{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.09378087210451055,
  "eval_steps": 1000,
  "global_step": 7001,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00013395353821527004,
      "grad_norm": 4.978700160980225,
      "learning_rate": 8.035714285714286e-07,
      "loss": 1.4198334693908692,
      "step": 10
    },
    {
      "epoch": 0.0002679070764305401,
      "grad_norm": 3.940929412841797,
      "learning_rate": 1.6964285714285713e-06,
      "loss": 1.3955652236938476,
      "step": 20
    },
    {
      "epoch": 0.0004018606146458101,
      "grad_norm": 2.915323495864868,
      "learning_rate": 2.5892857142857148e-06,
      "loss": 1.3398881912231446,
      "step": 30
    },
    {
      "epoch": 0.0005358141528610801,
      "grad_norm": 2.512028932571411,
      "learning_rate": 3.4821428571428576e-06,
      "loss": 1.2956685066223144,
      "step": 40
    },
    {
      "epoch": 0.0006697676910763502,
      "grad_norm": 2.561004161834717,
      "learning_rate": 4.375e-06,
      "loss": 1.2670017242431642,
      "step": 50
    },
    {
      "epoch": 0.0008037212292916202,
      "grad_norm": 2.296826124191284,
      "learning_rate": 5.267857142857143e-06,
      "loss": 1.2354182243347167,
      "step": 60
    },
    {
      "epoch": 0.0009376747675068903,
      "grad_norm": 2.3733971118927,
      "learning_rate": 6.160714285714286e-06,
      "loss": 1.1916924476623536,
      "step": 70
    },
    {
      "epoch": 0.0010716283057221603,
      "grad_norm": 2.520475387573242,
      "learning_rate": 7.0535714285714286e-06,
      "loss": 1.1873969078063964,
      "step": 80
    },
    {
      "epoch": 0.0012055818439374302,
      "grad_norm": 2.4990739822387695,
      "learning_rate": 7.946428571428571e-06,
      "loss": 1.1627039909362793,
      "step": 90
    },
    {
      "epoch": 0.0013395353821527003,
      "grad_norm": 2.3932080268859863,
      "learning_rate": 8.839285714285716e-06,
      "loss": 1.1169233322143555,
      "step": 100
    },
    {
      "epoch": 0.0014734889203679704,
      "grad_norm": 2.6543712615966797,
      "learning_rate": 9.732142857142858e-06,
      "loss": 1.1316115379333496,
      "step": 110
    },
    {
      "epoch": 0.0016074424585832403,
      "grad_norm": 2.6352458000183105,
      "learning_rate": 1.0625e-05,
      "loss": 1.0918511390686034,
      "step": 120
    },
    {
      "epoch": 0.0017413959967985105,
      "grad_norm": 2.4628708362579346,
      "learning_rate": 1.1517857142857142e-05,
      "loss": 1.1151651382446288,
      "step": 130
    },
    {
      "epoch": 0.0018753495350137806,
      "grad_norm": 2.482074737548828,
      "learning_rate": 1.2410714285714287e-05,
      "loss": 1.094371223449707,
      "step": 140
    },
    {
      "epoch": 0.0020093030732290505,
      "grad_norm": 2.431954860687256,
      "learning_rate": 1.3303571428571429e-05,
      "loss": 1.0961938858032227,
      "step": 150
    },
    {
      "epoch": 0.0021432566114443206,
      "grad_norm": 2.8230552673339844,
      "learning_rate": 1.4196428571428572e-05,
      "loss": 1.083104705810547,
      "step": 160
    },
    {
      "epoch": 0.0022772101496595907,
      "grad_norm": 2.721494436264038,
      "learning_rate": 1.5089285714285714e-05,
      "loss": 1.0713379859924317,
      "step": 170
    },
    {
      "epoch": 0.0024111636878748604,
      "grad_norm": 2.3565499782562256,
      "learning_rate": 1.5982142857142857e-05,
      "loss": 1.099677276611328,
      "step": 180
    },
    {
      "epoch": 0.0025451172260901305,
      "grad_norm": 2.345555543899536,
      "learning_rate": 1.6875000000000004e-05,
      "loss": 1.0562498092651367,
      "step": 190
    },
    {
      "epoch": 0.0026790707643054006,
      "grad_norm": 2.381127119064331,
      "learning_rate": 1.7767857142857143e-05,
      "loss": 1.0502027511596679,
      "step": 200
    },
    {
      "epoch": 0.0028130243025206707,
      "grad_norm": 2.4133875370025635,
      "learning_rate": 1.8660714285714287e-05,
      "loss": 1.0681496620178224,
      "step": 210
    },
    {
      "epoch": 0.002946977840735941,
      "grad_norm": 2.5238466262817383,
      "learning_rate": 1.955357142857143e-05,
      "loss": 1.0440272331237792,
      "step": 220
    },
    {
      "epoch": 0.003080931378951211,
      "grad_norm": 2.7593183517456055,
      "learning_rate": 2.0446428571428573e-05,
      "loss": 1.0714456558227539,
      "step": 230
    },
    {
      "epoch": 0.0032148849171664807,
      "grad_norm": 2.5562100410461426,
      "learning_rate": 2.1339285714285713e-05,
      "loss": 1.0298255920410155,
      "step": 240
    },
    {
      "epoch": 0.003348838455381751,
      "grad_norm": 2.6205103397369385,
      "learning_rate": 2.2232142857142856e-05,
      "loss": 1.050998592376709,
      "step": 250
    },
    {
      "epoch": 0.003482791993597021,
      "grad_norm": 2.5965018272399902,
      "learning_rate": 2.3125000000000003e-05,
      "loss": 1.0283641815185547,
      "step": 260
    },
    {
      "epoch": 0.003616745531812291,
      "grad_norm": 2.3729209899902344,
      "learning_rate": 2.4017857142857146e-05,
      "loss": 1.0502397537231445,
      "step": 270
    },
    {
      "epoch": 0.003750699070027561,
      "grad_norm": 2.378077983856201,
      "learning_rate": 2.4910714285714286e-05,
      "loss": 1.0391980171203614,
      "step": 280
    },
    {
      "epoch": 0.003884652608242831,
      "grad_norm": 2.4107515811920166,
      "learning_rate": 2.5803571428571433e-05,
      "loss": 1.0300152778625489,
      "step": 290
    },
    {
      "epoch": 0.004018606146458101,
      "grad_norm": 2.4866559505462646,
      "learning_rate": 2.6696428571428573e-05,
      "loss": 1.0498000144958497,
      "step": 300
    },
    {
      "epoch": 0.004152559684673371,
      "grad_norm": 2.3568379878997803,
      "learning_rate": 2.7589285714285716e-05,
      "loss": 1.0401558876037598,
      "step": 310
    },
    {
      "epoch": 0.004286513222888641,
      "grad_norm": 2.456519603729248,
      "learning_rate": 2.8482142857142856e-05,
      "loss": 1.0105724334716797,
      "step": 320
    },
    {
      "epoch": 0.004420466761103911,
      "grad_norm": 2.8484723567962646,
      "learning_rate": 2.9375000000000003e-05,
      "loss": 1.0271486282348632,
      "step": 330
    },
    {
      "epoch": 0.004554420299319181,
      "grad_norm": 2.829651117324829,
      "learning_rate": 3.0267857142857142e-05,
      "loss": 1.0125882148742675,
      "step": 340
    },
    {
      "epoch": 0.0046883738375344515,
      "grad_norm": 1.9819118976593018,
      "learning_rate": 3.116071428571429e-05,
      "loss": 1.0134092330932618,
      "step": 350
    },
    {
      "epoch": 0.004822327375749721,
      "grad_norm": 2.5545337200164795,
      "learning_rate": 3.205357142857143e-05,
      "loss": 1.006214714050293,
      "step": 360
    },
    {
      "epoch": 0.004956280913964991,
      "grad_norm": 2.3971850872039795,
      "learning_rate": 3.2946428571428576e-05,
      "loss": 1.014259147644043,
      "step": 370
    },
    {
      "epoch": 0.005090234452180261,
      "grad_norm": 2.188464641571045,
      "learning_rate": 3.383928571428572e-05,
      "loss": 1.0164517402648925,
      "step": 380
    },
    {
      "epoch": 0.005224187990395531,
      "grad_norm": 2.607635021209717,
      "learning_rate": 3.4732142857142855e-05,
      "loss": 1.0006816864013672,
      "step": 390
    },
    {
      "epoch": 0.005358141528610801,
      "grad_norm": 1.9828215837478638,
      "learning_rate": 3.5625000000000005e-05,
      "loss": 0.994498348236084,
      "step": 400
    },
    {
      "epoch": 0.005492095066826071,
      "grad_norm": 2.0254552364349365,
      "learning_rate": 3.651785714285714e-05,
      "loss": 1.0143715858459472,
      "step": 410
    },
    {
      "epoch": 0.0056260486050413415,
      "grad_norm": 2.088826894760132,
      "learning_rate": 3.7410714285714285e-05,
      "loss": 1.003225040435791,
      "step": 420
    },
    {
      "epoch": 0.005760002143256612,
      "grad_norm": 2.4125075340270996,
      "learning_rate": 3.8303571428571435e-05,
      "loss": 0.9795364379882813,
      "step": 430
    },
    {
      "epoch": 0.005893955681471882,
      "grad_norm": 2.0874435901641846,
      "learning_rate": 3.919642857142857e-05,
      "loss": 0.9827790260314941,
      "step": 440
    },
    {
      "epoch": 0.006027909219687152,
      "grad_norm": 1.7993626594543457,
      "learning_rate": 4.0089285714285715e-05,
      "loss": 0.992791748046875,
      "step": 450
    },
    {
      "epoch": 0.006161862757902422,
      "grad_norm": 2.1616885662078857,
      "learning_rate": 4.098214285714286e-05,
      "loss": 0.9879550933837891,
      "step": 460
    },
    {
      "epoch": 0.006295816296117691,
      "grad_norm": 2.0187010765075684,
      "learning_rate": 4.1875e-05,
      "loss": 0.9812018394470214,
      "step": 470
    },
    {
      "epoch": 0.006429769834332961,
      "grad_norm": 2.0268852710723877,
      "learning_rate": 4.2767857142857145e-05,
      "loss": 1.017441749572754,
      "step": 480
    },
    {
      "epoch": 0.0065637233725482315,
      "grad_norm": 2.1466526985168457,
      "learning_rate": 4.366071428571429e-05,
      "loss": 0.9769362449645996,
      "step": 490
    },
    {
      "epoch": 0.006697676910763502,
      "grad_norm": 2.0106499195098877,
      "learning_rate": 4.455357142857143e-05,
      "loss": 0.9902260780334473,
      "step": 500
    },
    {
      "epoch": 0.006831630448978772,
      "grad_norm": 1.843783974647522,
      "learning_rate": 4.5446428571428574e-05,
      "loss": 1.0007816314697267,
      "step": 510
    },
    {
      "epoch": 0.006965583987194042,
      "grad_norm": 1.8174176216125488,
      "learning_rate": 4.633928571428572e-05,
      "loss": 0.9792203903198242,
      "step": 520
    },
    {
      "epoch": 0.007099537525409312,
      "grad_norm": 1.8151051998138428,
      "learning_rate": 4.723214285714286e-05,
      "loss": 0.992054557800293,
      "step": 530
    },
    {
      "epoch": 0.007233491063624582,
      "grad_norm": 1.7471234798431396,
      "learning_rate": 4.8125000000000004e-05,
      "loss": 0.9979231834411622,
      "step": 540
    },
    {
      "epoch": 0.007367444601839852,
      "grad_norm": 1.7692842483520508,
      "learning_rate": 4.901785714285714e-05,
      "loss": 0.9751591682434082,
      "step": 550
    },
    {
      "epoch": 0.007501398140055122,
      "grad_norm": 1.6787338256835938,
      "learning_rate": 4.991071428571429e-05,
      "loss": 0.9884813308715821,
      "step": 560
    },
    {
      "epoch": 0.0076353516782703915,
      "grad_norm": 1.7012407779693604,
      "learning_rate": 5.080357142857143e-05,
      "loss": 0.9855763435363769,
      "step": 570
    },
    {
      "epoch": 0.007769305216485662,
      "grad_norm": 1.915123701095581,
      "learning_rate": 5.169642857142858e-05,
      "loss": 0.9511891365051269,
      "step": 580
    },
    {
      "epoch": 0.007903258754700932,
      "grad_norm": 1.833243727684021,
      "learning_rate": 5.258928571428572e-05,
      "loss": 0.9984745025634766,
      "step": 590
    },
    {
      "epoch": 0.008037212292916202,
      "grad_norm": 1.450955867767334,
      "learning_rate": 5.348214285714286e-05,
      "loss": 0.9740289688110352,
      "step": 600
    },
    {
      "epoch": 0.008171165831131472,
      "grad_norm": 1.7209330797195435,
      "learning_rate": 5.4375e-05,
      "loss": 0.9547987937927246,
      "step": 610
    },
    {
      "epoch": 0.008305119369346742,
      "grad_norm": 1.7298052310943604,
      "learning_rate": 5.526785714285715e-05,
      "loss": 0.9855975151062012,
      "step": 620
    },
    {
      "epoch": 0.008439072907562012,
      "grad_norm": 1.6286202669143677,
      "learning_rate": 5.616071428571429e-05,
      "loss": 0.9657292366027832,
      "step": 630
    },
    {
      "epoch": 0.008573026445777282,
      "grad_norm": 2.2121224403381348,
      "learning_rate": 5.705357142857143e-05,
      "loss": 0.9826187133789063,
      "step": 640
    },
    {
      "epoch": 0.008706979983992552,
      "grad_norm": 1.4359034299850464,
      "learning_rate": 5.794642857142858e-05,
      "loss": 0.9880683898925782,
      "step": 650
    },
    {
      "epoch": 0.008840933522207823,
      "grad_norm": 1.8250757455825806,
      "learning_rate": 5.883928571428572e-05,
      "loss": 0.9728729248046875,
      "step": 660
    },
    {
      "epoch": 0.008974887060423093,
      "grad_norm": 1.6160730123519897,
      "learning_rate": 5.973214285714286e-05,
      "loss": 0.9673951148986817,
      "step": 670
    },
    {
      "epoch": 0.009108840598638363,
      "grad_norm": 1.623353362083435,
      "learning_rate": 6.0624999999999996e-05,
      "loss": 0.9910868644714356,
      "step": 680
    },
    {
      "epoch": 0.009242794136853633,
      "grad_norm": 1.7099947929382324,
      "learning_rate": 6.151785714285715e-05,
      "loss": 0.9758204460144043,
      "step": 690
    },
    {
      "epoch": 0.009376747675068903,
      "grad_norm": 1.6103285551071167,
      "learning_rate": 6.241071428571428e-05,
      "loss": 0.9683603286743164,
      "step": 700
    },
    {
      "epoch": 0.009510701213284173,
      "grad_norm": 1.703170895576477,
      "learning_rate": 6.330357142857143e-05,
      "loss": 0.959897232055664,
      "step": 710
    },
    {
      "epoch": 0.009644654751499442,
      "grad_norm": 1.662630319595337,
      "learning_rate": 6.419642857142858e-05,
      "loss": 0.9767607688903809,
      "step": 720
    },
    {
      "epoch": 0.009778608289714712,
      "grad_norm": 1.5833293199539185,
      "learning_rate": 6.508928571428571e-05,
      "loss": 0.9608363151550293,
      "step": 730
    },
    {
      "epoch": 0.009912561827929982,
      "grad_norm": 1.7040482759475708,
      "learning_rate": 6.598214285714286e-05,
      "loss": 0.9693010330200196,
      "step": 740
    },
    {
      "epoch": 0.010046515366145252,
      "grad_norm": 1.505820393562317,
      "learning_rate": 6.6875e-05,
      "loss": 0.9575628280639649,
      "step": 750
    },
    {
      "epoch": 0.010180468904360522,
      "grad_norm": 1.812294602394104,
      "learning_rate": 6.776785714285716e-05,
      "loss": 0.9544397354125976,
      "step": 760
    },
    {
      "epoch": 0.010314422442575792,
      "grad_norm": 1.6276874542236328,
      "learning_rate": 6.866071428571429e-05,
      "loss": 0.9317039489746094,
      "step": 770
    },
    {
      "epoch": 0.010448375980791062,
      "grad_norm": 1.6243042945861816,
      "learning_rate": 6.955357142857143e-05,
      "loss": 0.9851680755615234,
      "step": 780
    },
    {
      "epoch": 0.010582329519006332,
      "grad_norm": 1.5983399152755737,
      "learning_rate": 7.044642857142859e-05,
      "loss": 0.9748747825622559,
      "step": 790
    },
    {
      "epoch": 0.010716283057221603,
      "grad_norm": 1.7678436040878296,
      "learning_rate": 7.133928571428572e-05,
      "loss": 0.9590046882629395,
      "step": 800
    },
    {
      "epoch": 0.010850236595436873,
      "grad_norm": 1.5059410333633423,
      "learning_rate": 7.223214285714286e-05,
      "loss": 0.9823746681213379,
      "step": 810
    },
    {
      "epoch": 0.010984190133652143,
      "grad_norm": 1.4575616121292114,
      "learning_rate": 7.3125e-05,
      "loss": 0.9733425140380859,
      "step": 820
    },
    {
      "epoch": 0.011118143671867413,
      "grad_norm": 1.6164171695709229,
      "learning_rate": 7.401785714285715e-05,
      "loss": 0.927185344696045,
      "step": 830
    },
    {
      "epoch": 0.011252097210082683,
      "grad_norm": 1.512903094291687,
      "learning_rate": 7.491071428571429e-05,
      "loss": 0.9673010826110839,
      "step": 840
    },
    {
      "epoch": 0.011386050748297953,
      "grad_norm": 1.4509457349777222,
      "learning_rate": 7.580357142857143e-05,
      "loss": 0.9444644927978516,
      "step": 850
    },
    {
      "epoch": 0.011520004286513223,
      "grad_norm": 1.3291261196136475,
      "learning_rate": 7.669642857142858e-05,
      "loss": 0.9484622001647949,
      "step": 860
    },
    {
      "epoch": 0.011653957824728493,
      "grad_norm": 1.6370415687561035,
      "learning_rate": 7.758928571428572e-05,
      "loss": 0.9409455299377442,
      "step": 870
    },
    {
      "epoch": 0.011787911362943763,
      "grad_norm": 1.5580850839614868,
      "learning_rate": 7.848214285714286e-05,
      "loss": 0.9556178092956543,
      "step": 880
    },
    {
      "epoch": 0.011921864901159034,
      "grad_norm": 1.4624580144882202,
      "learning_rate": 7.9375e-05,
      "loss": 0.9679806709289551,
      "step": 890
    },
    {
      "epoch": 0.012055818439374304,
      "grad_norm": 1.5962079763412476,
      "learning_rate": 8.026785714285715e-05,
      "loss": 0.9726933479309082,
      "step": 900
    },
    {
      "epoch": 0.012189771977589574,
      "grad_norm": 1.3573912382125854,
      "learning_rate": 8.116071428571429e-05,
      "loss": 0.9682340621948242,
      "step": 910
    },
    {
      "epoch": 0.012323725515804844,
      "grad_norm": 1.5596022605895996,
      "learning_rate": 8.205357142857143e-05,
      "loss": 0.9629892349243164,
      "step": 920
    },
    {
      "epoch": 0.012457679054020112,
      "grad_norm": 1.730648398399353,
      "learning_rate": 8.294642857142858e-05,
      "loss": 0.9566517829895019,
      "step": 930
    },
    {
      "epoch": 0.012591632592235382,
      "grad_norm": 1.5507956743240356,
      "learning_rate": 8.383928571428572e-05,
      "loss": 0.9654373168945313,
      "step": 940
    },
    {
      "epoch": 0.012725586130450653,
      "grad_norm": 1.4160267114639282,
      "learning_rate": 8.473214285714286e-05,
      "loss": 0.9531091690063477,
      "step": 950
    },
    {
      "epoch": 0.012859539668665923,
      "grad_norm": 1.617100477218628,
      "learning_rate": 8.5625e-05,
      "loss": 0.9591686248779296,
      "step": 960
    },
    {
      "epoch": 0.012993493206881193,
      "grad_norm": 1.296432375907898,
      "learning_rate": 8.651785714285715e-05,
      "loss": 0.971707534790039,
      "step": 970
    },
    {
      "epoch": 0.013127446745096463,
      "grad_norm": 1.402993083000183,
      "learning_rate": 8.74107142857143e-05,
      "loss": 0.9584191322326661,
      "step": 980
    },
    {
      "epoch": 0.013261400283311733,
      "grad_norm": 1.4196568727493286,
      "learning_rate": 8.830357142857142e-05,
      "loss": 0.9672185897827148,
      "step": 990
    },
    {
      "epoch": 0.013395353821527003,
      "grad_norm": 1.3073877096176147,
      "learning_rate": 8.919642857142858e-05,
      "loss": 0.9563102722167969,
      "step": 1000
    },
    {
      "epoch": 0.013395353821527003,
      "eval_loss": 0.9335520267486572,
      "eval_runtime": 282.864,
      "eval_samples_per_second": 10.652,
      "eval_steps_per_second": 10.652,
      "step": 1000
    },
    {
      "epoch": 0.013529307359742273,
      "grad_norm": 1.2876602411270142,
      "learning_rate": 9.008928571428572e-05,
      "loss": 0.9841785430908203,
      "step": 1010
    },
    {
      "epoch": 0.013663260897957543,
      "grad_norm": 1.302293062210083,
      "learning_rate": 9.098214285714285e-05,
      "loss": 0.9624536514282227,
      "step": 1020
    },
    {
      "epoch": 0.013797214436172813,
      "grad_norm": 1.4643065929412842,
      "learning_rate": 9.1875e-05,
      "loss": 0.9604171752929688,
      "step": 1030
    },
    {
      "epoch": 0.013931167974388084,
      "grad_norm": 1.5588184595108032,
      "learning_rate": 9.276785714285715e-05,
      "loss": 0.9482678413391114,
      "step": 1040
    },
    {
      "epoch": 0.014065121512603354,
      "grad_norm": 1.3061965703964233,
      "learning_rate": 9.36607142857143e-05,
      "loss": 0.9968178749084473,
      "step": 1050
    },
    {
      "epoch": 0.014199075050818624,
      "grad_norm": 1.49628746509552,
      "learning_rate": 9.455357142857143e-05,
      "loss": 0.965756893157959,
      "step": 1060
    },
    {
      "epoch": 0.014333028589033894,
      "grad_norm": 1.5427523851394653,
      "learning_rate": 9.544642857142858e-05,
      "loss": 0.9401347160339355,
      "step": 1070
    },
    {
      "epoch": 0.014466982127249164,
      "grad_norm": 1.44222891330719,
      "learning_rate": 9.633928571428573e-05,
      "loss": 0.9700592994689942,
      "step": 1080
    },
    {
      "epoch": 0.014600935665464434,
      "grad_norm": 1.2180535793304443,
      "learning_rate": 9.723214285714286e-05,
      "loss": 0.9597712516784668,
      "step": 1090
    },
    {
      "epoch": 0.014734889203679704,
      "grad_norm": 1.4644688367843628,
      "learning_rate": 9.8125e-05,
      "loss": 0.9620055198669434,
      "step": 1100
    },
    {
      "epoch": 0.014868842741894974,
      "grad_norm": 1.4942972660064697,
      "learning_rate": 9.901785714285716e-05,
      "loss": 0.9615335464477539,
      "step": 1110
    },
    {
      "epoch": 0.015002796280110245,
      "grad_norm": 1.1068652868270874,
      "learning_rate": 9.991071428571429e-05,
      "loss": 0.9622197151184082,
      "step": 1120
    },
    {
      "epoch": 0.015136749818325513,
      "grad_norm": 1.283275842666626,
      "learning_rate": 0.00010080357142857144,
      "loss": 0.9707681655883789,
      "step": 1130
    },
    {
      "epoch": 0.015270703356540783,
      "grad_norm": 1.197508454322815,
      "learning_rate": 0.00010169642857142859,
      "loss": 0.9689389228820801,
      "step": 1140
    },
    {
      "epoch": 0.015404656894756053,
      "grad_norm": 1.2500603199005127,
      "learning_rate": 0.00010258928571428572,
      "loss": 0.9651185035705566,
      "step": 1150
    },
    {
      "epoch": 0.015538610432971323,
      "grad_norm": 1.6802270412445068,
      "learning_rate": 0.00010348214285714286,
      "loss": 0.9665432929992676,
      "step": 1160
    },
    {
      "epoch": 0.015672563971186593,
      "grad_norm": 1.4018360376358032,
      "learning_rate": 0.000104375,
      "loss": 0.9392754554748535,
      "step": 1170
    },
    {
      "epoch": 0.015806517509401864,
      "grad_norm": 1.5453063249588013,
      "learning_rate": 0.00010526785714285715,
      "loss": 0.9618069648742675,
      "step": 1180
    },
    {
      "epoch": 0.015940471047617134,
      "grad_norm": 1.2306983470916748,
      "learning_rate": 0.00010616071428571428,
      "loss": 0.940494441986084,
      "step": 1190
    },
    {
      "epoch": 0.016074424585832404,
      "grad_norm": 1.1923718452453613,
      "learning_rate": 0.00010705357142857145,
      "loss": 0.9696602821350098,
      "step": 1200
    },
    {
      "epoch": 0.016208378124047674,
      "grad_norm": 1.3687347173690796,
      "learning_rate": 0.00010794642857142858,
      "loss": 0.9562197685241699,
      "step": 1210
    },
    {
      "epoch": 0.016342331662262944,
      "grad_norm": 1.3242262601852417,
      "learning_rate": 0.00010883928571428572,
      "loss": 0.941429328918457,
      "step": 1220
    },
    {
      "epoch": 0.016476285200478214,
      "grad_norm": 1.2945473194122314,
      "learning_rate": 0.00010973214285714286,
      "loss": 0.9632730484008789,
      "step": 1230
    },
    {
      "epoch": 0.016610238738693484,
      "grad_norm": 1.1794791221618652,
      "learning_rate": 0.000110625,
      "loss": 0.9528386116027832,
      "step": 1240
    },
    {
      "epoch": 0.016744192276908754,
      "grad_norm": 1.261305570602417,
      "learning_rate": 0.00011151785714285714,
      "loss": 0.945485782623291,
      "step": 1250
    },
    {
      "epoch": 0.016878145815124024,
      "grad_norm": 1.310412883758545,
      "learning_rate": 0.00011241071428571428,
      "loss": 0.9500171661376953,
      "step": 1260
    },
    {
      "epoch": 0.017012099353339295,
      "grad_norm": 1.4417651891708374,
      "learning_rate": 0.00011330357142857145,
      "loss": 0.965580940246582,
      "step": 1270
    },
    {
      "epoch": 0.017146052891554565,
      "grad_norm": 1.2227723598480225,
      "learning_rate": 0.00011419642857142858,
      "loss": 0.9576460838317871,
      "step": 1280
    },
    {
      "epoch": 0.017280006429769835,
      "grad_norm": 1.278008222579956,
      "learning_rate": 0.00011508928571428572,
      "loss": 0.9216486930847168,
      "step": 1290
    },
    {
      "epoch": 0.017413959967985105,
      "grad_norm": 1.2406392097473145,
      "learning_rate": 0.00011598214285714287,
      "loss": 0.9882441520690918,
      "step": 1300
    },
    {
      "epoch": 0.017547913506200375,
      "grad_norm": 1.2370669841766357,
      "learning_rate": 0.000116875,
      "loss": 0.9419057846069336,
      "step": 1310
    },
    {
      "epoch": 0.017681867044415645,
      "grad_norm": 1.1850550174713135,
      "learning_rate": 0.00011776785714285714,
      "loss": 0.976494026184082,
      "step": 1320
    },
    {
      "epoch": 0.017815820582630915,
      "grad_norm": 1.1672782897949219,
      "learning_rate": 0.00011866071428571428,
      "loss": 0.9599658012390136,
      "step": 1330
    },
    {
      "epoch": 0.017949774120846185,
      "grad_norm": 1.2031351327896118,
      "learning_rate": 0.00011955357142857144,
      "loss": 0.9466784477233887,
      "step": 1340
    },
    {
      "epoch": 0.018083727659061456,
      "grad_norm": 1.423079013824463,
      "learning_rate": 0.00012044642857142858,
      "loss": 0.9467257499694824,
      "step": 1350
    },
    {
      "epoch": 0.018217681197276726,
      "grad_norm": 1.3132128715515137,
      "learning_rate": 0.00012133928571428572,
      "loss": 0.95529203414917,
      "step": 1360
    },
    {
      "epoch": 0.018351634735491996,
      "grad_norm": 1.3859386444091797,
      "learning_rate": 0.00012223214285714287,
      "loss": 0.9389223098754883,
      "step": 1370
    },
    {
      "epoch": 0.018485588273707266,
      "grad_norm": 1.1685659885406494,
      "learning_rate": 0.000123125,
      "loss": 0.94387845993042,
      "step": 1380
    },
    {
      "epoch": 0.018619541811922536,
      "grad_norm": 1.2127257585525513,
      "learning_rate": 0.00012401785714285715,
      "loss": 0.918062400817871,
      "step": 1390
    },
    {
      "epoch": 0.018753495350137806,
      "grad_norm": 1.1995147466659546,
      "learning_rate": 0.00012491071428571428,
      "loss": 0.9396192550659179,
      "step": 1400
    },
    {
      "epoch": 0.018887448888353076,
      "grad_norm": 1.2880109548568726,
      "learning_rate": 0.00012580357142857144,
      "loss": 0.9649562835693359,
      "step": 1410
    },
    {
      "epoch": 0.019021402426568346,
      "grad_norm": 1.2551807165145874,
      "learning_rate": 0.00012669642857142857,
      "loss": 0.9511116027832032,
      "step": 1420
    },
    {
      "epoch": 0.019155355964783616,
      "grad_norm": 1.184615135192871,
      "learning_rate": 0.00012758928571428573,
      "loss": 0.9791507720947266,
      "step": 1430
    },
    {
      "epoch": 0.019289309502998883,
      "grad_norm": 1.1787488460540771,
      "learning_rate": 0.00012848214285714286,
      "loss": 0.9635199546813965,
      "step": 1440
    },
    {
      "epoch": 0.019423263041214153,
      "grad_norm": 1.0999921560287476,
      "learning_rate": 0.00012937500000000001,
      "loss": 0.9462879180908204,
      "step": 1450
    },
    {
      "epoch": 0.019557216579429423,
      "grad_norm": 1.3516278266906738,
      "learning_rate": 0.00013026785714285714,
      "loss": 0.9511301040649414,
      "step": 1460
    },
    {
      "epoch": 0.019691170117644693,
      "grad_norm": 1.2714283466339111,
      "learning_rate": 0.00013116071428571427,
      "loss": 0.934202003479004,
      "step": 1470
    },
    {
      "epoch": 0.019825123655859964,
      "grad_norm": 1.1116617918014526,
      "learning_rate": 0.00013205357142857143,
      "loss": 0.9517221450805664,
      "step": 1480
    },
    {
      "epoch": 0.019959077194075234,
      "grad_norm": 1.1716386079788208,
      "learning_rate": 0.0001329464285714286,
      "loss": 0.9329569816589356,
      "step": 1490
    },
    {
      "epoch": 0.020093030732290504,
      "grad_norm": 1.1489286422729492,
      "learning_rate": 0.00013383928571428572,
      "loss": 0.9706562995910645,
      "step": 1500
    },
    {
      "epoch": 0.020226984270505774,
      "grad_norm": 1.3109049797058105,
      "learning_rate": 0.00013473214285714287,
      "loss": 0.9514205932617188,
      "step": 1510
    },
    {
      "epoch": 0.020360937808721044,
      "grad_norm": 1.218209147453308,
      "learning_rate": 0.000135625,
      "loss": 0.9647247314453125,
      "step": 1520
    },
    {
      "epoch": 0.020494891346936314,
      "grad_norm": 1.1620960235595703,
      "learning_rate": 0.00013651785714285713,
      "loss": 0.9633931159973145,
      "step": 1530
    },
    {
      "epoch": 0.020628844885151584,
      "grad_norm": 1.1143990755081177,
      "learning_rate": 0.0001374107142857143,
      "loss": 0.9660541534423828,
      "step": 1540
    },
    {
      "epoch": 0.020762798423366854,
      "grad_norm": 1.12709641456604,
      "learning_rate": 0.00013830357142857145,
      "loss": 0.9445898056030273,
      "step": 1550
    },
    {
      "epoch": 0.020896751961582125,
      "grad_norm": 1.2789790630340576,
      "learning_rate": 0.00013919642857142858,
      "loss": 0.9396877288818359,
      "step": 1560
    },
    {
      "epoch": 0.021030705499797395,
      "grad_norm": 1.3755370378494263,
      "learning_rate": 0.00014008928571428573,
      "loss": 0.9562576293945313,
      "step": 1570
    },
    {
      "epoch": 0.021164659038012665,
      "grad_norm": 1.106921911239624,
      "learning_rate": 0.00014098214285714286,
      "loss": 0.9208305358886719,
      "step": 1580
    },
    {
      "epoch": 0.021298612576227935,
      "grad_norm": 1.2218865156173706,
      "learning_rate": 0.000141875,
      "loss": 0.923252010345459,
      "step": 1590
    },
    {
      "epoch": 0.021432566114443205,
      "grad_norm": 1.1265087127685547,
      "learning_rate": 0.00014276785714285715,
      "loss": 0.9264353752136231,
      "step": 1600
    },
    {
      "epoch": 0.021566519652658475,
      "grad_norm": 1.3815466165542603,
      "learning_rate": 0.00014366071428571428,
      "loss": 0.9643337249755859,
      "step": 1610
    },
    {
      "epoch": 0.021700473190873745,
      "grad_norm": 1.1408675909042358,
      "learning_rate": 0.00014455357142857144,
      "loss": 0.9644880294799805,
      "step": 1620
    },
    {
      "epoch": 0.021834426729089015,
      "grad_norm": 1.202513575553894,
      "learning_rate": 0.0001454464285714286,
      "loss": 0.9403377532958984,
      "step": 1630
    },
    {
      "epoch": 0.021968380267304286,
      "grad_norm": 1.024915337562561,
      "learning_rate": 0.00014633928571428572,
      "loss": 0.9280579566955567,
      "step": 1640
    },
    {
      "epoch": 0.022102333805519556,
      "grad_norm": 1.126615047454834,
      "learning_rate": 0.00014723214285714285,
      "loss": 0.9409577369689941,
      "step": 1650
    },
    {
      "epoch": 0.022236287343734826,
      "grad_norm": 1.0569106340408325,
      "learning_rate": 0.000148125,
      "loss": 0.9590131759643554,
      "step": 1660
    },
    {
      "epoch": 0.022370240881950096,
      "grad_norm": 1.1735954284667969,
      "learning_rate": 0.00014901785714285714,
      "loss": 0.9538652420043945,
      "step": 1670
    },
    {
      "epoch": 0.022504194420165366,
      "grad_norm": 1.0429877042770386,
      "learning_rate": 0.0001499107142857143,
      "loss": 0.9786016464233398,
      "step": 1680
    },
    {
      "epoch": 0.022638147958380636,
      "grad_norm": 1.189475417137146,
      "learning_rate": 0.00015080357142857145,
      "loss": 0.955595874786377,
      "step": 1690
    },
    {
      "epoch": 0.022772101496595906,
      "grad_norm": 1.0221614837646484,
      "learning_rate": 0.00015169642857142858,
      "loss": 0.9822449684143066,
      "step": 1700
    },
    {
      "epoch": 0.022906055034811176,
      "grad_norm": 1.1084915399551392,
      "learning_rate": 0.0001525892857142857,
      "loss": 0.9466286659240722,
      "step": 1710
    },
    {
      "epoch": 0.023040008573026446,
      "grad_norm": 1.1933523416519165,
      "learning_rate": 0.00015348214285714287,
      "loss": 0.9203065872192383,
      "step": 1720
    },
    {
      "epoch": 0.023173962111241717,
      "grad_norm": 1.0246378183364868,
      "learning_rate": 0.000154375,
      "loss": 0.947907829284668,
      "step": 1730
    },
    {
      "epoch": 0.023307915649456987,
      "grad_norm": 1.2799917459487915,
      "learning_rate": 0.00015526785714285716,
      "loss": 0.9746535301208497,
      "step": 1740
    },
    {
      "epoch": 0.023441869187672257,
      "grad_norm": 0.9834460616111755,
      "learning_rate": 0.00015616071428571428,
      "loss": 0.9352348327636719,
      "step": 1750
    },
    {
      "epoch": 0.023575822725887527,
      "grad_norm": 1.1289196014404297,
      "learning_rate": 0.00015705357142857144,
      "loss": 0.9422959327697754,
      "step": 1760
    },
    {
      "epoch": 0.023709776264102797,
      "grad_norm": 1.1601067781448364,
      "learning_rate": 0.00015794642857142857,
      "loss": 0.9540572166442871,
      "step": 1770
    },
    {
      "epoch": 0.023843729802318067,
      "grad_norm": 1.0564992427825928,
      "learning_rate": 0.00015883928571428573,
      "loss": 0.9810628890991211,
      "step": 1780
    },
    {
      "epoch": 0.023977683340533337,
      "grad_norm": 1.4195152521133423,
      "learning_rate": 0.00015973214285714286,
      "loss": 0.9447280883789062,
      "step": 1790
    },
    {
      "epoch": 0.024111636878748607,
      "grad_norm": 1.1684353351593018,
      "learning_rate": 0.00016062500000000001,
      "loss": 0.9330779075622558,
      "step": 1800
    },
    {
      "epoch": 0.024245590416963878,
      "grad_norm": 1.0490233898162842,
      "learning_rate": 0.00016151785714285714,
      "loss": 0.9627155303955078,
      "step": 1810
    },
    {
      "epoch": 0.024379543955179148,
      "grad_norm": 1.1879408359527588,
      "learning_rate": 0.00016241071428571427,
      "loss": 0.9369135856628418,
      "step": 1820
    },
    {
      "epoch": 0.024513497493394418,
      "grad_norm": 1.038156509399414,
      "learning_rate": 0.00016330357142857146,
      "loss": 0.9641859054565429,
      "step": 1830
    },
    {
      "epoch": 0.024647451031609688,
      "grad_norm": 1.281324028968811,
      "learning_rate": 0.0001641964285714286,
      "loss": 0.9745452880859375,
      "step": 1840
    },
    {
      "epoch": 0.024781404569824955,
      "grad_norm": 1.0497291088104248,
      "learning_rate": 0.00016508928571428572,
      "loss": 0.9583146095275878,
      "step": 1850
    },
    {
      "epoch": 0.024915358108040225,
      "grad_norm": 1.2138155698776245,
      "learning_rate": 0.00016598214285714287,
      "loss": 0.94415283203125,
      "step": 1860
    },
    {
      "epoch": 0.025049311646255495,
      "grad_norm": 1.0328693389892578,
      "learning_rate": 0.000166875,
      "loss": 0.9546913146972656,
      "step": 1870
    },
    {
      "epoch": 0.025183265184470765,
      "grad_norm": 0.9754835367202759,
      "learning_rate": 0.00016776785714285713,
      "loss": 0.941204833984375,
      "step": 1880
    },
    {
      "epoch": 0.025317218722686035,
      "grad_norm": 1.1597093343734741,
      "learning_rate": 0.0001686607142857143,
      "loss": 0.9581302642822266,
      "step": 1890
    },
    {
      "epoch": 0.025451172260901305,
      "grad_norm": 0.9526762366294861,
      "learning_rate": 0.00016955357142857145,
      "loss": 0.9658097267150879,
      "step": 1900
    },
    {
      "epoch": 0.025585125799116575,
      "grad_norm": 1.0967228412628174,
      "learning_rate": 0.00017044642857142858,
      "loss": 0.9606739997863769,
      "step": 1910
    },
    {
      "epoch": 0.025719079337331845,
      "grad_norm": 1.2501522302627563,
      "learning_rate": 0.00017133928571428573,
      "loss": 0.9462210655212402,
      "step": 1920
    },
    {
      "epoch": 0.025853032875547115,
      "grad_norm": 1.2555738687515259,
      "learning_rate": 0.00017223214285714286,
      "loss": 0.978199577331543,
      "step": 1930
    },
    {
      "epoch": 0.025986986413762386,
      "grad_norm": 1.2749037742614746,
      "learning_rate": 0.000173125,
      "loss": 0.9561247825622559,
      "step": 1940
    },
    {
      "epoch": 0.026120939951977656,
      "grad_norm": 1.1155575513839722,
      "learning_rate": 0.00017401785714285715,
      "loss": 0.9505109786987305,
      "step": 1950
    },
    {
      "epoch": 0.026254893490192926,
      "grad_norm": 1.1074986457824707,
      "learning_rate": 0.00017491071428571428,
      "loss": 0.964531135559082,
      "step": 1960
    },
    {
      "epoch": 0.026388847028408196,
      "grad_norm": 1.0668226480484009,
      "learning_rate": 0.00017580357142857144,
      "loss": 0.9583681106567383,
      "step": 1970
    },
    {
      "epoch": 0.026522800566623466,
      "grad_norm": 1.0635207891464233,
      "learning_rate": 0.0001766964285714286,
      "loss": 0.9607739448547363,
      "step": 1980
    },
    {
      "epoch": 0.026656754104838736,
      "grad_norm": 1.1774014234542847,
      "learning_rate": 0.00017758928571428572,
      "loss": 0.938353443145752,
      "step": 1990
    },
    {
      "epoch": 0.026790707643054006,
      "grad_norm": 0.9912426471710205,
      "learning_rate": 0.00017848214285714285,
      "loss": 0.9563817977905273,
      "step": 2000
    },
    {
      "epoch": 0.026790707643054006,
      "eval_loss": 0.9294622540473938,
      "eval_runtime": 282.9209,
      "eval_samples_per_second": 10.65,
      "eval_steps_per_second": 10.65,
      "step": 2000
    },
    {
      "epoch": 0.026924661181269276,
      "grad_norm": 1.1386301517486572,
      "learning_rate": 0.000179375,
      "loss": 0.9594663619995117,
      "step": 2010
    },
    {
      "epoch": 0.027058614719484547,
      "grad_norm": 1.0435603857040405,
      "learning_rate": 0.00018026785714285714,
      "loss": 0.9453421592712402,
      "step": 2020
    },
    {
      "epoch": 0.027192568257699817,
      "grad_norm": 1.1574785709381104,
      "learning_rate": 0.0001811607142857143,
      "loss": 0.9297013282775879,
      "step": 2030
    },
    {
      "epoch": 0.027326521795915087,
      "grad_norm": 1.1167242527008057,
      "learning_rate": 0.00018205357142857145,
      "loss": 0.9530548095703125,
      "step": 2040
    },
    {
      "epoch": 0.027460475334130357,
      "grad_norm": 1.0436396598815918,
      "learning_rate": 0.00018294642857142858,
      "loss": 0.9412404060363769,
      "step": 2050
    },
    {
      "epoch": 0.027594428872345627,
      "grad_norm": 1.184274435043335,
      "learning_rate": 0.0001838392857142857,
      "loss": 0.9605161666870117,
      "step": 2060
    },
    {
      "epoch": 0.027728382410560897,
      "grad_norm": 0.9393510222434998,
      "learning_rate": 0.00018473214285714287,
      "loss": 0.9509821891784668,
      "step": 2070
    },
    {
      "epoch": 0.027862335948776167,
      "grad_norm": 1.0484318733215332,
      "learning_rate": 0.000185625,
      "loss": 0.9221634864807129,
      "step": 2080
    },
    {
      "epoch": 0.027996289486991437,
      "grad_norm": 0.9934367537498474,
      "learning_rate": 0.00018651785714285716,
      "loss": 0.9437132835388183,
      "step": 2090
    },
    {
      "epoch": 0.028130243025206707,
      "grad_norm": 0.9708594679832458,
      "learning_rate": 0.00018741071428571429,
      "loss": 0.9631563186645508,
      "step": 2100
    },
    {
      "epoch": 0.028264196563421978,
      "grad_norm": 1.0960266590118408,
      "learning_rate": 0.00018830357142857144,
      "loss": 0.9347660064697265,
      "step": 2110
    },
    {
      "epoch": 0.028398150101637248,
      "grad_norm": 1.4339450597763062,
      "learning_rate": 0.00018919642857142857,
      "loss": 0.9685755729675293,
      "step": 2120
    },
    {
      "epoch": 0.028532103639852518,
      "grad_norm": 0.9974057078361511,
      "learning_rate": 0.00019008928571428573,
      "loss": 0.9545693397521973,
      "step": 2130
    },
    {
      "epoch": 0.028666057178067788,
      "grad_norm": 1.0973010063171387,
      "learning_rate": 0.00019098214285714286,
      "loss": 0.9272083282470703,
      "step": 2140
    },
    {
      "epoch": 0.028800010716283058,
      "grad_norm": 0.9838769435882568,
      "learning_rate": 0.00019187500000000002,
      "loss": 0.9372439384460449,
      "step": 2150
    },
    {
      "epoch": 0.028933964254498328,
      "grad_norm": 1.0080406665802002,
      "learning_rate": 0.00019276785714285715,
      "loss": 0.9752328872680665,
      "step": 2160
    },
    {
      "epoch": 0.0290679177927136,
      "grad_norm": 0.9965051412582397,
      "learning_rate": 0.00019366071428571427,
      "loss": 0.9454973220825196,
      "step": 2170
    },
    {
      "epoch": 0.02920187133092887,
      "grad_norm": 0.9947363138198853,
      "learning_rate": 0.00019455357142857146,
      "loss": 0.9513129234313965,
      "step": 2180
    },
    {
      "epoch": 0.02933582486914414,
      "grad_norm": 1.1437948942184448,
      "learning_rate": 0.0001954464285714286,
      "loss": 0.9673953056335449,
      "step": 2190
    },
    {
      "epoch": 0.02946977840735941,
      "grad_norm": 1.0892202854156494,
      "learning_rate": 0.00019633928571428572,
      "loss": 0.9653889656066894,
      "step": 2200
    },
    {
      "epoch": 0.02960373194557468,
      "grad_norm": 1.149424433708191,
      "learning_rate": 0.00019723214285714287,
      "loss": 0.9513349533081055,
      "step": 2210
    },
    {
      "epoch": 0.02973768548378995,
      "grad_norm": 1.014336347579956,
      "learning_rate": 0.000198125,
      "loss": 0.9285200119018555,
      "step": 2220
    },
    {
      "epoch": 0.02987163902200522,
      "grad_norm": 1.0589832067489624,
      "learning_rate": 0.00019901785714285713,
      "loss": 0.9304631233215332,
      "step": 2230
    },
    {
      "epoch": 0.03000559256022049,
      "grad_norm": 0.9858236312866211,
      "learning_rate": 0.0001999107142857143,
      "loss": 0.9667643547058106,
      "step": 2240
    },
    {
      "epoch": 0.03013954609843576,
      "grad_norm": 0.9708908796310425,
      "learning_rate": 0.0001999999923770744,
      "loss": 0.9756393432617188,
      "step": 2250
    },
    {
      "epoch": 0.030273499636651026,
      "grad_norm": 1.029149055480957,
      "learning_rate": 0.00019999996602622187,
      "loss": 0.9660782814025879,
      "step": 2260
    },
    {
      "epoch": 0.030407453174866296,
      "grad_norm": 0.9579015374183655,
      "learning_rate": 0.00019999992085333718,
      "loss": 0.9369187355041504,
      "step": 2270
    },
    {
      "epoch": 0.030541406713081566,
      "grad_norm": 1.0455657243728638,
      "learning_rate": 0.0001999998568584289,
      "loss": 0.957420539855957,
      "step": 2280
    },
    {
      "epoch": 0.030675360251296836,
      "grad_norm": 1.1843011379241943,
      "learning_rate": 0.00019999977404150902,
      "loss": 0.9495012283325195,
      "step": 2290
    },
    {
      "epoch": 0.030809313789512106,
      "grad_norm": 1.0620977878570557,
      "learning_rate": 0.00019999967240259315,
      "loss": 0.9708341598510742,
      "step": 2300
    },
    {
      "epoch": 0.030943267327727376,
      "grad_norm": 1.0837901830673218,
      "learning_rate": 0.0001999995519417004,
      "loss": 0.9840412139892578,
      "step": 2310
    },
    {
      "epoch": 0.031077220865942647,
      "grad_norm": 0.9072780013084412,
      "learning_rate": 0.00019999941265885348,
      "loss": 0.9503616333007813,
      "step": 2320
    },
    {
      "epoch": 0.031211174404157917,
      "grad_norm": 0.8991392850875854,
      "learning_rate": 0.00019999925455407855,
      "loss": 0.960198974609375,
      "step": 2330
    },
    {
      "epoch": 0.03134512794237319,
      "grad_norm": 0.8700608015060425,
      "learning_rate": 0.00019999907762740542,
      "loss": 0.9340170860290528,
      "step": 2340
    },
    {
      "epoch": 0.03147908148058846,
      "grad_norm": 1.0618027448654175,
      "learning_rate": 0.00019999888187886735,
      "loss": 0.9670058250427246,
      "step": 2350
    },
    {
      "epoch": 0.03161303501880373,
      "grad_norm": 1.0053410530090332,
      "learning_rate": 0.0001999986673085012,
      "loss": 0.9664807319641113,
      "step": 2360
    },
    {
      "epoch": 0.031746988557019,
      "grad_norm": 1.0876250267028809,
      "learning_rate": 0.0001999984339163474,
      "loss": 0.9429721832275391,
      "step": 2370
    },
    {
      "epoch": 0.03188094209523427,
      "grad_norm": 1.1047455072402954,
      "learning_rate": 0.00019999818170244983,
      "loss": 0.9404646873474121,
      "step": 2380
    },
    {
      "epoch": 0.03201489563344954,
      "grad_norm": 1.0188040733337402,
      "learning_rate": 0.00019999791066685596,
      "loss": 0.9436681747436524,
      "step": 2390
    },
    {
      "epoch": 0.03214884917166481,
      "grad_norm": 0.9918760657310486,
      "learning_rate": 0.0001999976208096168,
      "loss": 0.9435869216918945,
      "step": 2400
    },
    {
      "epoch": 0.03228280270988008,
      "grad_norm": 1.0088706016540527,
      "learning_rate": 0.00019999731213078697,
      "loss": 0.9398737907409668,
      "step": 2410
    },
    {
      "epoch": 0.03241675624809535,
      "grad_norm": 0.9979799389839172,
      "learning_rate": 0.00019999698463042448,
      "loss": 0.9468537330627441,
      "step": 2420
    },
    {
      "epoch": 0.03255070978631062,
      "grad_norm": 0.9645936489105225,
      "learning_rate": 0.00019999663830859104,
      "loss": 0.9193555831909179,
      "step": 2430
    },
    {
      "epoch": 0.03268466332452589,
      "grad_norm": 0.9175757169723511,
      "learning_rate": 0.00019999627316535178,
      "loss": 0.9474472999572754,
      "step": 2440
    },
    {
      "epoch": 0.03281861686274116,
      "grad_norm": 0.9668617844581604,
      "learning_rate": 0.0001999958892007755,
      "loss": 0.9571184158325196,
      "step": 2450
    },
    {
      "epoch": 0.03295257040095643,
      "grad_norm": 1.0439015626907349,
      "learning_rate": 0.00019999548641493438,
      "loss": 0.9383399009704589,
      "step": 2460
    },
    {
      "epoch": 0.0330865239391717,
      "grad_norm": 1.014618158340454,
      "learning_rate": 0.00019999506480790434,
      "loss": 0.8999407768249512,
      "step": 2470
    },
    {
      "epoch": 0.03322047747738697,
      "grad_norm": 0.9947917461395264,
      "learning_rate": 0.00019999462437976463,
      "loss": 0.9517890930175781,
      "step": 2480
    },
    {
      "epoch": 0.03335443101560224,
      "grad_norm": 0.9013609290122986,
      "learning_rate": 0.0001999941651305982,
      "loss": 0.9355831146240234,
      "step": 2490
    },
    {
      "epoch": 0.03348838455381751,
      "grad_norm": 1.045623540878296,
      "learning_rate": 0.0001999936870604915,
      "loss": 0.9567943572998047,
      "step": 2500
    },
    {
      "epoch": 0.03362233809203278,
      "grad_norm": 1.0059949159622192,
      "learning_rate": 0.00019999319016953448,
      "loss": 0.9816353797912598,
      "step": 2510
    },
    {
      "epoch": 0.03375629163024805,
      "grad_norm": 0.9956102967262268,
      "learning_rate": 0.0001999926744578207,
      "loss": 0.9603646278381348,
      "step": 2520
    },
    {
      "epoch": 0.03389024516846332,
      "grad_norm": 0.9674333333969116,
      "learning_rate": 0.0001999921399254472,
      "loss": 0.9202863693237304,
      "step": 2530
    },
    {
      "epoch": 0.03402419870667859,
      "grad_norm": 1.149797797203064,
      "learning_rate": 0.00019999158657251458,
      "loss": 0.9565104484558106,
      "step": 2540
    },
    {
      "epoch": 0.03415815224489386,
      "grad_norm": 0.8696104288101196,
      "learning_rate": 0.00019999101439912702,
      "loss": 0.9499526977539062,
      "step": 2550
    },
    {
      "epoch": 0.03429210578310913,
      "grad_norm": 0.9146940112113953,
      "learning_rate": 0.0001999904234053922,
      "loss": 0.9595541000366211,
      "step": 2560
    },
    {
      "epoch": 0.0344260593213244,
      "grad_norm": 1.1251157522201538,
      "learning_rate": 0.00019998981359142136,
      "loss": 0.9481052398681641,
      "step": 2570
    },
    {
      "epoch": 0.03456001285953967,
      "grad_norm": 0.951270341873169,
      "learning_rate": 0.0001999891849573293,
      "loss": 0.9557564735412598,
      "step": 2580
    },
    {
      "epoch": 0.03469396639775494,
      "grad_norm": 0.931270956993103,
      "learning_rate": 0.0001999885375032343,
      "loss": 0.9400008201599122,
      "step": 2590
    },
    {
      "epoch": 0.03482791993597021,
      "grad_norm": 0.9405684471130371,
      "learning_rate": 0.00019998787122925827,
      "loss": 0.9368779182434082,
      "step": 2600
    },
    {
      "epoch": 0.03496187347418548,
      "grad_norm": 1.1437263488769531,
      "learning_rate": 0.00019998718613552658,
      "loss": 0.9466694831848145,
      "step": 2610
    },
    {
      "epoch": 0.03509582701240075,
      "grad_norm": 1.1577832698822021,
      "learning_rate": 0.0001999864822221682,
      "loss": 0.9129865646362305,
      "step": 2620
    },
    {
      "epoch": 0.03522978055061602,
      "grad_norm": 1.0464088916778564,
      "learning_rate": 0.0001999857594893156,
      "loss": 0.9279396057128906,
      "step": 2630
    },
    {
      "epoch": 0.03536373408883129,
      "grad_norm": 1.053524136543274,
      "learning_rate": 0.00019998501793710484,
      "loss": 0.9761667251586914,
      "step": 2640
    },
    {
      "epoch": 0.03549768762704656,
      "grad_norm": 0.902043342590332,
      "learning_rate": 0.0001999842575656755,
      "loss": 0.9456755638122558,
      "step": 2650
    },
    {
      "epoch": 0.03563164116526183,
      "grad_norm": 1.0060131549835205,
      "learning_rate": 0.00019998347837517063,
      "loss": 0.9625540733337402,
      "step": 2660
    },
    {
      "epoch": 0.0357655947034771,
      "grad_norm": 1.1090943813323975,
      "learning_rate": 0.00019998268036573697,
      "loss": 0.9310230255126953,
      "step": 2670
    },
    {
      "epoch": 0.03589954824169237,
      "grad_norm": 1.0450677871704102,
      "learning_rate": 0.00019998186353752467,
      "loss": 0.9483952522277832,
      "step": 2680
    },
    {
      "epoch": 0.03603350177990764,
      "grad_norm": 1.1651952266693115,
      "learning_rate": 0.00019998102789068752,
      "loss": 0.9506073951721191,
      "step": 2690
    },
    {
      "epoch": 0.03616745531812291,
      "grad_norm": 1.033385157585144,
      "learning_rate": 0.00019998017342538274,
      "loss": 0.9435412406921386,
      "step": 2700
    },
    {
      "epoch": 0.03630140885633818,
      "grad_norm": 0.9711186289787292,
      "learning_rate": 0.00019997930014177125,
      "loss": 0.9563213348388672,
      "step": 2710
    },
    {
      "epoch": 0.03643536239455345,
      "grad_norm": 1.0140986442565918,
      "learning_rate": 0.00019997840804001732,
      "loss": 0.9228851318359375,
      "step": 2720
    },
    {
      "epoch": 0.03656931593276872,
      "grad_norm": 1.0165528059005737,
      "learning_rate": 0.00019997749712028892,
      "loss": 0.9404541969299316,
      "step": 2730
    },
    {
      "epoch": 0.03670326947098399,
      "grad_norm": 0.9036928415298462,
      "learning_rate": 0.00019997656738275751,
      "loss": 0.9369661331176757,
      "step": 2740
    },
    {
      "epoch": 0.03683722300919926,
      "grad_norm": 0.9310727119445801,
      "learning_rate": 0.00019997561882759807,
      "loss": 0.9242416381835937,
      "step": 2750
    },
    {
      "epoch": 0.03697117654741453,
      "grad_norm": 0.9902796745300293,
      "learning_rate": 0.00019997465145498914,
      "loss": 0.9326069831848145,
      "step": 2760
    },
    {
      "epoch": 0.0371051300856298,
      "grad_norm": 1.0533158779144287,
      "learning_rate": 0.00019997366526511274,
      "loss": 0.9423623085021973,
      "step": 2770
    },
    {
      "epoch": 0.03723908362384507,
      "grad_norm": 1.045711874961853,
      "learning_rate": 0.0001999726602581546,
      "loss": 0.9074527740478515,
      "step": 2780
    },
    {
      "epoch": 0.03737303716206034,
      "grad_norm": 0.9685200452804565,
      "learning_rate": 0.00019997163643430382,
      "loss": 0.9323711395263672,
      "step": 2790
    },
    {
      "epoch": 0.03750699070027561,
      "grad_norm": 1.0000747442245483,
      "learning_rate": 0.0001999705937937531,
      "loss": 0.9754042625427246,
      "step": 2800
    },
    {
      "epoch": 0.03764094423849088,
      "grad_norm": 0.9732158184051514,
      "learning_rate": 0.0001999695323366987,
      "loss": 0.9792140007019043,
      "step": 2810
    },
    {
      "epoch": 0.03777489777670615,
      "grad_norm": 0.990386426448822,
      "learning_rate": 0.0001999684520633404,
      "loss": 0.9250985145568847,
      "step": 2820
    },
    {
      "epoch": 0.03790885131492142,
      "grad_norm": 1.0131051540374756,
      "learning_rate": 0.00019996735297388157,
      "loss": 0.9338252067565918,
      "step": 2830
    },
    {
      "epoch": 0.03804280485313669,
      "grad_norm": 0.9239967465400696,
      "learning_rate": 0.000199966235068529,
      "loss": 0.9674043655395508,
      "step": 2840
    },
    {
      "epoch": 0.03817675839135196,
      "grad_norm": 0.9224802255630493,
      "learning_rate": 0.00019996509834749318,
      "loss": 0.9093979835510254,
      "step": 2850
    },
    {
      "epoch": 0.03831071192956723,
      "grad_norm": 0.9065786004066467,
      "learning_rate": 0.00019996394281098802,
      "loss": 0.9495321273803711,
      "step": 2860
    },
    {
      "epoch": 0.0384446654677825,
      "grad_norm": 1.0348045825958252,
      "learning_rate": 0.00019996276845923104,
      "loss": 0.9240209579467773,
      "step": 2870
    },
    {
      "epoch": 0.038578619005997766,
      "grad_norm": 0.9366620182991028,
      "learning_rate": 0.0001999615752924433,
      "loss": 0.9238608360290528,
      "step": 2880
    },
    {
      "epoch": 0.038712572544213036,
      "grad_norm": 1.122186303138733,
      "learning_rate": 0.0001999603633108493,
      "loss": 0.9154872894287109,
      "step": 2890
    },
    {
      "epoch": 0.038846526082428307,
      "grad_norm": 0.8368102312088013,
      "learning_rate": 0.0001999591325146772,
      "loss": 0.9037948608398437,
      "step": 2900
    },
    {
      "epoch": 0.03898047962064358,
      "grad_norm": 0.8084287047386169,
      "learning_rate": 0.00019995788290415867,
      "loss": 0.9526560783386231,
      "step": 2910
    },
    {
      "epoch": 0.03911443315885885,
      "grad_norm": 0.9385167360305786,
      "learning_rate": 0.00019995661447952894,
      "loss": 0.9274347305297852,
      "step": 2920
    },
    {
      "epoch": 0.03924838669707412,
      "grad_norm": 1.0126752853393555,
      "learning_rate": 0.00019995532724102668,
      "loss": 0.9336477279663086,
      "step": 2930
    },
    {
      "epoch": 0.03938234023528939,
      "grad_norm": 1.0768028497695923,
      "learning_rate": 0.00019995402118889424,
      "loss": 0.9312108039855957,
      "step": 2940
    },
    {
      "epoch": 0.03951629377350466,
      "grad_norm": 0.9114157557487488,
      "learning_rate": 0.00019995269632337742,
      "loss": 0.9413155555725098,
      "step": 2950
    },
    {
      "epoch": 0.03965024731171993,
      "grad_norm": 0.8750312924385071,
      "learning_rate": 0.00019995135264472557,
      "loss": 0.9509553909301758,
      "step": 2960
    },
    {
      "epoch": 0.0397842008499352,
      "grad_norm": 1.0956413745880127,
      "learning_rate": 0.00019994999015319164,
      "loss": 0.9265983581542969,
      "step": 2970
    },
    {
      "epoch": 0.03991815438815047,
      "grad_norm": 0.9434391856193542,
      "learning_rate": 0.00019994860884903204,
      "loss": 0.9004026412963867,
      "step": 2980
    },
    {
      "epoch": 0.04005210792636574,
      "grad_norm": 0.8206298351287842,
      "learning_rate": 0.00019994720873250677,
      "loss": 0.9307710647583007,
      "step": 2990
    },
    {
      "epoch": 0.04018606146458101,
      "grad_norm": 1.0000940561294556,
      "learning_rate": 0.00019994578980387937,
      "loss": 0.9217443466186523,
      "step": 3000
    },
    {
      "epoch": 0.04018606146458101,
      "eval_loss": 0.9095186591148376,
      "eval_runtime": 282.8579,
      "eval_samples_per_second": 10.652,
      "eval_steps_per_second": 10.652,
      "step": 3000
    },
    {
      "epoch": 0.04032001500279628,
      "grad_norm": 0.8734673857688904,
      "learning_rate": 0.0001999443520634169,
      "loss": 0.941364860534668,
      "step": 3010
    },
    {
      "epoch": 0.04045396854101155,
      "grad_norm": 0.8950884938240051,
      "learning_rate": 0.00019994289551139,
      "loss": 0.9462560653686524,
      "step": 3020
    },
    {
      "epoch": 0.04058792207922682,
      "grad_norm": 1.0811048746109009,
      "learning_rate": 0.00019994142014807277,
      "loss": 0.9378127098083496,
      "step": 3030
    },
    {
      "epoch": 0.04072187561744209,
      "grad_norm": 1.3110917806625366,
      "learning_rate": 0.00019993992597374297,
      "loss": 0.9490732192993164,
      "step": 3040
    },
    {
      "epoch": 0.04085582915565736,
      "grad_norm": 0.892918050289154,
      "learning_rate": 0.00019993841298868175,
      "loss": 0.9350913047790528,
      "step": 3050
    },
    {
      "epoch": 0.04098978269387263,
      "grad_norm": 0.8810139298439026,
      "learning_rate": 0.00019993688119317396,
      "loss": 0.9095968246459961,
      "step": 3060
    },
    {
      "epoch": 0.0411237362320879,
      "grad_norm": 0.9049996137619019,
      "learning_rate": 0.00019993533058750788,
      "loss": 0.9268231391906738,
      "step": 3070
    },
    {
      "epoch": 0.04125768977030317,
      "grad_norm": 0.8555030226707458,
      "learning_rate": 0.0001999337611719754,
      "loss": 0.9202340126037598,
      "step": 3080
    },
    {
      "epoch": 0.04139164330851844,
      "grad_norm": 0.8748414516448975,
      "learning_rate": 0.00019993217294687185,
      "loss": 0.9081048965454102,
      "step": 3090
    },
    {
      "epoch": 0.04152559684673371,
      "grad_norm": 0.9201762080192566,
      "learning_rate": 0.00019993056591249624,
      "loss": 0.9255125045776367,
      "step": 3100
    },
    {
      "epoch": 0.04165955038494898,
      "grad_norm": 0.8035100102424622,
      "learning_rate": 0.00019992894006915098,
      "loss": 0.9364505767822265,
      "step": 3110
    },
    {
      "epoch": 0.04179350392316425,
      "grad_norm": 0.8489510416984558,
      "learning_rate": 0.00019992729541714214,
      "loss": 0.9114150047302246,
      "step": 3120
    },
    {
      "epoch": 0.04192745746137952,
      "grad_norm": 0.8885366916656494,
      "learning_rate": 0.00019992563195677927,
      "loss": 0.9378640174865722,
      "step": 3130
    },
    {
      "epoch": 0.04206141099959479,
      "grad_norm": 0.916833221912384,
      "learning_rate": 0.00019992394968837546,
      "loss": 0.935064697265625,
      "step": 3140
    },
    {
      "epoch": 0.04219536453781006,
      "grad_norm": 0.8948741555213928,
      "learning_rate": 0.0001999222486122473,
      "loss": 0.9183822631835937,
      "step": 3150
    },
    {
      "epoch": 0.04232931807602533,
      "grad_norm": 1.0157934427261353,
      "learning_rate": 0.000199920528728715,
      "loss": 0.946108055114746,
      "step": 3160
    },
    {
      "epoch": 0.0424632716142406,
      "grad_norm": 0.8575133085250854,
      "learning_rate": 0.00019991879003810231,
      "loss": 0.9633334159851075,
      "step": 3170
    },
    {
      "epoch": 0.04259722515245587,
      "grad_norm": 0.9529309272766113,
      "learning_rate": 0.00019991703254073647,
      "loss": 0.9019985198974609,
      "step": 3180
    },
    {
      "epoch": 0.04273117869067114,
      "grad_norm": 0.9916787147521973,
      "learning_rate": 0.00019991525623694827,
      "loss": 0.9476448059082031,
      "step": 3190
    },
    {
      "epoch": 0.04286513222888641,
      "grad_norm": 0.8963183164596558,
      "learning_rate": 0.00019991346112707204,
      "loss": 0.9350907325744628,
      "step": 3200
    },
    {
      "epoch": 0.04299908576710168,
      "grad_norm": 0.9970526695251465,
      "learning_rate": 0.00019991164721144564,
      "loss": 0.9250227928161621,
      "step": 3210
    },
    {
      "epoch": 0.04313303930531695,
      "grad_norm": 0.8881255388259888,
      "learning_rate": 0.00019990981449041054,
      "loss": 0.9250835418701172,
      "step": 3220
    },
    {
      "epoch": 0.04326699284353222,
      "grad_norm": 1.0573492050170898,
      "learning_rate": 0.00019990796296431165,
      "loss": 0.9150334358215332,
      "step": 3230
    },
    {
      "epoch": 0.04340094638174749,
      "grad_norm": 1.0080435276031494,
      "learning_rate": 0.00019990609263349746,
      "loss": 0.9671030044555664,
      "step": 3240
    },
    {
      "epoch": 0.04353489991996276,
      "grad_norm": 0.9293102025985718,
      "learning_rate": 0.00019990420349832004,
      "loss": 0.9286737442016602,
      "step": 3250
    },
    {
      "epoch": 0.04366885345817803,
      "grad_norm": 0.9268156886100769,
      "learning_rate": 0.00019990229555913493,
      "loss": 0.9326079368591309,
      "step": 3260
    },
    {
      "epoch": 0.0438028069963933,
      "grad_norm": 0.9221058487892151,
      "learning_rate": 0.0001999003688163013,
      "loss": 0.9419919013977051,
      "step": 3270
    },
    {
      "epoch": 0.04393676053460857,
      "grad_norm": 0.8882941007614136,
      "learning_rate": 0.0001998984232701817,
      "loss": 0.9261360168457031,
      "step": 3280
    },
    {
      "epoch": 0.04407071407282384,
      "grad_norm": 0.844829797744751,
      "learning_rate": 0.00019989645892114243,
      "loss": 0.927768325805664,
      "step": 3290
    },
    {
      "epoch": 0.04420466761103911,
      "grad_norm": 1.0086514949798584,
      "learning_rate": 0.00019989447576955317,
      "loss": 0.9342447280883789,
      "step": 3300
    },
    {
      "epoch": 0.04433862114925438,
      "grad_norm": 0.9930247068405151,
      "learning_rate": 0.00019989247381578719,
      "loss": 0.9354818344116211,
      "step": 3310
    },
    {
      "epoch": 0.04447257468746965,
      "grad_norm": 0.9238676428794861,
      "learning_rate": 0.0001998904530602213,
      "loss": 0.95006103515625,
      "step": 3320
    },
    {
      "epoch": 0.04460652822568492,
      "grad_norm": 0.8584195971488953,
      "learning_rate": 0.0001998884135032358,
      "loss": 0.9177757263183594,
      "step": 3330
    },
    {
      "epoch": 0.04474048176390019,
      "grad_norm": 0.9681214094161987,
      "learning_rate": 0.0001998863551452147,
      "loss": 0.9331873893737793,
      "step": 3340
    },
    {
      "epoch": 0.04487443530211546,
      "grad_norm": 0.9767482876777649,
      "learning_rate": 0.0001998842779865453,
      "loss": 0.8993824005126954,
      "step": 3350
    },
    {
      "epoch": 0.04500838884033073,
      "grad_norm": 0.9796115159988403,
      "learning_rate": 0.00019988218202761867,
      "loss": 0.9324639320373536,
      "step": 3360
    },
    {
      "epoch": 0.045142342378546,
      "grad_norm": 1.0159355401992798,
      "learning_rate": 0.0001998800672688292,
      "loss": 0.9196332931518555,
      "step": 3370
    },
    {
      "epoch": 0.04527629591676127,
      "grad_norm": 0.9687798619270325,
      "learning_rate": 0.000199877933710575,
      "loss": 0.9572076797485352,
      "step": 3380
    },
    {
      "epoch": 0.04541024945497654,
      "grad_norm": 0.9174926280975342,
      "learning_rate": 0.00019987578135325764,
      "loss": 0.9035839080810547,
      "step": 3390
    },
    {
      "epoch": 0.04554420299319181,
      "grad_norm": 0.8543787002563477,
      "learning_rate": 0.00019987361019728223,
      "loss": 0.9235164642333984,
      "step": 3400
    },
    {
      "epoch": 0.04567815653140708,
      "grad_norm": 0.870416522026062,
      "learning_rate": 0.00019987142024305743,
      "loss": 0.9195279121398926,
      "step": 3410
    },
    {
      "epoch": 0.04581211006962235,
      "grad_norm": 0.8485437631607056,
      "learning_rate": 0.00019986921149099543,
      "loss": 0.9079272270202636,
      "step": 3420
    },
    {
      "epoch": 0.04594606360783762,
      "grad_norm": 0.8823762536048889,
      "learning_rate": 0.00019986698394151197,
      "loss": 0.9315704345703125,
      "step": 3430
    },
    {
      "epoch": 0.04608001714605289,
      "grad_norm": 0.8347775340080261,
      "learning_rate": 0.0001998647375950263,
      "loss": 0.8943398475646973,
      "step": 3440
    },
    {
      "epoch": 0.04621397068426816,
      "grad_norm": 0.8320932388305664,
      "learning_rate": 0.0001998624724519613,
      "loss": 0.9377827644348145,
      "step": 3450
    },
    {
      "epoch": 0.04634792422248343,
      "grad_norm": 1.092563271522522,
      "learning_rate": 0.0001998601885127432,
      "loss": 0.9222759246826172,
      "step": 3460
    },
    {
      "epoch": 0.0464818777606987,
      "grad_norm": 0.9765579104423523,
      "learning_rate": 0.00019985788577780194,
      "loss": 0.9315988540649414,
      "step": 3470
    },
    {
      "epoch": 0.04661583129891397,
      "grad_norm": 0.8835811614990234,
      "learning_rate": 0.00019985556424757098,
      "loss": 0.9130311012268066,
      "step": 3480
    },
    {
      "epoch": 0.046749784837129243,
      "grad_norm": 0.8476258516311646,
      "learning_rate": 0.00019985322392248724,
      "loss": 0.9215794563293457,
      "step": 3490
    },
    {
      "epoch": 0.046883738375344514,
      "grad_norm": 0.9369233846664429,
      "learning_rate": 0.00019985086480299121,
      "loss": 0.9167479515075684,
      "step": 3500
    },
    {
      "epoch": 0.047017691913559784,
      "grad_norm": 0.9186804890632629,
      "learning_rate": 0.00019984848688952696,
      "loss": 0.9383148193359375,
      "step": 3510
    },
    {
      "epoch": 0.047151645451775054,
      "grad_norm": 0.9544302821159363,
      "learning_rate": 0.000199846090182542,
      "loss": 0.9238338470458984,
      "step": 3520
    },
    {
      "epoch": 0.047285598989990324,
      "grad_norm": 0.9397124648094177,
      "learning_rate": 0.00019984367468248748,
      "loss": 0.9094552993774414,
      "step": 3530
    },
    {
      "epoch": 0.047419552528205594,
      "grad_norm": 0.8298595547676086,
      "learning_rate": 0.00019984124038981808,
      "loss": 0.9345894813537597,
      "step": 3540
    },
    {
      "epoch": 0.047553506066420864,
      "grad_norm": 0.9456910490989685,
      "learning_rate": 0.00019983878730499193,
      "loss": 0.9422292709350586,
      "step": 3550
    },
    {
      "epoch": 0.047687459604636134,
      "grad_norm": 0.8198268413543701,
      "learning_rate": 0.00019983631542847075,
      "loss": 0.9278390884399415,
      "step": 3560
    },
    {
      "epoch": 0.047821413142851404,
      "grad_norm": 0.8617146015167236,
      "learning_rate": 0.00019983382476071983,
      "loss": 0.9171154975891114,
      "step": 3570
    },
    {
      "epoch": 0.047955366681066675,
      "grad_norm": 0.8447020649909973,
      "learning_rate": 0.00019983131530220793,
      "loss": 0.9147285461425781,
      "step": 3580
    },
    {
      "epoch": 0.048089320219281945,
      "grad_norm": 0.8757874965667725,
      "learning_rate": 0.00019982878705340743,
      "loss": 0.936093807220459,
      "step": 3590
    },
    {
      "epoch": 0.048223273757497215,
      "grad_norm": 0.8899102807044983,
      "learning_rate": 0.00019982624001479415,
      "loss": 0.9166989326477051,
      "step": 3600
    },
    {
      "epoch": 0.048357227295712485,
      "grad_norm": 0.8696548342704773,
      "learning_rate": 0.0001998236741868475,
      "loss": 0.9303563117980957,
      "step": 3610
    },
    {
      "epoch": 0.048491180833927755,
      "grad_norm": 0.9574766159057617,
      "learning_rate": 0.00019982108957005047,
      "loss": 0.9008687019348145,
      "step": 3620
    },
    {
      "epoch": 0.048625134372143025,
      "grad_norm": 0.988602876663208,
      "learning_rate": 0.00019981848616488948,
      "loss": 0.926761245727539,
      "step": 3630
    },
    {
      "epoch": 0.048759087910358295,
      "grad_norm": 0.9067688584327698,
      "learning_rate": 0.00019981586397185456,
      "loss": 0.9246554374694824,
      "step": 3640
    },
    {
      "epoch": 0.048893041448573565,
      "grad_norm": 0.8702062368392944,
      "learning_rate": 0.00019981322299143927,
      "loss": 0.9095652580261231,
      "step": 3650
    },
    {
      "epoch": 0.049026994986788835,
      "grad_norm": 0.9878699779510498,
      "learning_rate": 0.00019981056322414072,
      "loss": 0.8977669715881348,
      "step": 3660
    },
    {
      "epoch": 0.049160948525004106,
      "grad_norm": 0.8746298551559448,
      "learning_rate": 0.00019980788467045945,
      "loss": 0.9227336883544922,
      "step": 3670
    },
    {
      "epoch": 0.049294902063219376,
      "grad_norm": 0.9934686422348022,
      "learning_rate": 0.0001998051873308997,
      "loss": 0.9178746223449707,
      "step": 3680
    },
    {
      "epoch": 0.049428855601434646,
      "grad_norm": 0.9328760504722595,
      "learning_rate": 0.00019980247120596916,
      "loss": 0.9450766563415527,
      "step": 3690
    },
    {
      "epoch": 0.04956280913964991,
      "grad_norm": 0.8085207343101501,
      "learning_rate": 0.00019979973629617902,
      "loss": 0.9002262115478515,
      "step": 3700
    },
    {
      "epoch": 0.04969676267786518,
      "grad_norm": 0.9033879041671753,
      "learning_rate": 0.00019979698260204405,
      "loss": 0.9274055480957031,
      "step": 3710
    },
    {
      "epoch": 0.04983071621608045,
      "grad_norm": 0.9597737193107605,
      "learning_rate": 0.00019979421012408257,
      "loss": 0.8976191520690918,
      "step": 3720
    },
    {
      "epoch": 0.04996466975429572,
      "grad_norm": 0.8724941611289978,
      "learning_rate": 0.00019979141886281642,
      "loss": 0.9168087005615234,
      "step": 3730
    },
    {
      "epoch": 0.05009862329251099,
      "grad_norm": 0.8592389225959778,
      "learning_rate": 0.00019978860881877096,
      "loss": 0.9181900024414062,
      "step": 3740
    },
    {
      "epoch": 0.05023257683072626,
      "grad_norm": 0.9238866567611694,
      "learning_rate": 0.0001997857799924751,
      "loss": 0.9294312477111817,
      "step": 3750
    },
    {
      "epoch": 0.05036653036894153,
      "grad_norm": 0.9541118144989014,
      "learning_rate": 0.0001997829323844613,
      "loss": 0.9044769287109375,
      "step": 3760
    },
    {
      "epoch": 0.0505004839071568,
      "grad_norm": 0.8983111381530762,
      "learning_rate": 0.0001997800659952655,
      "loss": 0.9011771202087402,
      "step": 3770
    },
    {
      "epoch": 0.05063443744537207,
      "grad_norm": 0.9186615943908691,
      "learning_rate": 0.00019977718082542722,
      "loss": 0.9071616172790528,
      "step": 3780
    },
    {
      "epoch": 0.05076839098358734,
      "grad_norm": 0.8877631425857544,
      "learning_rate": 0.00019977427687548955,
      "loss": 0.9309605598449707,
      "step": 3790
    },
    {
      "epoch": 0.05090234452180261,
      "grad_norm": 0.8731006979942322,
      "learning_rate": 0.00019977135414599905,
      "loss": 0.9176346778869628,
      "step": 3800
    },
    {
      "epoch": 0.05103629806001788,
      "grad_norm": 0.9903236627578735,
      "learning_rate": 0.00019976841263750583,
      "loss": 0.9287304878234863,
      "step": 3810
    },
    {
      "epoch": 0.05117025159823315,
      "grad_norm": 1.0615185499191284,
      "learning_rate": 0.00019976545235056352,
      "loss": 0.924833106994629,
      "step": 3820
    },
    {
      "epoch": 0.05130420513644842,
      "grad_norm": 0.9164450764656067,
      "learning_rate": 0.00019976247328572938,
      "loss": 0.9383651733398437,
      "step": 3830
    },
    {
      "epoch": 0.05143815867466369,
      "grad_norm": 1.0046485662460327,
      "learning_rate": 0.000199759475443564,
      "loss": 0.9078041076660156,
      "step": 3840
    },
    {
      "epoch": 0.05157211221287896,
      "grad_norm": 0.886355996131897,
      "learning_rate": 0.0001997564588246318,
      "loss": 0.8979540824890136,
      "step": 3850
    },
    {
      "epoch": 0.05170606575109423,
      "grad_norm": 0.8408886194229126,
      "learning_rate": 0.00019975342342950044,
      "loss": 0.900606346130371,
      "step": 3860
    },
    {
      "epoch": 0.0518400192893095,
      "grad_norm": 0.881829023361206,
      "learning_rate": 0.0001997503692587413,
      "loss": 0.8997422218322754,
      "step": 3870
    },
    {
      "epoch": 0.05197397282752477,
      "grad_norm": 1.0530098676681519,
      "learning_rate": 0.0001997472963129292,
      "loss": 0.9246989250183105,
      "step": 3880
    },
    {
      "epoch": 0.05210792636574004,
      "grad_norm": 0.9508712291717529,
      "learning_rate": 0.0001997442045926426,
      "loss": 0.9115647315979004,
      "step": 3890
    },
    {
      "epoch": 0.05224187990395531,
      "grad_norm": 0.9167556166648865,
      "learning_rate": 0.00019974109409846337,
      "loss": 0.9261734962463379,
      "step": 3900
    },
    {
      "epoch": 0.05237583344217058,
      "grad_norm": 0.9922856092453003,
      "learning_rate": 0.00019973796483097696,
      "loss": 0.9105630874633789,
      "step": 3910
    },
    {
      "epoch": 0.05250978698038585,
      "grad_norm": 0.9163697957992554,
      "learning_rate": 0.00019973481679077238,
      "loss": 0.9143681526184082,
      "step": 3920
    },
    {
      "epoch": 0.05264374051860112,
      "grad_norm": 0.8309245109558105,
      "learning_rate": 0.0001997316499784422,
      "loss": 0.9261678695678711,
      "step": 3930
    },
    {
      "epoch": 0.05277769405681639,
      "grad_norm": 1.0028313398361206,
      "learning_rate": 0.0001997284643945824,
      "loss": 0.8874808311462402,
      "step": 3940
    },
    {
      "epoch": 0.05291164759503166,
      "grad_norm": 0.9336344003677368,
      "learning_rate": 0.0001997252600397926,
      "loss": 0.9186717987060546,
      "step": 3950
    },
    {
      "epoch": 0.05304560113324693,
      "grad_norm": 0.9435511231422424,
      "learning_rate": 0.00019972203691467596,
      "loss": 0.9015359878540039,
      "step": 3960
    },
    {
      "epoch": 0.0531795546714622,
      "grad_norm": 0.8353652954101562,
      "learning_rate": 0.0001997187950198391,
      "loss": 0.9050702095031739,
      "step": 3970
    },
    {
      "epoch": 0.05331350820967747,
      "grad_norm": 0.9390276074409485,
      "learning_rate": 0.00019971553435589224,
      "loss": 0.9628662109375,
      "step": 3980
    },
    {
      "epoch": 0.05344746174789274,
      "grad_norm": 0.9069141149520874,
      "learning_rate": 0.00019971225492344905,
      "loss": 0.8978482246398926,
      "step": 3990
    },
    {
      "epoch": 0.05358141528610801,
      "grad_norm": 0.9027342796325684,
      "learning_rate": 0.00019970895672312685,
      "loss": 0.9324676513671875,
      "step": 4000
    },
    {
      "epoch": 0.05358141528610801,
      "eval_loss": 0.8909962773323059,
      "eval_runtime": 282.9255,
      "eval_samples_per_second": 10.649,
      "eval_steps_per_second": 10.649,
      "step": 4000
    },
    {
      "epoch": 0.05371536882432328,
      "grad_norm": 0.8347414135932922,
      "learning_rate": 0.0001997056397555464,
      "loss": 0.9355817794799804,
      "step": 4010
    },
    {
      "epoch": 0.05384932236253855,
      "grad_norm": 0.9647659063339233,
      "learning_rate": 0.000199702304021332,
      "loss": 0.9352194786071777,
      "step": 4020
    },
    {
      "epoch": 0.05398327590075382,
      "grad_norm": 0.9430063962936401,
      "learning_rate": 0.0001996989495211115,
      "loss": 0.9100798606872559,
      "step": 4030
    },
    {
      "epoch": 0.05411722943896909,
      "grad_norm": 0.8570693731307983,
      "learning_rate": 0.00019969557625551635,
      "loss": 0.8997575759887695,
      "step": 4040
    },
    {
      "epoch": 0.05425118297718436,
      "grad_norm": 0.8614152073860168,
      "learning_rate": 0.0001996921842251814,
      "loss": 0.919250202178955,
      "step": 4050
    },
    {
      "epoch": 0.05438513651539963,
      "grad_norm": 1.0438923835754395,
      "learning_rate": 0.00019968877343074515,
      "loss": 0.9120957374572753,
      "step": 4060
    },
    {
      "epoch": 0.0545190900536149,
      "grad_norm": 0.8420084118843079,
      "learning_rate": 0.00019968534387284956,
      "loss": 0.9383444786071777,
      "step": 4070
    },
    {
      "epoch": 0.054653043591830174,
      "grad_norm": 0.9416694641113281,
      "learning_rate": 0.00019968189555214012,
      "loss": 0.9132110595703125,
      "step": 4080
    },
    {
      "epoch": 0.054786997130045444,
      "grad_norm": 0.9215363264083862,
      "learning_rate": 0.0001996784284692659,
      "loss": 0.9173206329345703,
      "step": 4090
    },
    {
      "epoch": 0.054920950668260714,
      "grad_norm": 1.1527929306030273,
      "learning_rate": 0.00019967494262487946,
      "loss": 0.9360475540161133,
      "step": 4100
    },
    {
      "epoch": 0.055054904206475984,
      "grad_norm": 0.9070743322372437,
      "learning_rate": 0.00019967143801963693,
      "loss": 0.9068495750427246,
      "step": 4110
    },
    {
      "epoch": 0.055188857744691254,
      "grad_norm": 0.8676186800003052,
      "learning_rate": 0.00019966791465419795,
      "loss": 0.9177586555480957,
      "step": 4120
    },
    {
      "epoch": 0.055322811282906524,
      "grad_norm": 0.9942190051078796,
      "learning_rate": 0.00019966437252922563,
      "loss": 0.937378215789795,
      "step": 4130
    },
    {
      "epoch": 0.055456764821121794,
      "grad_norm": 1.0533899068832397,
      "learning_rate": 0.00019966081164538672,
      "loss": 0.9167658805847168,
      "step": 4140
    },
    {
      "epoch": 0.055590718359337064,
      "grad_norm": 0.9439006447792053,
      "learning_rate": 0.00019965723200335146,
      "loss": 0.9186976432800293,
      "step": 4150
    },
    {
      "epoch": 0.055724671897552334,
      "grad_norm": 0.8586645126342773,
      "learning_rate": 0.0001996536336037936,
      "loss": 0.9189139366149902,
      "step": 4160
    },
    {
      "epoch": 0.055858625435767605,
      "grad_norm": 0.9884462952613831,
      "learning_rate": 0.00019965001644739043,
      "loss": 0.9102540969848633,
      "step": 4170
    },
    {
      "epoch": 0.055992578973982875,
      "grad_norm": 0.8929790258407593,
      "learning_rate": 0.00019964638053482277,
      "loss": 0.9131571769714355,
      "step": 4180
    },
    {
      "epoch": 0.056126532512198145,
      "grad_norm": 0.8665063977241516,
      "learning_rate": 0.00019964272586677498,
      "loss": 0.9067962646484375,
      "step": 4190
    },
    {
      "epoch": 0.056260486050413415,
      "grad_norm": 0.8934453129768372,
      "learning_rate": 0.0001996390524439349,
      "loss": 0.9221485137939454,
      "step": 4200
    },
    {
      "epoch": 0.056394439588628685,
      "grad_norm": 1.073230504989624,
      "learning_rate": 0.000199635360266994,
      "loss": 0.9390836715698242,
      "step": 4210
    },
    {
      "epoch": 0.056528393126843955,
      "grad_norm": 0.9033562541007996,
      "learning_rate": 0.00019963164933664722,
      "loss": 0.9366939544677735,
      "step": 4220
    },
    {
      "epoch": 0.056662346665059225,
      "grad_norm": 1.003556489944458,
      "learning_rate": 0.000199627919653593,
      "loss": 0.9272829055786133,
      "step": 4230
    },
    {
      "epoch": 0.056796300203274495,
      "grad_norm": 0.8745154738426208,
      "learning_rate": 0.00019962417121853333,
      "loss": 0.9143185615539551,
      "step": 4240
    },
    {
      "epoch": 0.056930253741489766,
      "grad_norm": 0.8166207671165466,
      "learning_rate": 0.0001996204040321738,
      "loss": 0.9050333976745606,
      "step": 4250
    },
    {
      "epoch": 0.057064207279705036,
      "grad_norm": 0.9641032218933105,
      "learning_rate": 0.00019961661809522348,
      "loss": 0.9403782844543457,
      "step": 4260
    },
    {
      "epoch": 0.057198160817920306,
      "grad_norm": 0.9547359347343445,
      "learning_rate": 0.00019961281340839487,
      "loss": 0.9166964530944824,
      "step": 4270
    },
    {
      "epoch": 0.057332114356135576,
      "grad_norm": 0.881603479385376,
      "learning_rate": 0.00019960898997240415,
      "loss": 0.8962611198425293,
      "step": 4280
    },
    {
      "epoch": 0.057466067894350846,
      "grad_norm": 0.9667623043060303,
      "learning_rate": 0.00019960514778797096,
      "loss": 0.9078075408935546,
      "step": 4290
    },
    {
      "epoch": 0.057600021432566116,
      "grad_norm": 0.9174531698226929,
      "learning_rate": 0.0001996012868558185,
      "loss": 0.906414794921875,
      "step": 4300
    },
    {
      "epoch": 0.057733974970781386,
      "grad_norm": 0.9098462462425232,
      "learning_rate": 0.0001995974071766734,
      "loss": 0.9087407112121582,
      "step": 4310
    },
    {
      "epoch": 0.057867928508996656,
      "grad_norm": 0.9336055517196655,
      "learning_rate": 0.000199593508751266,
      "loss": 0.9243258476257324,
      "step": 4320
    },
    {
      "epoch": 0.058001882047211926,
      "grad_norm": 0.9590570330619812,
      "learning_rate": 0.00019958959158033,
      "loss": 0.9220330238342285,
      "step": 4330
    },
    {
      "epoch": 0.0581358355854272,
      "grad_norm": 0.9370716214179993,
      "learning_rate": 0.00019958565566460267,
      "loss": 0.9094621658325195,
      "step": 4340
    },
    {
      "epoch": 0.05826978912364247,
      "grad_norm": 0.8216221928596497,
      "learning_rate": 0.0001995817010048249,
      "loss": 0.9052978515625,
      "step": 4350
    },
    {
      "epoch": 0.05840374266185774,
      "grad_norm": 0.9418812990188599,
      "learning_rate": 0.00019957772760174097,
      "loss": 0.8963290214538574,
      "step": 4360
    },
    {
      "epoch": 0.05853769620007301,
      "grad_norm": 0.8006338477134705,
      "learning_rate": 0.0001995737354560988,
      "loss": 0.9135848999023437,
      "step": 4370
    },
    {
      "epoch": 0.05867164973828828,
      "grad_norm": 1.0015288591384888,
      "learning_rate": 0.00019956972456864976,
      "loss": 0.894831371307373,
      "step": 4380
    },
    {
      "epoch": 0.05880560327650355,
      "grad_norm": 0.807648241519928,
      "learning_rate": 0.00019956569494014885,
      "loss": 0.9128003120422363,
      "step": 4390
    },
    {
      "epoch": 0.05893955681471882,
      "grad_norm": 0.8862670063972473,
      "learning_rate": 0.00019956164657135444,
      "loss": 0.8800803184509277,
      "step": 4400
    },
    {
      "epoch": 0.05907351035293409,
      "grad_norm": 0.8406708836555481,
      "learning_rate": 0.00019955757946302859,
      "loss": 0.9087419509887695,
      "step": 4410
    },
    {
      "epoch": 0.05920746389114936,
      "grad_norm": 1.0521955490112305,
      "learning_rate": 0.00019955349361593672,
      "loss": 0.9132774353027344,
      "step": 4420
    },
    {
      "epoch": 0.05934141742936463,
      "grad_norm": 0.9215775728225708,
      "learning_rate": 0.00019954938903084794,
      "loss": 0.9055030822753907,
      "step": 4430
    },
    {
      "epoch": 0.0594753709675799,
      "grad_norm": 0.9627100825309753,
      "learning_rate": 0.00019954526570853482,
      "loss": 0.8979693412780761,
      "step": 4440
    },
    {
      "epoch": 0.05960932450579517,
      "grad_norm": 0.9040695428848267,
      "learning_rate": 0.0001995411236497734,
      "loss": 0.9099948883056641,
      "step": 4450
    },
    {
      "epoch": 0.05974327804401044,
      "grad_norm": 0.8744531273841858,
      "learning_rate": 0.00019953696285534338,
      "loss": 0.9203990936279297,
      "step": 4460
    },
    {
      "epoch": 0.05987723158222571,
      "grad_norm": 0.8961984515190125,
      "learning_rate": 0.00019953278332602786,
      "loss": 0.883176326751709,
      "step": 4470
    },
    {
      "epoch": 0.06001118512044098,
      "grad_norm": 0.9887130856513977,
      "learning_rate": 0.0001995285850626135,
      "loss": 0.9187640190124512,
      "step": 4480
    },
    {
      "epoch": 0.06014513865865625,
      "grad_norm": 0.9350399374961853,
      "learning_rate": 0.00019952436806589048,
      "loss": 0.9422298431396484,
      "step": 4490
    },
    {
      "epoch": 0.06027909219687152,
      "grad_norm": 0.978025496006012,
      "learning_rate": 0.00019952013233665258,
      "loss": 0.903172779083252,
      "step": 4500
    },
    {
      "epoch": 0.06041304573508679,
      "grad_norm": 0.8744170665740967,
      "learning_rate": 0.00019951587787569707,
      "loss": 0.8885890960693359,
      "step": 4510
    },
    {
      "epoch": 0.06054699927330205,
      "grad_norm": 0.9057816863059998,
      "learning_rate": 0.0001995116046838246,
      "loss": 0.9105055809020997,
      "step": 4520
    },
    {
      "epoch": 0.06068095281151732,
      "grad_norm": 0.9650569558143616,
      "learning_rate": 0.00019950731276183965,
      "loss": 0.9167463302612304,
      "step": 4530
    },
    {
      "epoch": 0.06081490634973259,
      "grad_norm": 0.7593880891799927,
      "learning_rate": 0.0001995030021105499,
      "loss": 0.9327964782714844,
      "step": 4540
    },
    {
      "epoch": 0.06094885988794786,
      "grad_norm": 0.8758352994918823,
      "learning_rate": 0.00019949867273076676,
      "loss": 0.9022039413452149,
      "step": 4550
    },
    {
      "epoch": 0.06108281342616313,
      "grad_norm": 0.8169499635696411,
      "learning_rate": 0.0001994943246233051,
      "loss": 0.8898695945739746,
      "step": 4560
    },
    {
      "epoch": 0.0612167669643784,
      "grad_norm": 0.8713719844818115,
      "learning_rate": 0.00019948995778898334,
      "loss": 0.9046549797058105,
      "step": 4570
    },
    {
      "epoch": 0.06135072050259367,
      "grad_norm": 0.9203425049781799,
      "learning_rate": 0.00019948557222862338,
      "loss": 0.9264613151550293,
      "step": 4580
    },
    {
      "epoch": 0.06148467404080894,
      "grad_norm": 0.9501252174377441,
      "learning_rate": 0.00019948116794305068,
      "loss": 0.9253769874572754,
      "step": 4590
    },
    {
      "epoch": 0.06161862757902421,
      "grad_norm": 0.9844063520431519,
      "learning_rate": 0.00019947674493309424,
      "loss": 0.9189377784729004,
      "step": 4600
    },
    {
      "epoch": 0.06175258111723948,
      "grad_norm": 0.9375742673873901,
      "learning_rate": 0.00019947230319958658,
      "loss": 0.9230340957641602,
      "step": 4610
    },
    {
      "epoch": 0.06188653465545475,
      "grad_norm": 0.9792372584342957,
      "learning_rate": 0.0001994678427433636,
      "loss": 0.9204941749572754,
      "step": 4620
    },
    {
      "epoch": 0.06202048819367002,
      "grad_norm": 0.8661828637123108,
      "learning_rate": 0.00019946336356526498,
      "loss": 0.9278335571289062,
      "step": 4630
    },
    {
      "epoch": 0.06215444173188529,
      "grad_norm": 0.9424053430557251,
      "learning_rate": 0.00019945886566613376,
      "loss": 0.8783210754394531,
      "step": 4640
    },
    {
      "epoch": 0.06228839527010056,
      "grad_norm": 1.065873622894287,
      "learning_rate": 0.0001994543490468165,
      "loss": 0.8889184951782226,
      "step": 4650
    },
    {
      "epoch": 0.06242234880831583,
      "grad_norm": 0.8346952795982361,
      "learning_rate": 0.00019944981370816336,
      "loss": 0.9113599777221679,
      "step": 4660
    },
    {
      "epoch": 0.0625563023465311,
      "grad_norm": 0.8639548420906067,
      "learning_rate": 0.00019944525965102797,
      "loss": 0.8904115676879882,
      "step": 4670
    },
    {
      "epoch": 0.06269025588474637,
      "grad_norm": 0.8392593264579773,
      "learning_rate": 0.00019944068687626748,
      "loss": 0.9148737907409668,
      "step": 4680
    },
    {
      "epoch": 0.06282420942296164,
      "grad_norm": 0.9527535438537598,
      "learning_rate": 0.0001994360953847426,
      "loss": 0.9055316925048829,
      "step": 4690
    },
    {
      "epoch": 0.06295816296117691,
      "grad_norm": 0.8947551250457764,
      "learning_rate": 0.00019943148517731753,
      "loss": 0.908508014678955,
      "step": 4700
    },
    {
      "epoch": 0.06309211649939218,
      "grad_norm": 0.8721895217895508,
      "learning_rate": 0.00019942685625486,
      "loss": 0.9218954086303711,
      "step": 4710
    },
    {
      "epoch": 0.06322607003760745,
      "grad_norm": 0.9207981824874878,
      "learning_rate": 0.0001994222086182413,
      "loss": 0.9213238716125488,
      "step": 4720
    },
    {
      "epoch": 0.06336002357582272,
      "grad_norm": 0.9650294780731201,
      "learning_rate": 0.00019941754226833618,
      "loss": 0.9041328430175781,
      "step": 4730
    },
    {
      "epoch": 0.063493977114038,
      "grad_norm": 0.9149458408355713,
      "learning_rate": 0.00019941285720602293,
      "loss": 0.9019318580627441,
      "step": 4740
    },
    {
      "epoch": 0.06362793065225326,
      "grad_norm": 0.8800939321517944,
      "learning_rate": 0.00019940815343218345,
      "loss": 0.9098714828491211,
      "step": 4750
    },
    {
      "epoch": 0.06376188419046853,
      "grad_norm": 0.8645089864730835,
      "learning_rate": 0.000199403430947703,
      "loss": 0.9096037864685058,
      "step": 4760
    },
    {
      "epoch": 0.0638958377286838,
      "grad_norm": 0.9408107399940491,
      "learning_rate": 0.00019939868975347045,
      "loss": 0.9122167587280273,
      "step": 4770
    },
    {
      "epoch": 0.06402979126689907,
      "grad_norm": 0.8798220157623291,
      "learning_rate": 0.00019939392985037825,
      "loss": 0.8967701911926269,
      "step": 4780
    },
    {
      "epoch": 0.06416374480511434,
      "grad_norm": 0.9887925386428833,
      "learning_rate": 0.00019938915123932227,
      "loss": 0.9241874694824219,
      "step": 4790
    },
    {
      "epoch": 0.06429769834332962,
      "grad_norm": 0.9510945081710815,
      "learning_rate": 0.00019938435392120196,
      "loss": 0.8980023384094238,
      "step": 4800
    },
    {
      "epoch": 0.06443165188154489,
      "grad_norm": 0.8272057771682739,
      "learning_rate": 0.00019937953789692025,
      "loss": 0.8926514625549317,
      "step": 4810
    },
    {
      "epoch": 0.06456560541976016,
      "grad_norm": 0.8406584858894348,
      "learning_rate": 0.00019937470316738366,
      "loss": 0.9133185386657715,
      "step": 4820
    },
    {
      "epoch": 0.06469955895797543,
      "grad_norm": 0.9292959570884705,
      "learning_rate": 0.0001993698497335021,
      "loss": 0.9174582481384277,
      "step": 4830
    },
    {
      "epoch": 0.0648335124961907,
      "grad_norm": 0.9233940243721008,
      "learning_rate": 0.00019936497759618916,
      "loss": 0.9082266807556152,
      "step": 4840
    },
    {
      "epoch": 0.06496746603440597,
      "grad_norm": 0.9036064147949219,
      "learning_rate": 0.00019936008675636188,
      "loss": 0.9060539245605469,
      "step": 4850
    },
    {
      "epoch": 0.06510141957262124,
      "grad_norm": 0.9129693508148193,
      "learning_rate": 0.00019935517721494074,
      "loss": 0.9315420150756836,
      "step": 4860
    },
    {
      "epoch": 0.0652353731108365,
      "grad_norm": 0.9471613168716431,
      "learning_rate": 0.0001993502489728499,
      "loss": 0.9134193420410156,
      "step": 4870
    },
    {
      "epoch": 0.06536932664905178,
      "grad_norm": 1.0248454809188843,
      "learning_rate": 0.0001993453020310169,
      "loss": 0.8997479438781738,
      "step": 4880
    },
    {
      "epoch": 0.06550328018726705,
      "grad_norm": 1.078979253768921,
      "learning_rate": 0.00019934033639037289,
      "loss": 0.9184505462646484,
      "step": 4890
    },
    {
      "epoch": 0.06563723372548232,
      "grad_norm": 0.8346269726753235,
      "learning_rate": 0.0001993353520518525,
      "loss": 0.9187174797058105,
      "step": 4900
    },
    {
      "epoch": 0.06577118726369759,
      "grad_norm": 0.8458219170570374,
      "learning_rate": 0.00019933034901639383,
      "loss": 0.9144869804382324,
      "step": 4910
    },
    {
      "epoch": 0.06590514080191286,
      "grad_norm": 0.8314769864082336,
      "learning_rate": 0.00019932532728493863,
      "loss": 0.9098584175109863,
      "step": 4920
    },
    {
      "epoch": 0.06603909434012813,
      "grad_norm": 0.9744229316711426,
      "learning_rate": 0.00019932028685843206,
      "loss": 0.9002020835876465,
      "step": 4930
    },
    {
      "epoch": 0.0661730478783434,
      "grad_norm": 0.8311421871185303,
      "learning_rate": 0.00019931522773782283,
      "loss": 0.908812141418457,
      "step": 4940
    },
    {
      "epoch": 0.06630700141655867,
      "grad_norm": 0.9931102395057678,
      "learning_rate": 0.0001993101499240632,
      "loss": 0.8970493316650391,
      "step": 4950
    },
    {
      "epoch": 0.06644095495477394,
      "grad_norm": 0.8851704001426697,
      "learning_rate": 0.00019930505341810885,
      "loss": 0.9012079238891602,
      "step": 4960
    },
    {
      "epoch": 0.06657490849298921,
      "grad_norm": 0.8322035074234009,
      "learning_rate": 0.0001992999382209191,
      "loss": 0.9162311553955078,
      "step": 4970
    },
    {
      "epoch": 0.06670886203120448,
      "grad_norm": 0.923004150390625,
      "learning_rate": 0.0001992948043334567,
      "loss": 0.9190991401672364,
      "step": 4980
    },
    {
      "epoch": 0.06684281556941975,
      "grad_norm": 1.058236002922058,
      "learning_rate": 0.00019928965175668796,
      "loss": 0.8755815505981446,
      "step": 4990
    },
    {
      "epoch": 0.06697676910763502,
      "grad_norm": 0.9285317063331604,
      "learning_rate": 0.00019928448049158276,
      "loss": 0.9065299034118652,
      "step": 5000
    },
    {
      "epoch": 0.06697676910763502,
      "eval_loss": 0.8810214400291443,
      "eval_runtime": 283.0005,
      "eval_samples_per_second": 10.647,
      "eval_steps_per_second": 10.647,
      "step": 5000
    },
    {
      "epoch": 0.06711072264585029,
      "grad_norm": 0.8535721302032471,
      "learning_rate": 0.00019927929053911435,
      "loss": 0.9006348609924316,
      "step": 5010
    },
    {
      "epoch": 0.06724467618406556,
      "grad_norm": 0.8682570457458496,
      "learning_rate": 0.00019927408190025965,
      "loss": 0.8905268669128418,
      "step": 5020
    },
    {
      "epoch": 0.06737862972228083,
      "grad_norm": 0.905788242816925,
      "learning_rate": 0.000199268854575999,
      "loss": 0.9189523696899414,
      "step": 5030
    },
    {
      "epoch": 0.0675125832604961,
      "grad_norm": 0.9985746741294861,
      "learning_rate": 0.00019926360856731628,
      "loss": 0.9133862495422364,
      "step": 5040
    },
    {
      "epoch": 0.06764653679871137,
      "grad_norm": 0.9106592535972595,
      "learning_rate": 0.00019925834387519891,
      "loss": 0.8930727005004883,
      "step": 5050
    },
    {
      "epoch": 0.06778049033692664,
      "grad_norm": 0.8151264190673828,
      "learning_rate": 0.00019925306050063782,
      "loss": 0.9098219871520996,
      "step": 5060
    },
    {
      "epoch": 0.06791444387514191,
      "grad_norm": 0.8801676034927368,
      "learning_rate": 0.00019924775844462747,
      "loss": 0.9096726417541504,
      "step": 5070
    },
    {
      "epoch": 0.06804839741335718,
      "grad_norm": 0.8600324988365173,
      "learning_rate": 0.00019924243770816577,
      "loss": 0.8851785659790039,
      "step": 5080
    },
    {
      "epoch": 0.06818235095157245,
      "grad_norm": 0.8714832067489624,
      "learning_rate": 0.0001992370982922542,
      "loss": 0.9042956352233886,
      "step": 5090
    },
    {
      "epoch": 0.06831630448978772,
      "grad_norm": 0.883526623249054,
      "learning_rate": 0.00019923174019789776,
      "loss": 0.888262939453125,
      "step": 5100
    },
    {
      "epoch": 0.06845025802800299,
      "grad_norm": 0.8125107288360596,
      "learning_rate": 0.00019922636342610496,
      "loss": 0.9034280776977539,
      "step": 5110
    },
    {
      "epoch": 0.06858421156621826,
      "grad_norm": 0.8528799414634705,
      "learning_rate": 0.00019922096797788777,
      "loss": 0.9065326690673828,
      "step": 5120
    },
    {
      "epoch": 0.06871816510443353,
      "grad_norm": 0.812098503112793,
      "learning_rate": 0.00019921555385426175,
      "loss": 0.8851298332214356,
      "step": 5130
    },
    {
      "epoch": 0.0688521186426488,
      "grad_norm": 0.8713356256484985,
      "learning_rate": 0.000199210121056246,
      "loss": 0.910951042175293,
      "step": 5140
    },
    {
      "epoch": 0.06898607218086407,
      "grad_norm": 0.8418512940406799,
      "learning_rate": 0.00019920466958486303,
      "loss": 0.8865553855895996,
      "step": 5150
    },
    {
      "epoch": 0.06912002571907934,
      "grad_norm": 0.9700740575790405,
      "learning_rate": 0.00019919919944113895,
      "loss": 0.8811507225036621,
      "step": 5160
    },
    {
      "epoch": 0.06925397925729461,
      "grad_norm": 0.8555593490600586,
      "learning_rate": 0.0001991937106261033,
      "loss": 0.9122533798217773,
      "step": 5170
    },
    {
      "epoch": 0.06938793279550988,
      "grad_norm": 0.8585926294326782,
      "learning_rate": 0.00019918820314078923,
      "loss": 0.9141427993774414,
      "step": 5180
    },
    {
      "epoch": 0.06952188633372515,
      "grad_norm": 0.9098888039588928,
      "learning_rate": 0.00019918267698623333,
      "loss": 0.9057223320007324,
      "step": 5190
    },
    {
      "epoch": 0.06965583987194042,
      "grad_norm": 1.4986580610275269,
      "learning_rate": 0.00019917713216347575,
      "loss": 0.8974419593811035,
      "step": 5200
    },
    {
      "epoch": 0.06978979341015569,
      "grad_norm": 0.8064653277397156,
      "learning_rate": 0.0001991715686735602,
      "loss": 0.8973417282104492,
      "step": 5210
    },
    {
      "epoch": 0.06992374694837096,
      "grad_norm": 0.9617411494255066,
      "learning_rate": 0.00019916598651753371,
      "loss": 0.9342555999755859,
      "step": 5220
    },
    {
      "epoch": 0.07005770048658623,
      "grad_norm": 1.0437543392181396,
      "learning_rate": 0.0001991603856964471,
      "loss": 0.9038956642150879,
      "step": 5230
    },
    {
      "epoch": 0.0701916540248015,
      "grad_norm": 0.7741673588752747,
      "learning_rate": 0.00019915476621135443,
      "loss": 0.8806075096130371,
      "step": 5240
    },
    {
      "epoch": 0.07032560756301677,
      "grad_norm": 0.872779369354248,
      "learning_rate": 0.0001991491280633135,
      "loss": 0.8836995124816894,
      "step": 5250
    },
    {
      "epoch": 0.07045956110123204,
      "grad_norm": 1.0274291038513184,
      "learning_rate": 0.00019914347125338545,
      "loss": 0.8840160369873047,
      "step": 5260
    },
    {
      "epoch": 0.07059351463944731,
      "grad_norm": 0.9586661458015442,
      "learning_rate": 0.00019913779578263507,
      "loss": 0.8941290855407715,
      "step": 5270
    },
    {
      "epoch": 0.07072746817766258,
      "grad_norm": 0.8342275619506836,
      "learning_rate": 0.0001991321016521306,
      "loss": 0.8942935943603516,
      "step": 5280
    },
    {
      "epoch": 0.07086142171587785,
      "grad_norm": 0.8630613088607788,
      "learning_rate": 0.00019912638886294373,
      "loss": 0.9337220191955566,
      "step": 5290
    },
    {
      "epoch": 0.07099537525409312,
      "grad_norm": 0.8034462332725525,
      "learning_rate": 0.00019912065741614973,
      "loss": 0.9068234443664551,
      "step": 5300
    },
    {
      "epoch": 0.07112932879230839,
      "grad_norm": 0.818018913269043,
      "learning_rate": 0.00019911490731282743,
      "loss": 0.9113452911376954,
      "step": 5310
    },
    {
      "epoch": 0.07126328233052366,
      "grad_norm": 1.0437874794006348,
      "learning_rate": 0.0001991091385540591,
      "loss": 0.8976190567016602,
      "step": 5320
    },
    {
      "epoch": 0.07139723586873893,
      "grad_norm": 0.877898097038269,
      "learning_rate": 0.0001991033511409305,
      "loss": 0.8937252044677735,
      "step": 5330
    },
    {
      "epoch": 0.0715311894069542,
      "grad_norm": 0.8473853468894958,
      "learning_rate": 0.00019909754507453097,
      "loss": 0.9105791091918946,
      "step": 5340
    },
    {
      "epoch": 0.07166514294516947,
      "grad_norm": 0.9755077958106995,
      "learning_rate": 0.00019909172035595334,
      "loss": 0.8925551414489746,
      "step": 5350
    },
    {
      "epoch": 0.07179909648338474,
      "grad_norm": 1.1288537979125977,
      "learning_rate": 0.0001990858769862939,
      "loss": 0.9144526481628418,
      "step": 5360
    },
    {
      "epoch": 0.07193305002160001,
      "grad_norm": 0.8881911635398865,
      "learning_rate": 0.00019908001496665256,
      "loss": 0.9116629600524903,
      "step": 5370
    },
    {
      "epoch": 0.07206700355981528,
      "grad_norm": 1.008434534072876,
      "learning_rate": 0.0001990741342981326,
      "loss": 0.8908850669860839,
      "step": 5380
    },
    {
      "epoch": 0.07220095709803055,
      "grad_norm": 0.924994170665741,
      "learning_rate": 0.0001990682349818409,
      "loss": 0.892117691040039,
      "step": 5390
    },
    {
      "epoch": 0.07233491063624582,
      "grad_norm": 1.0123432874679565,
      "learning_rate": 0.00019906231701888788,
      "loss": 0.8864407539367676,
      "step": 5400
    },
    {
      "epoch": 0.07246886417446109,
      "grad_norm": 0.8557733297348022,
      "learning_rate": 0.00019905638041038735,
      "loss": 0.8893271446228027,
      "step": 5410
    },
    {
      "epoch": 0.07260281771267636,
      "grad_norm": 0.8255430459976196,
      "learning_rate": 0.00019905042515745675,
      "loss": 0.9042425155639648,
      "step": 5420
    },
    {
      "epoch": 0.07273677125089163,
      "grad_norm": 1.0060651302337646,
      "learning_rate": 0.00019904445126121695,
      "loss": 0.8978655815124512,
      "step": 5430
    },
    {
      "epoch": 0.0728707247891069,
      "grad_norm": 0.8825175166130066,
      "learning_rate": 0.00019903845872279238,
      "loss": 0.8874133110046387,
      "step": 5440
    },
    {
      "epoch": 0.07300467832732217,
      "grad_norm": 0.8155242800712585,
      "learning_rate": 0.00019903244754331095,
      "loss": 0.9240117073059082,
      "step": 5450
    },
    {
      "epoch": 0.07313863186553744,
      "grad_norm": 0.9396053552627563,
      "learning_rate": 0.00019902641772390412,
      "loss": 0.9122659683227539,
      "step": 5460
    },
    {
      "epoch": 0.07327258540375271,
      "grad_norm": 0.8111370801925659,
      "learning_rate": 0.00019902036926570675,
      "loss": 0.8991929054260254,
      "step": 5470
    },
    {
      "epoch": 0.07340653894196798,
      "grad_norm": 0.7930305600166321,
      "learning_rate": 0.00019901430216985735,
      "loss": 0.8710382461547852,
      "step": 5480
    },
    {
      "epoch": 0.07354049248018325,
      "grad_norm": 0.8627564311027527,
      "learning_rate": 0.00019900821643749784,
      "loss": 0.9189313888549805,
      "step": 5490
    },
    {
      "epoch": 0.07367444601839852,
      "grad_norm": 0.8343905806541443,
      "learning_rate": 0.00019900211206977367,
      "loss": 0.8809655189514161,
      "step": 5500
    },
    {
      "epoch": 0.0738083995566138,
      "grad_norm": 0.8543509244918823,
      "learning_rate": 0.00019899598906783386,
      "loss": 0.9055264472961426,
      "step": 5510
    },
    {
      "epoch": 0.07394235309482906,
      "grad_norm": 0.8609879016876221,
      "learning_rate": 0.0001989898474328308,
      "loss": 0.8854434013366699,
      "step": 5520
    },
    {
      "epoch": 0.07407630663304433,
      "grad_norm": 0.9053690433502197,
      "learning_rate": 0.00019898368716592056,
      "loss": 0.8991393089294434,
      "step": 5530
    },
    {
      "epoch": 0.0742102601712596,
      "grad_norm": 0.9232255220413208,
      "learning_rate": 0.00019897750826826255,
      "loss": 0.9063379287719726,
      "step": 5540
    },
    {
      "epoch": 0.07434421370947487,
      "grad_norm": 0.8195554614067078,
      "learning_rate": 0.00019897131074101982,
      "loss": 0.8966815948486329,
      "step": 5550
    },
    {
      "epoch": 0.07447816724769014,
      "grad_norm": 0.8728640079498291,
      "learning_rate": 0.00019896509458535886,
      "loss": 0.9156026840209961,
      "step": 5560
    },
    {
      "epoch": 0.07461212078590541,
      "grad_norm": 1.0263285636901855,
      "learning_rate": 0.00019895885980244964,
      "loss": 0.9155710220336915,
      "step": 5570
    },
    {
      "epoch": 0.07474607432412068,
      "grad_norm": 0.9898778200149536,
      "learning_rate": 0.0001989526063934657,
      "loss": 0.9170265197753906,
      "step": 5580
    },
    {
      "epoch": 0.07488002786233595,
      "grad_norm": 0.9437372088432312,
      "learning_rate": 0.00019894633435958408,
      "loss": 0.9208130836486816,
      "step": 5590
    },
    {
      "epoch": 0.07501398140055122,
      "grad_norm": 0.8335493803024292,
      "learning_rate": 0.00019894004370198527,
      "loss": 0.9066865921020508,
      "step": 5600
    },
    {
      "epoch": 0.0751479349387665,
      "grad_norm": 0.9112666845321655,
      "learning_rate": 0.00019893373442185332,
      "loss": 0.9198005676269532,
      "step": 5610
    },
    {
      "epoch": 0.07528188847698176,
      "grad_norm": 0.85235595703125,
      "learning_rate": 0.00019892740652037577,
      "loss": 0.9002386093139648,
      "step": 5620
    },
    {
      "epoch": 0.07541584201519703,
      "grad_norm": 0.9408565163612366,
      "learning_rate": 0.00019892105999874362,
      "loss": 0.8927816390991211,
      "step": 5630
    },
    {
      "epoch": 0.0755497955534123,
      "grad_norm": 0.943725049495697,
      "learning_rate": 0.00019891469485815147,
      "loss": 0.9069494247436524,
      "step": 5640
    },
    {
      "epoch": 0.07568374909162758,
      "grad_norm": 0.8113754987716675,
      "learning_rate": 0.00019890831109979735,
      "loss": 0.9259742736816406,
      "step": 5650
    },
    {
      "epoch": 0.07581770262984285,
      "grad_norm": 0.8559271693229675,
      "learning_rate": 0.0001989019087248828,
      "loss": 0.8804376602172852,
      "step": 5660
    },
    {
      "epoch": 0.07595165616805812,
      "grad_norm": 0.9119812250137329,
      "learning_rate": 0.00019889548773461288,
      "loss": 0.9078903198242188,
      "step": 5670
    },
    {
      "epoch": 0.07608560970627339,
      "grad_norm": 0.7937560081481934,
      "learning_rate": 0.00019888904813019614,
      "loss": 0.880002498626709,
      "step": 5680
    },
    {
      "epoch": 0.07621956324448866,
      "grad_norm": 0.9108148217201233,
      "learning_rate": 0.00019888258991284467,
      "loss": 0.9070451736450196,
      "step": 5690
    },
    {
      "epoch": 0.07635351678270393,
      "grad_norm": 0.9098515510559082,
      "learning_rate": 0.00019887611308377405,
      "loss": 0.9075231552124023,
      "step": 5700
    },
    {
      "epoch": 0.0764874703209192,
      "grad_norm": 1.0761123895645142,
      "learning_rate": 0.00019886961764420332,
      "loss": 0.9096757888793945,
      "step": 5710
    },
    {
      "epoch": 0.07662142385913447,
      "grad_norm": 0.7704610228538513,
      "learning_rate": 0.00019886310359535506,
      "loss": 0.8939119338989258,
      "step": 5720
    },
    {
      "epoch": 0.07675537739734974,
      "grad_norm": 0.8866276741027832,
      "learning_rate": 0.00019885657093845534,
      "loss": 0.9135266304016113,
      "step": 5730
    },
    {
      "epoch": 0.076889330935565,
      "grad_norm": 0.9191501140594482,
      "learning_rate": 0.00019885001967473375,
      "loss": 0.8904623985290527,
      "step": 5740
    },
    {
      "epoch": 0.07702328447378026,
      "grad_norm": 0.9312494397163391,
      "learning_rate": 0.00019884344980542338,
      "loss": 0.8734722137451172,
      "step": 5750
    },
    {
      "epoch": 0.07715723801199553,
      "grad_norm": 0.8679017424583435,
      "learning_rate": 0.00019883686133176083,
      "loss": 0.8858957290649414,
      "step": 5760
    },
    {
      "epoch": 0.0772911915502108,
      "grad_norm": 1.0164021253585815,
      "learning_rate": 0.00019883025425498612,
      "loss": 0.9009102821350098,
      "step": 5770
    },
    {
      "epoch": 0.07742514508842607,
      "grad_norm": 0.9356606006622314,
      "learning_rate": 0.0001988236285763429,
      "loss": 0.8916072845458984,
      "step": 5780
    },
    {
      "epoch": 0.07755909862664134,
      "grad_norm": 1.0990995168685913,
      "learning_rate": 0.0001988169842970782,
      "loss": 0.8782832145690918,
      "step": 5790
    },
    {
      "epoch": 0.07769305216485661,
      "grad_norm": 0.8013916015625,
      "learning_rate": 0.0001988103214184427,
      "loss": 0.8873194694519043,
      "step": 5800
    },
    {
      "epoch": 0.07782700570307188,
      "grad_norm": 0.9014554619789124,
      "learning_rate": 0.00019880363994169038,
      "loss": 0.8783865928649902,
      "step": 5810
    },
    {
      "epoch": 0.07796095924128715,
      "grad_norm": 0.9066067337989807,
      "learning_rate": 0.0001987969398680789,
      "loss": 0.9075798034667969,
      "step": 5820
    },
    {
      "epoch": 0.07809491277950242,
      "grad_norm": 0.912848949432373,
      "learning_rate": 0.00019879022119886935,
      "loss": 0.9085654258728028,
      "step": 5830
    },
    {
      "epoch": 0.0782288663177177,
      "grad_norm": 0.8937046527862549,
      "learning_rate": 0.00019878348393532628,
      "loss": 0.8880997657775879,
      "step": 5840
    },
    {
      "epoch": 0.07836281985593296,
      "grad_norm": 0.9600427746772766,
      "learning_rate": 0.00019877672807871782,
      "loss": 0.8978815078735352,
      "step": 5850
    },
    {
      "epoch": 0.07849677339414823,
      "grad_norm": 0.796197772026062,
      "learning_rate": 0.00019876995363031554,
      "loss": 0.9105263710021972,
      "step": 5860
    },
    {
      "epoch": 0.0786307269323635,
      "grad_norm": 0.970268726348877,
      "learning_rate": 0.00019876316059139456,
      "loss": 0.9024666786193848,
      "step": 5870
    },
    {
      "epoch": 0.07876468047057877,
      "grad_norm": 0.879298746585846,
      "learning_rate": 0.0001987563489632334,
      "loss": 0.8722917556762695,
      "step": 5880
    },
    {
      "epoch": 0.07889863400879404,
      "grad_norm": 0.9075465202331543,
      "learning_rate": 0.0001987495187471142,
      "loss": 0.9163165092468262,
      "step": 5890
    },
    {
      "epoch": 0.07903258754700931,
      "grad_norm": 0.9039913415908813,
      "learning_rate": 0.00019874266994432253,
      "loss": 0.8797751426696777,
      "step": 5900
    },
    {
      "epoch": 0.07916654108522458,
      "grad_norm": 0.7920781970024109,
      "learning_rate": 0.0001987358025561475,
      "loss": 0.9124955177307129,
      "step": 5910
    },
    {
      "epoch": 0.07930049462343985,
      "grad_norm": 0.779161810874939,
      "learning_rate": 0.00019872891658388167,
      "loss": 0.9420258522033691,
      "step": 5920
    },
    {
      "epoch": 0.07943444816165512,
      "grad_norm": 0.8104649782180786,
      "learning_rate": 0.00019872201202882108,
      "loss": 0.9008892059326172,
      "step": 5930
    },
    {
      "epoch": 0.0795684016998704,
      "grad_norm": 0.9147506356239319,
      "learning_rate": 0.0001987150888922654,
      "loss": 0.9117377281188965,
      "step": 5940
    },
    {
      "epoch": 0.07970235523808566,
      "grad_norm": 0.8242200016975403,
      "learning_rate": 0.00019870814717551762,
      "loss": 0.8753026962280274,
      "step": 5950
    },
    {
      "epoch": 0.07983630877630093,
      "grad_norm": 0.7932929992675781,
      "learning_rate": 0.00019870118687988436,
      "loss": 0.9308527946472168,
      "step": 5960
    },
    {
      "epoch": 0.0799702623145162,
      "grad_norm": 0.9321459531784058,
      "learning_rate": 0.00019869420800667568,
      "loss": 0.9189542770385742,
      "step": 5970
    },
    {
      "epoch": 0.08010421585273148,
      "grad_norm": 0.9252848625183105,
      "learning_rate": 0.00019868721055720515,
      "loss": 0.8846657752990723,
      "step": 5980
    },
    {
      "epoch": 0.08023816939094675,
      "grad_norm": 0.9462780952453613,
      "learning_rate": 0.00019868019453278982,
      "loss": 0.8895047187805176,
      "step": 5990
    },
    {
      "epoch": 0.08037212292916202,
      "grad_norm": 0.8938540816307068,
      "learning_rate": 0.00019867315993475027,
      "loss": 0.9139046669006348,
      "step": 6000
    },
    {
      "epoch": 0.08037212292916202,
      "eval_loss": 0.870812714099884,
      "eval_runtime": 282.991,
      "eval_samples_per_second": 10.647,
      "eval_steps_per_second": 10.647,
      "step": 6000
    },
    {
      "epoch": 0.08050607646737729,
      "grad_norm": 0.9877068996429443,
      "learning_rate": 0.00019866610676441053,
      "loss": 0.88814697265625,
      "step": 6010
    },
    {
      "epoch": 0.08064003000559256,
      "grad_norm": 0.8751171231269836,
      "learning_rate": 0.00019865903502309815,
      "loss": 0.9264678001403809,
      "step": 6020
    },
    {
      "epoch": 0.08077398354380783,
      "grad_norm": 0.8859562277793884,
      "learning_rate": 0.0001986519447121442,
      "loss": 0.9087067604064941,
      "step": 6030
    },
    {
      "epoch": 0.0809079370820231,
      "grad_norm": 0.9175627827644348,
      "learning_rate": 0.00019864483583288318,
      "loss": 0.8855989456176758,
      "step": 6040
    },
    {
      "epoch": 0.08104189062023837,
      "grad_norm": 0.8336227536201477,
      "learning_rate": 0.0001986377083866532,
      "loss": 0.8984386444091796,
      "step": 6050
    },
    {
      "epoch": 0.08117584415845364,
      "grad_norm": 0.8608767986297607,
      "learning_rate": 0.00019863056237479573,
      "loss": 0.8643695831298828,
      "step": 6060
    },
    {
      "epoch": 0.0813097976966689,
      "grad_norm": 0.8834564685821533,
      "learning_rate": 0.00019862339779865576,
      "loss": 0.8660571098327636,
      "step": 6070
    },
    {
      "epoch": 0.08144375123488418,
      "grad_norm": 0.8985983729362488,
      "learning_rate": 0.00019861621465958192,
      "loss": 0.9212307929992676,
      "step": 6080
    },
    {
      "epoch": 0.08157770477309945,
      "grad_norm": 0.9581639766693115,
      "learning_rate": 0.00019860901295892612,
      "loss": 0.8630023002624512,
      "step": 6090
    },
    {
      "epoch": 0.08171165831131472,
      "grad_norm": 1.0424174070358276,
      "learning_rate": 0.00019860179269804394,
      "loss": 0.9141819953918457,
      "step": 6100
    },
    {
      "epoch": 0.08184561184952999,
      "grad_norm": 0.9702780246734619,
      "learning_rate": 0.00019859455387829432,
      "loss": 0.8898551940917969,
      "step": 6110
    },
    {
      "epoch": 0.08197956538774526,
      "grad_norm": 0.7757335305213928,
      "learning_rate": 0.0001985872965010398,
      "loss": 0.9024291038513184,
      "step": 6120
    },
    {
      "epoch": 0.08211351892596053,
      "grad_norm": 0.8599688410758972,
      "learning_rate": 0.00019858002056764634,
      "loss": 0.9251861572265625,
      "step": 6130
    },
    {
      "epoch": 0.0822474724641758,
      "grad_norm": 0.888074517250061,
      "learning_rate": 0.00019857272607948343,
      "loss": 0.8888583183288574,
      "step": 6140
    },
    {
      "epoch": 0.08238142600239107,
      "grad_norm": 0.8135772943496704,
      "learning_rate": 0.00019856541303792403,
      "loss": 0.91273193359375,
      "step": 6150
    },
    {
      "epoch": 0.08251537954060634,
      "grad_norm": 0.8697153329849243,
      "learning_rate": 0.00019855808144434461,
      "loss": 0.8498779296875,
      "step": 6160
    },
    {
      "epoch": 0.08264933307882161,
      "grad_norm": 0.7868149876594543,
      "learning_rate": 0.0001985507313001252,
      "loss": 0.9167130470275879,
      "step": 6170
    },
    {
      "epoch": 0.08278328661703688,
      "grad_norm": 0.829988956451416,
      "learning_rate": 0.00019854336260664908,
      "loss": 0.9186518669128418,
      "step": 6180
    },
    {
      "epoch": 0.08291724015525215,
      "grad_norm": 0.8120090961456299,
      "learning_rate": 0.00019853597536530332,
      "loss": 0.8929192543029785,
      "step": 6190
    },
    {
      "epoch": 0.08305119369346742,
      "grad_norm": 0.8368801474571228,
      "learning_rate": 0.0001985285695774783,
      "loss": 0.9175758361816406,
      "step": 6200
    },
    {
      "epoch": 0.08318514723168269,
      "grad_norm": 0.9555438160896301,
      "learning_rate": 0.00019852114524456798,
      "loss": 0.8993669509887695,
      "step": 6210
    },
    {
      "epoch": 0.08331910076989796,
      "grad_norm": 0.879269540309906,
      "learning_rate": 0.00019851370236796974,
      "loss": 0.8785482406616211,
      "step": 6220
    },
    {
      "epoch": 0.08345305430811323,
      "grad_norm": 0.9049955010414124,
      "learning_rate": 0.00019850624094908446,
      "loss": 0.9055108070373535,
      "step": 6230
    },
    {
      "epoch": 0.0835870078463285,
      "grad_norm": 0.8032420873641968,
      "learning_rate": 0.00019849876098931656,
      "loss": 0.8821776390075684,
      "step": 6240
    },
    {
      "epoch": 0.08372096138454377,
      "grad_norm": 0.8842592239379883,
      "learning_rate": 0.00019849126249007391,
      "loss": 0.8976425170898438,
      "step": 6250
    },
    {
      "epoch": 0.08385491492275904,
      "grad_norm": 0.8668513298034668,
      "learning_rate": 0.0001984837454527679,
      "loss": 0.8919437408447266,
      "step": 6260
    },
    {
      "epoch": 0.08398886846097431,
      "grad_norm": 0.8331581950187683,
      "learning_rate": 0.00019847620987881339,
      "loss": 0.9026307106018067,
      "step": 6270
    },
    {
      "epoch": 0.08412282199918958,
      "grad_norm": 0.9931880831718445,
      "learning_rate": 0.0001984686557696287,
      "loss": 0.9175630569458008,
      "step": 6280
    },
    {
      "epoch": 0.08425677553740485,
      "grad_norm": 0.9612919688224792,
      "learning_rate": 0.00019846108312663568,
      "loss": 0.8934223175048828,
      "step": 6290
    },
    {
      "epoch": 0.08439072907562012,
      "grad_norm": 0.8972668051719666,
      "learning_rate": 0.0001984534919512597,
      "loss": 0.8968947410583497,
      "step": 6300
    },
    {
      "epoch": 0.08452468261383539,
      "grad_norm": 0.8106444478034973,
      "learning_rate": 0.00019844588224492946,
      "loss": 0.8991499900817871,
      "step": 6310
    },
    {
      "epoch": 0.08465863615205066,
      "grad_norm": 0.8595272898674011,
      "learning_rate": 0.00019843825400907736,
      "loss": 0.8847417831420898,
      "step": 6320
    },
    {
      "epoch": 0.08479258969026593,
      "grad_norm": 0.9921483993530273,
      "learning_rate": 0.00019843060724513916,
      "loss": 0.8878026962280273,
      "step": 6330
    },
    {
      "epoch": 0.0849265432284812,
      "grad_norm": 0.8371380567550659,
      "learning_rate": 0.0001984229419545542,
      "loss": 0.8880431175231933,
      "step": 6340
    },
    {
      "epoch": 0.08506049676669647,
      "grad_norm": 0.9585050940513611,
      "learning_rate": 0.00019841525813876513,
      "loss": 0.9015412330627441,
      "step": 6350
    },
    {
      "epoch": 0.08519445030491174,
      "grad_norm": 1.0291354656219482,
      "learning_rate": 0.00019840755579921825,
      "loss": 0.9376767158508301,
      "step": 6360
    },
    {
      "epoch": 0.08532840384312701,
      "grad_norm": 0.9010260701179504,
      "learning_rate": 0.00019839983493736334,
      "loss": 0.8865902900695801,
      "step": 6370
    },
    {
      "epoch": 0.08546235738134228,
      "grad_norm": 0.908366858959198,
      "learning_rate": 0.00019839209555465355,
      "loss": 0.8736093521118165,
      "step": 6380
    },
    {
      "epoch": 0.08559631091955755,
      "grad_norm": 0.9020763039588928,
      "learning_rate": 0.00019838433765254564,
      "loss": 0.9092628479003906,
      "step": 6390
    },
    {
      "epoch": 0.08573026445777282,
      "grad_norm": 0.8950169086456299,
      "learning_rate": 0.0001983765612324998,
      "loss": 0.8904929161071777,
      "step": 6400
    },
    {
      "epoch": 0.08586421799598809,
      "grad_norm": 0.9111661314964294,
      "learning_rate": 0.00019836876629597967,
      "loss": 0.9055502891540528,
      "step": 6410
    },
    {
      "epoch": 0.08599817153420336,
      "grad_norm": 0.8205604553222656,
      "learning_rate": 0.00019836095284445247,
      "loss": 0.8952221870422363,
      "step": 6420
    },
    {
      "epoch": 0.08613212507241863,
      "grad_norm": 0.8009998798370361,
      "learning_rate": 0.00019835312087938882,
      "loss": 0.8730879783630371,
      "step": 6430
    },
    {
      "epoch": 0.0862660786106339,
      "grad_norm": 0.9848208427429199,
      "learning_rate": 0.0001983452704022629,
      "loss": 0.918694019317627,
      "step": 6440
    },
    {
      "epoch": 0.08640003214884917,
      "grad_norm": 0.9056805968284607,
      "learning_rate": 0.00019833740141455227,
      "loss": 0.8988473892211915,
      "step": 6450
    },
    {
      "epoch": 0.08653398568706444,
      "grad_norm": 0.8487341403961182,
      "learning_rate": 0.00019832951391773803,
      "loss": 0.8815153121948243,
      "step": 6460
    },
    {
      "epoch": 0.08666793922527971,
      "grad_norm": 0.8682738542556763,
      "learning_rate": 0.00019832160791330482,
      "loss": 0.8735312461853028,
      "step": 6470
    },
    {
      "epoch": 0.08680189276349498,
      "grad_norm": 0.918135941028595,
      "learning_rate": 0.0001983136834027407,
      "loss": 0.8753129005432129,
      "step": 6480
    },
    {
      "epoch": 0.08693584630171025,
      "grad_norm": 0.9785839319229126,
      "learning_rate": 0.0001983057403875372,
      "loss": 0.9091508865356446,
      "step": 6490
    },
    {
      "epoch": 0.08706979983992552,
      "grad_norm": 0.8380211591720581,
      "learning_rate": 0.00019829777886918936,
      "loss": 0.9081586837768555,
      "step": 6500
    },
    {
      "epoch": 0.08720375337814079,
      "grad_norm": 0.9928073883056641,
      "learning_rate": 0.0001982897988491957,
      "loss": 0.886607551574707,
      "step": 6510
    },
    {
      "epoch": 0.08733770691635606,
      "grad_norm": 0.889708936214447,
      "learning_rate": 0.00019828180032905826,
      "loss": 0.903848934173584,
      "step": 6520
    },
    {
      "epoch": 0.08747166045457133,
      "grad_norm": 0.860169529914856,
      "learning_rate": 0.00019827378331028247,
      "loss": 0.8999492645263671,
      "step": 6530
    },
    {
      "epoch": 0.0876056139927866,
      "grad_norm": 0.866796612739563,
      "learning_rate": 0.00019826574779437738,
      "loss": 0.8896894454956055,
      "step": 6540
    },
    {
      "epoch": 0.08773956753100187,
      "grad_norm": 0.9426800608634949,
      "learning_rate": 0.0001982576937828553,
      "loss": 0.884849739074707,
      "step": 6550
    },
    {
      "epoch": 0.08787352106921714,
      "grad_norm": 0.8826320171356201,
      "learning_rate": 0.00019824962127723228,
      "loss": 0.8952905654907226,
      "step": 6560
    },
    {
      "epoch": 0.08800747460743241,
      "grad_norm": 0.8080610632896423,
      "learning_rate": 0.0001982415302790277,
      "loss": 0.8895891189575196,
      "step": 6570
    },
    {
      "epoch": 0.08814142814564768,
      "grad_norm": 0.9116784930229187,
      "learning_rate": 0.00019823342078976439,
      "loss": 0.8802170753479004,
      "step": 6580
    },
    {
      "epoch": 0.08827538168386295,
      "grad_norm": 0.8237122893333435,
      "learning_rate": 0.0001982252928109688,
      "loss": 0.9002214431762695,
      "step": 6590
    },
    {
      "epoch": 0.08840933522207822,
      "grad_norm": 0.9646167159080505,
      "learning_rate": 0.0001982171463441708,
      "loss": 0.9047221183776856,
      "step": 6600
    },
    {
      "epoch": 0.08854328876029349,
      "grad_norm": 1.0312793254852295,
      "learning_rate": 0.0001982089813909036,
      "loss": 0.9000032424926758,
      "step": 6610
    },
    {
      "epoch": 0.08867724229850876,
      "grad_norm": 0.8922534584999084,
      "learning_rate": 0.00019820079795270412,
      "loss": 0.9081113815307618,
      "step": 6620
    },
    {
      "epoch": 0.08881119583672403,
      "grad_norm": 0.8138467669487,
      "learning_rate": 0.00019819259603111262,
      "loss": 0.8970637321472168,
      "step": 6630
    },
    {
      "epoch": 0.0889451493749393,
      "grad_norm": 0.911562442779541,
      "learning_rate": 0.00019818437562767286,
      "loss": 0.8693688392639161,
      "step": 6640
    },
    {
      "epoch": 0.08907910291315457,
      "grad_norm": 0.8164427280426025,
      "learning_rate": 0.00019817613674393206,
      "loss": 0.8692523956298828,
      "step": 6650
    },
    {
      "epoch": 0.08921305645136984,
      "grad_norm": 0.8683072328567505,
      "learning_rate": 0.000198167879381441,
      "loss": 0.872105598449707,
      "step": 6660
    },
    {
      "epoch": 0.08934700998958511,
      "grad_norm": 0.882523775100708,
      "learning_rate": 0.00019815960354175384,
      "loss": 0.9052138328552246,
      "step": 6670
    },
    {
      "epoch": 0.08948096352780038,
      "grad_norm": 0.8588592410087585,
      "learning_rate": 0.0001981513092264283,
      "loss": 0.8927006721496582,
      "step": 6680
    },
    {
      "epoch": 0.08961491706601565,
      "grad_norm": 0.9427072405815125,
      "learning_rate": 0.00019814299643702553,
      "loss": 0.8979676246643067,
      "step": 6690
    },
    {
      "epoch": 0.08974887060423092,
      "grad_norm": 0.9437255859375,
      "learning_rate": 0.00019813466517511015,
      "loss": 0.8920472145080567,
      "step": 6700
    },
    {
      "epoch": 0.0898828241424462,
      "grad_norm": 1.0207903385162354,
      "learning_rate": 0.00019812631544225028,
      "loss": 0.9171392440795898,
      "step": 6710
    },
    {
      "epoch": 0.09001677768066146,
      "grad_norm": 0.9653688669204712,
      "learning_rate": 0.0001981179472400175,
      "loss": 0.8855166435241699,
      "step": 6720
    },
    {
      "epoch": 0.09015073121887673,
      "grad_norm": 0.8295390009880066,
      "learning_rate": 0.00019810956056998694,
      "loss": 0.885284423828125,
      "step": 6730
    },
    {
      "epoch": 0.090284684757092,
      "grad_norm": 0.9237222671508789,
      "learning_rate": 0.00019810115543373704,
      "loss": 0.8777644157409668,
      "step": 6740
    },
    {
      "epoch": 0.09041863829530727,
      "grad_norm": 0.869898796081543,
      "learning_rate": 0.00019809273183284985,
      "loss": 0.90859375,
      "step": 6750
    },
    {
      "epoch": 0.09055259183352254,
      "grad_norm": 0.8586453199386597,
      "learning_rate": 0.00019808428976891094,
      "loss": 0.8833423614501953,
      "step": 6760
    },
    {
      "epoch": 0.09068654537173781,
      "grad_norm": 0.8746774196624756,
      "learning_rate": 0.00019807582924350915,
      "loss": 0.8715258598327636,
      "step": 6770
    },
    {
      "epoch": 0.09082049890995308,
      "grad_norm": 0.9243775010108948,
      "learning_rate": 0.00019806735025823703,
      "loss": 0.903589916229248,
      "step": 6780
    },
    {
      "epoch": 0.09095445244816835,
      "grad_norm": 0.9908545613288879,
      "learning_rate": 0.00019805885281469048,
      "loss": 0.8912477493286133,
      "step": 6790
    },
    {
      "epoch": 0.09108840598638362,
      "grad_norm": 0.9097034931182861,
      "learning_rate": 0.00019805033691446883,
      "loss": 0.8512225151062012,
      "step": 6800
    },
    {
      "epoch": 0.0912223595245989,
      "grad_norm": 0.8254439830780029,
      "learning_rate": 0.00019804180255917502,
      "loss": 0.8855799674987793,
      "step": 6810
    },
    {
      "epoch": 0.09135631306281417,
      "grad_norm": 0.8863989114761353,
      "learning_rate": 0.00019803324975041538,
      "loss": 0.8820269584655762,
      "step": 6820
    },
    {
      "epoch": 0.09149026660102944,
      "grad_norm": 0.9456990361213684,
      "learning_rate": 0.00019802467848979968,
      "loss": 0.9022522926330566,
      "step": 6830
    },
    {
      "epoch": 0.0916242201392447,
      "grad_norm": 0.9565940499305725,
      "learning_rate": 0.0001980160887789412,
      "loss": 0.8896581649780273,
      "step": 6840
    },
    {
      "epoch": 0.09175817367745998,
      "grad_norm": 0.9246744513511658,
      "learning_rate": 0.00019800748061945674,
      "loss": 0.8820788383483886,
      "step": 6850
    },
    {
      "epoch": 0.09189212721567525,
      "grad_norm": 0.9936888813972473,
      "learning_rate": 0.00019799885401296653,
      "loss": 0.8832516670227051,
      "step": 6860
    },
    {
      "epoch": 0.09202608075389052,
      "grad_norm": 0.8579689264297485,
      "learning_rate": 0.00019799020896109424,
      "loss": 0.8947846412658691,
      "step": 6870
    },
    {
      "epoch": 0.09216003429210579,
      "grad_norm": 0.8918428421020508,
      "learning_rate": 0.00019798154546546708,
      "loss": 0.901578712463379,
      "step": 6880
    },
    {
      "epoch": 0.09229398783032106,
      "grad_norm": 1.207512378692627,
      "learning_rate": 0.00019797286352771568,
      "loss": 0.9050963401794434,
      "step": 6890
    },
    {
      "epoch": 0.09242794136853633,
      "grad_norm": 0.8686789274215698,
      "learning_rate": 0.00019796416314947415,
      "loss": 0.9166510581970215,
      "step": 6900
    },
    {
      "epoch": 0.0925618949067516,
      "grad_norm": 0.8333821296691895,
      "learning_rate": 0.0001979554443323801,
      "loss": 0.8954951286315918,
      "step": 6910
    },
    {
      "epoch": 0.09269584844496687,
      "grad_norm": 0.9899786710739136,
      "learning_rate": 0.00019794670707807456,
      "loss": 0.9007176399230957,
      "step": 6920
    },
    {
      "epoch": 0.09282980198318214,
      "grad_norm": 0.8100647330284119,
      "learning_rate": 0.00019793795138820208,
      "loss": 0.9082587242126465,
      "step": 6930
    },
    {
      "epoch": 0.0929637555213974,
      "grad_norm": 0.9281670451164246,
      "learning_rate": 0.0001979291772644107,
      "loss": 0.8884916305541992,
      "step": 6940
    },
    {
      "epoch": 0.09309770905961268,
      "grad_norm": 0.8830415606498718,
      "learning_rate": 0.0001979203847083518,
      "loss": 0.8823875427246094,
      "step": 6950
    },
    {
      "epoch": 0.09323166259782795,
      "grad_norm": 0.8673772811889648,
      "learning_rate": 0.00019791157372168038,
      "loss": 0.87948579788208,
      "step": 6960
    },
    {
      "epoch": 0.09336561613604322,
      "grad_norm": 0.922353982925415,
      "learning_rate": 0.00019790274430605482,
      "loss": 0.8732827186584473,
      "step": 6970
    },
    {
      "epoch": 0.09349956967425849,
      "grad_norm": 0.9420790672302246,
      "learning_rate": 0.000197893896463137,
      "loss": 0.8895031929016113,
      "step": 6980
    },
    {
      "epoch": 0.09363352321247376,
      "grad_norm": 0.8268752694129944,
      "learning_rate": 0.0001978850301945923,
      "loss": 0.8912209510803223,
      "step": 6990
    },
    {
      "epoch": 0.09376747675068903,
      "grad_norm": 0.8777807354927063,
      "learning_rate": 0.00019787614550208945,
      "loss": 0.8920040130615234,
      "step": 7000
    },
    {
      "epoch": 0.09376747675068903,
      "eval_loss": 0.8644543886184692,
      "eval_runtime": 282.9926,
      "eval_samples_per_second": 10.647,
      "eval_steps_per_second": 10.647,
      "step": 7000
    }
  ],
  "logging_steps": 10,
  "max_steps": 74653,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10000000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.386491436203103e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}