505644569303e-06,
      "loss": 0.4837,
      "mean_token_accuracy": 0.8566939587596106,
      "step": 605
    },
    {
      "epoch": 0.7099214431189991,
      "grad_norm": 0.4147651195526123,
      "learning_rate": 4.698126933703031e-06,
      "loss": 0.483,
      "mean_token_accuracy": 0.8570666700504006,
      "step": 610
    },
    {
      "epoch": 0.715740471341286,
      "grad_norm": 0.40449991822242737,
      "learning_rate": 4.526937471773919e-06,
      "loss": 0.4689,
      "mean_token_accuracy": 0.860008777707927,
      "step": 615
    },
    {
      "epoch": 0.7215594995635729,
      "grad_norm": 0.40807604789733887,
      "learning_rate": 4.35800794631879e-06,
      "loss": 0.4826,
      "mean_token_accuracy": 0.8565822278909835,
      "step": 620
    },
    {
      "epoch": 0.7273785277858598,
      "grad_norm": 0.39830535650253296,
      "learning_rate": 4.191408111701693e-06,
      "loss": 0.4731,
      "mean_token_accuracy": 0.8593161088649728,
      "step": 625
    },
    {
      "epoch": 0.7331975560081466,
      "grad_norm": 0.3948498070240021,
      "learning_rate": 4.0272067603109646e-06,
      "loss": 0.4676,
      "mean_token_accuracy": 0.8615339302083884,
      "step": 630
    },
    {
      "epoch": 0.7390165842304335,
      "grad_norm": 0.3837417960166931,
      "learning_rate": 3.865471694153465e-06,
      "loss": 0.4638,
      "mean_token_accuracy": 0.8620474819313568,
      "step": 635
    },
    {
      "epoch": 0.7448356124527203,
      "grad_norm": 0.4020540118217468,
      "learning_rate": 3.706269696857785e-06,
      "loss": 0.4794,
      "mean_token_accuracy": 0.8580428959872017,
      "step": 640
    },
    {
      "epoch": 0.7506546406750073,
      "grad_norm": 0.3988768756389618,
      "learning_rate": 3.5496665060979563e-06,
      "loss": 0.4791,
      "mean_token_accuracy": 0.857951603270609,
      "step": 645
    },
    {
      "epoch": 0.7564736688972942,
      "grad_norm": 0.38841554522514343,
      "learning_rate": 3.395726786449044e-06,
      "loss": 0.4795,
      "mean_token_accuracy": 0.8580783028985468,
      "step": 650
    },
    {
      "epoch": 0.762292697119581,
      "grad_norm": 0.3969273865222931,
      "learning_rate": 3.244514102685881e-06,
      "loss": 0.4752,
      "mean_token_accuracy": 0.8586160745112773,
      "step": 655
    },
    {
      "epoch": 0.7681117253418679,
      "grad_norm": 0.40074124932289124,
      "learning_rate": 3.0960908935358904e-06,
      "loss": 0.4881,
      "mean_token_accuracy": 0.855172939595195,
      "step": 660
    },
    {
      "epoch": 0.7739307535641547,
      "grad_norm": 0.4185120463371277,
      "learning_rate": 2.9505184458968925e-06,
      "loss": 0.4689,
      "mean_token_accuracy": 0.8608836993671453,
      "step": 665
    },
    {
      "epoch": 0.7797497817864416,
      "grad_norm": 0.42662736773490906,
      "learning_rate": 2.807856869530534e-06,
      "loss": 0.4906,
      "mean_token_accuracy": 0.854303324792995,
      "step": 670
    },
    {
      "epoch": 0.7855688100087286,
      "grad_norm": 0.3838253915309906,
      "learning_rate": 2.6681650722417517e-06,
      "loss": 0.4739,
      "mean_token_accuracy": 0.859679923783397,
      "step": 675
    },
    {
      "epoch": 0.7913878382310154,
      "grad_norm": 0.4014156460762024,
      "learning_rate": 2.5315007355545983e-06,
      "loss": 0.4647,
      "mean_token_accuracy": 0.8613118683252459,
      "step": 680
    },
    {
      "epoch": 0.7972068664533023,
      "grad_norm": 0.41356369853019714,
      "learning_rate": 2.3979202908943576e-06,
      "loss": 0.4647,
      "mean_token_accuracy": 0.8605610072930571,
      "step": 685
    },
    {
      "epoch": 0.8030258946755892,
      "grad_norm": 0.4056306481361389,
      "learning_rate": 2.267478896285913e-06,
      "loss": 0.48,
      "mean_token_accuracy": 0.8584646023840529,
      "step": 690
    },
    {
      "epoch": 0.808844922897876,
      "grad_norm": 0.3952696919441223,
      "learning_rate": 2.140230413577882e-06,
      "loss": 0.4773,
      "mean_token_accuracy": 0.8575413077069948,
      "step": 695
    },
    {
      "epoch": 0.814663951120163,
      "grad_norm": 0.40555790066719055,
      "learning_rate": 2.0162273862019965e-06,
      "loss": 0.4822,
      "mean_token_accuracy": 0.8564873526871629,
      "step": 700
    },
    {
      "epoch": 0.8204829793424498,
      "grad_norm": 0.39178261160850525,
      "learning_rate": 1.8955210174768857e-06,
      "loss": 0.4665,
      "mean_token_accuracy": 0.8613343347514766,
      "step": 705
    },
    {
      "epoch": 0.8263020075647367,
      "grad_norm": 0.39173874258995056,
      "learning_rate": 1.7781611494651729e-06,
      "loss": 0.4736,
      "mean_token_accuracy": 0.8586594796262844,
      "step": 710
    },
    {
      "epoch": 0.8321210357870236,
      "grad_norm": 0.3980768620967865,
      "learning_rate": 1.6641962423927294e-06,
      "loss": 0.4742,
      "mean_token_accuracy": 0.8592263804363405,
      "step": 715
    },
    {
      "epoch": 0.8379400640093104,
      "grad_norm": 0.3935108780860901,
      "learning_rate": 1.5536733546384574e-06,
      "loss": 0.4861,
      "mean_token_accuracy": 0.8557533487541985,
      "step": 720
    },
    {
      "epoch": 0.8437590922315973,
      "grad_norm": 0.41904276609420776,
      "learning_rate": 1.4466381233029781e-06,
      "loss": 0.4806,
      "mean_token_accuracy": 0.8569492588154313,
      "step": 725
    },
    {
      "epoch": 0.8495781204538843,
      "grad_norm": 0.3956993818283081,
      "learning_rate": 1.3431347453641253e-06,
      "loss": 0.4772,
      "mean_token_accuracy": 0.8591885796899528,
      "step": 730
    },
    {
      "epoch": 0.8553971486761711,
      "grad_norm": 0.413522869348526,
      "learning_rate": 1.2432059594271684e-06,
      "loss": 0.4737,
      "mean_token_accuracy": 0.8588345902893689,
      "step": 735
    },
    {
      "epoch": 0.861216176898458,
      "grad_norm": 0.40007099509239197,
      "learning_rate": 1.1468930280771728e-06,
      "loss": 0.4713,
      "mean_token_accuracy": 0.8596527738293955,
      "step": 740
    },
    {
      "epoch": 0.8670352051207448,
      "grad_norm": 0.4135393798351288,
      "learning_rate": 1.0542357208408572e-06,
      "loss": 0.5017,
      "mean_token_accuracy": 0.8527636202781883,
      "step": 745
    },
    {
      "epoch": 0.8728542333430317,
      "grad_norm": 0.404310941696167,
      "learning_rate": 9.652722977649797e-07,
      "loss": 0.453,
      "mean_token_accuracy": 0.8641921613151137,
      "step": 750
    },
    {
      "epoch": 0.8786732615653186,
      "grad_norm": 0.3922938406467438,
      "learning_rate": 8.800394936179812e-07,
      "loss": 0.4654,
      "mean_token_accuracy": 0.8607802659264431,
      "step": 755
    },
    {
      "epoch": 0.8844922897876055,
      "grad_norm": 0.39852988719940186,
      "learning_rate": 7.985725027214841e-07,
      "loss": 0.4657,
      "mean_token_accuracy": 0.8610265562422169,
      "step": 760
    },
    {
      "epoch": 0.8903113180098924,
      "grad_norm": 0.4099702835083008,
      "learning_rate": 7.209049644178489e-07,
      "loss": 0.4645,
      "mean_token_accuracy": 0.8611578725473782,
      "step": 765
    },
    {
      "epoch": 0.8961303462321792,
      "grad_norm": 0.41344019770622253,
      "learning_rate": 6.470689491798232e-07,
      "loss": 0.4709,
      "mean_token_accuracy": 0.8593041269818004,
      "step": 770
    },
    {
      "epoch": 0.9019493744544661,
      "grad_norm": 0.38776156306266785,
      "learning_rate": 5.770949453680064e-07,
      "loss": 0.4719,
      "mean_token_accuracy": 0.8599895481770243,
      "step": 775
    },
    {
      "epoch": 0.907768402676753,
      "grad_norm": 0.39070042967796326,
      "learning_rate": 5.110118466415903e-07,
      "loss": 0.485,
      "mean_token_accuracy": 0.8541616461336453,
      "step": 780
    },
    {
      "epoch": 0.9135874308990398,
      "grad_norm": 0.39178237318992615,
      "learning_rate": 4.4884694002760297e-07,
      "loss": 0.4733,
      "mean_token_accuracy": 0.8594136677203281,
      "step": 785
    },
    {
      "epoch": 0.9194064591213268,
      "grad_norm": 0.3782612383365631,
      "learning_rate": 3.906258946535446e-07,
      "loss": 0.4762,
      "mean_token_accuracy": 0.8573084054617641,
      "step": 790
    },
    {
      "epoch": 0.9252254873436137,
      "grad_norm": 0.39697375893592834,
      "learning_rate": 3.363727511480974e-07,
      "loss": 0.4765,
      "mean_token_accuracy": 0.8577650392622143,
      "step": 795
    },
    {
      "epoch": 0.9310445155659005,
      "grad_norm": 0.3849934935569763,
      "learning_rate": 2.861099117142718e-07,
      "loss": 0.461,
      "mean_token_accuracy": 0.8616937448944618,
      "step": 800
    },
    {
      "epoch": 0.9368635437881874,
      "grad_norm": 0.4121275842189789,
      "learning_rate": 2.398581308790848e-07,
      "loss": 0.4827,
      "mean_token_accuracy": 0.857346642587542,
      "step": 805
    },
    {
      "epoch": 0.9426825720104742,
      "grad_norm": 0.39553239941596985,
      "learning_rate": 1.976365069236108e-07,
      "loss": 0.475,
      "mean_token_accuracy": 0.8578932364447682,
      "step": 810
    },
    {
      "epoch": 0.9485016002327611,
      "grad_norm": 0.3881802558898926,
      "learning_rate": 1.5946247399691638e-07,
      "loss": 0.4706,
      "mean_token_accuracy": 0.8601075021722997,
      "step": 815
    },
    {
      "epoch": 0.9543206284550481,
      "grad_norm": 0.3950815498828888,
      "learning_rate": 1.2535179491715453e-07,
      "loss": 0.4666,
      "mean_token_accuracy": 0.8601555329046816,
      "step": 820
    },
    {
      "epoch": 0.9601396566773349,
      "grad_norm": 0.4037560522556305,
      "learning_rate": 9.531855466278218e-08,
      "loss": 0.4636,
      "mean_token_accuracy": 0.8610238882101969,
      "step": 825
    },
    {
      "epoch": 0.9659586848996218,
      "grad_norm": 0.3907164931297302,
      "learning_rate": 6.937515455659128e-08,
      "loss": 0.4686,
      "mean_token_accuracy": 0.8608224325086367,
      "step": 830
    },
    {
      "epoch": 0.9717777131219086,
      "grad_norm": 0.41244828701019287,
      "learning_rate": 4.753230714496071e-08,
      "loss": 0.4832,
      "mean_token_accuracy": 0.8551789295502588,
      "step": 835
    },
    {
      "epoch": 0.9775967413441955,
      "grad_norm": 0.39467689394950867,
      "learning_rate": 2.9799031774427888e-08,
      "loss": 0.4877,
      "mean_token_accuracy": 0.8558349555197562,
      "step": 840
    },
    {
      "epoch": 0.9834157695664824,
      "grad_norm": 0.39321061968803406,
      "learning_rate": 1.6182650867421213e-08,
      "loss": 0.4634,
      "mean_token_accuracy": 0.8616077323763098,
      "step": 845
    },
    {
      "epoch": 0.9892347977887693,
      "grad_norm": 0.4110186696052551,
      "learning_rate": 6.688786898688682e-09,
      "loss": 0.4885,
      "mean_token_accuracy": 0.8555144290494805,
      "step": 850
    },
    {
      "epoch": 0.9950538260110562,
      "grad_norm": 0.388217955827713,
      "learning_rate": 1.321360073662792e-09,
      "loss": 0.4607,
      "mean_token_accuracy": 0.8616522037739971,
      "step": 855
    },
    {
      "epoch": 0.9997090485888857,
      "mean_token_accuracy": 0.8550544311901966,
      "step": 859,
      "total_flos": 42703635087360.0,
      "train_loss": 0.504338482067388,
      "train_runtime": 61692.734,
      "train_samples_per_second": 0.446,
      "train_steps_per_second": 0.014
    }
  ],
  "logging_steps": 5,
  "max_steps": 859,
  "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": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 42703635087360.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}