{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 75.0,
  "eval_steps": 500,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.25,
      "grad_norm": 9.3125,
      "learning_rate": 1.8e-05,
      "loss": 2.419099235534668,
      "loss_d0": 2.4237823486328125,
      "step": 10
    },
    {
      "epoch": 0.5,
      "grad_norm": 4.46875,
      "learning_rate": 3.8e-05,
      "loss": 1.016735553741455,
      "loss_d0": 1.0169126868247986,
      "step": 20
    },
    {
      "epoch": 0.75,
      "grad_norm": 3.03125,
      "learning_rate": 5.8e-05,
      "loss": 0.7163782119750977,
      "loss_d0": 0.7170086741447449,
      "step": 30
    },
    {
      "epoch": 1.0,
      "grad_norm": 3.015625,
      "learning_rate": 7.800000000000001e-05,
      "loss": 0.5825047969818116,
      "loss_d0": 0.5834829866886139,
      "step": 40
    },
    {
      "epoch": 1.25,
      "grad_norm": 2.640625,
      "learning_rate": 9.8e-05,
      "loss": 0.5070408821105957,
      "loss_d0": 0.5083645194768905,
      "step": 50
    },
    {
      "epoch": 1.5,
      "grad_norm": 2.625,
      "learning_rate": 0.000118,
      "loss": 0.5112638473510742,
      "loss_d0": 0.5115580916404724,
      "step": 60
    },
    {
      "epoch": 1.75,
      "grad_norm": 2.25,
      "learning_rate": 0.000138,
      "loss": 0.4753141403198242,
      "loss_d0": 0.4750445276498795,
      "step": 70
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.453125,
      "learning_rate": 0.00015800000000000002,
      "loss": 0.44187202453613283,
      "loss_d0": 0.4409120589494705,
      "step": 80
    },
    {
      "epoch": 2.25,
      "grad_norm": 2.109375,
      "learning_rate": 0.00017800000000000002,
      "loss": 0.3495074510574341,
      "loss_d0": 0.3506994903087616,
      "step": 90
    },
    {
      "epoch": 2.5,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019800000000000002,
      "loss": 0.355522608757019,
      "loss_d0": 0.3573369801044464,
      "step": 100
    },
    {
      "epoch": 2.75,
      "grad_norm": 2.40625,
      "learning_rate": 0.0002,
      "loss": 0.36629390716552734,
      "loss_d0": 0.36523382663726806,
      "step": 110
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.3125,
      "learning_rate": 0.0002,
      "loss": 0.32001023292541503,
      "loss_d0": 0.3198924452066422,
      "step": 120
    },
    {
      "epoch": 3.25,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002,
      "loss": 0.2629648923873901,
      "loss_d0": 0.26279704421758654,
      "step": 130
    },
    {
      "epoch": 3.5,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002,
      "loss": 0.2764778137207031,
      "loss_d0": 0.27699622362852094,
      "step": 140
    },
    {
      "epoch": 3.75,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002,
      "loss": 0.2808023691177368,
      "loss_d0": 0.27955625057220457,
      "step": 150
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002,
      "loss": 0.2766923666000366,
      "loss_d0": 0.2778839275240898,
      "step": 160
    },
    {
      "epoch": 4.25,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0002,
      "loss": 0.2177126407623291,
      "loss_d0": 0.21817942708730698,
      "step": 170
    },
    {
      "epoch": 4.5,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002,
      "loss": 0.22354187965393066,
      "loss_d0": 0.2253672033548355,
      "step": 180
    },
    {
      "epoch": 4.75,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002,
      "loss": 0.2261255979537964,
      "loss_d0": 0.22708745449781417,
      "step": 190
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002,
      "loss": 0.23856868743896484,
      "loss_d0": 0.2382099911570549,
      "step": 200
    },
    {
      "epoch": 5.25,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002,
      "loss": 0.1836856484413147,
      "loss_d0": 0.18439767211675645,
      "step": 210
    },
    {
      "epoch": 5.5,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002,
      "loss": 0.20460774898529052,
      "loss_d0": 0.2053621456027031,
      "step": 220
    },
    {
      "epoch": 5.75,
      "grad_norm": 1.125,
      "learning_rate": 0.0002,
      "loss": 0.20773332118988036,
      "loss_d0": 0.2081416815519333,
      "step": 230
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.375,
      "learning_rate": 0.0002,
      "loss": 0.20451078414916993,
      "loss_d0": 0.20580717772245408,
      "step": 240
    },
    {
      "epoch": 6.25,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002,
      "loss": 0.16367343664169312,
      "loss_d0": 0.1637965440750122,
      "step": 250
    },
    {
      "epoch": 6.5,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002,
      "loss": 0.1768767476081848,
      "loss_d0": 0.17705970853567124,
      "step": 260
    },
    {
      "epoch": 6.75,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002,
      "loss": 0.19422578811645508,
      "loss_d0": 0.1945399060845375,
      "step": 270
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.84375,
      "learning_rate": 0.0002,
      "loss": 0.19753103256225585,
      "loss_d0": 0.1971623793244362,
      "step": 280
    },
    {
      "epoch": 7.25,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0002,
      "loss": 0.14055638313293456,
      "loss_d0": 0.1412511058151722,
      "step": 290
    },
    {
      "epoch": 7.5,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0002,
      "loss": 0.15400553941726686,
      "loss_d0": 0.15403806418180466,
      "step": 300
    },
    {
      "epoch": 7.75,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0002,
      "loss": 0.1660689115524292,
      "loss_d0": 0.16630121916532517,
      "step": 310
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002,
      "loss": 0.1730404496192932,
      "loss_d0": 0.17368634641170502,
      "step": 320
    },
    {
      "epoch": 8.25,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0002,
      "loss": 0.1311247706413269,
      "loss_d0": 0.13191080316901208,
      "step": 330
    },
    {
      "epoch": 8.5,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0002,
      "loss": 0.13660377264022827,
      "loss_d0": 0.13707814514636993,
      "step": 340
    },
    {
      "epoch": 8.75,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0002,
      "loss": 0.16057606935501098,
      "loss_d0": 0.16095266342163086,
      "step": 350
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0002,
      "loss": 0.15158145427703856,
      "loss_d0": 0.1519700437784195,
      "step": 360
    },
    {
      "epoch": 9.25,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0002,
      "loss": 0.12598543167114257,
      "loss_d0": 0.12593471333384515,
      "step": 370
    },
    {
      "epoch": 9.5,
      "grad_norm": 0.8125,
      "learning_rate": 0.0002,
      "loss": 0.12625789642333984,
      "loss_d0": 0.12605250850319863,
      "step": 380
    },
    {
      "epoch": 9.75,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002,
      "loss": 0.15585073232650756,
      "loss_d0": 0.15560536012053489,
      "step": 390
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0002,
      "loss": 0.15288279056549073,
      "loss_d0": 0.15293605029582977,
      "step": 400
    },
    {
      "epoch": 10.25,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0002,
      "loss": 0.1126750111579895,
      "loss_d0": 0.11266628131270409,
      "step": 410
    },
    {
      "epoch": 10.5,
      "grad_norm": 11.75,
      "learning_rate": 0.0002,
      "loss": 0.1392864465713501,
      "loss_d0": 0.13897545784711837,
      "step": 420
    },
    {
      "epoch": 10.75,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0002,
      "loss": 0.1241998553276062,
      "loss_d0": 0.12495879083871841,
      "step": 430
    },
    {
      "epoch": 11.0,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0002,
      "loss": 0.13399370908737182,
      "loss_d0": 0.1345785766839981,
      "step": 440
    },
    {
      "epoch": 11.25,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002,
      "loss": 0.11483274698257447,
      "loss_d0": 0.11512825042009353,
      "step": 450
    },
    {
      "epoch": 11.5,
      "grad_norm": 0.65234375,
      "learning_rate": 0.0002,
      "loss": 0.115890634059906,
      "loss_d0": 0.11643885746598244,
      "step": 460
    },
    {
      "epoch": 11.75,
      "grad_norm": 0.703125,
      "learning_rate": 0.0002,
      "loss": 0.12420456409454346,
      "loss_d0": 0.12463393807411194,
      "step": 470
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0002,
      "loss": 0.12463445663452148,
      "loss_d0": 0.12508776113390924,
      "step": 480
    },
    {
      "epoch": 12.25,
      "grad_norm": 0.66015625,
      "learning_rate": 0.0002,
      "loss": 0.08669283390045165,
      "loss_d0": 0.08694541230797767,
      "step": 490
    },
    {
      "epoch": 12.5,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0002,
      "loss": 0.10650432109832764,
      "loss_d0": 0.10702361688017845,
      "step": 500
    },
    {
      "epoch": 12.5,
      "eval_loss": 7.686856746673584,
      "eval_runtime": 0.346,
      "eval_samples_per_second": 722.458,
      "eval_steps_per_second": 72.246,
      "step": 500
    },
    {
      "epoch": 12.75,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0002,
      "loss": 0.11304057836532592,
      "loss_d0": 0.11300384327769279,
      "step": 510
    },
    {
      "epoch": 13.0,
      "grad_norm": 0.765625,
      "learning_rate": 0.0002,
      "loss": 0.12059590816497803,
      "loss_d0": 0.12071148529648781,
      "step": 520
    },
    {
      "epoch": 13.25,
      "grad_norm": 0.53125,
      "learning_rate": 0.0002,
      "loss": 0.08754821419715882,
      "loss_d0": 0.08753191828727722,
      "step": 530
    },
    {
      "epoch": 13.5,
      "grad_norm": 0.69140625,
      "learning_rate": 0.0002,
      "loss": 0.09736397266387939,
      "loss_d0": 0.09748184978961945,
      "step": 540
    },
    {
      "epoch": 13.75,
      "grad_norm": 0.703125,
      "learning_rate": 0.0002,
      "loss": 0.10162832736968994,
      "loss_d0": 0.10216130167245865,
      "step": 550
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0002,
      "loss": 0.11431772708892822,
      "loss_d0": 0.11476395502686501,
      "step": 560
    },
    {
      "epoch": 14.25,
      "grad_norm": 0.66015625,
      "learning_rate": 0.0002,
      "loss": 0.09515688419342042,
      "loss_d0": 0.09562777429819107,
      "step": 570
    },
    {
      "epoch": 14.5,
      "grad_norm": 0.671875,
      "learning_rate": 0.0002,
      "loss": 0.10088273286819457,
      "loss_d0": 0.10082438737154006,
      "step": 580
    },
    {
      "epoch": 14.75,
      "grad_norm": 0.71484375,
      "learning_rate": 0.0002,
      "loss": 0.10143396854400635,
      "loss_d0": 0.10186150297522545,
      "step": 590
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.66796875,
      "learning_rate": 0.0002,
      "loss": 0.11544075012207031,
      "loss_d0": 0.11603940427303314,
      "step": 600
    },
    {
      "epoch": 15.25,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0002,
      "loss": 0.08207579851150512,
      "loss_d0": 0.08233651667833328,
      "step": 610
    },
    {
      "epoch": 15.5,
      "grad_norm": 0.671875,
      "learning_rate": 0.0002,
      "loss": 0.09775862693786622,
      "loss_d0": 0.09806842431426048,
      "step": 620
    },
    {
      "epoch": 15.75,
      "grad_norm": 0.703125,
      "learning_rate": 0.0002,
      "loss": 0.0941968321800232,
      "loss_d0": 0.09450600519776345,
      "step": 630
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0002,
      "loss": 0.11077705621719361,
      "loss_d0": 0.111270122975111,
      "step": 640
    },
    {
      "epoch": 16.25,
      "grad_norm": 0.65234375,
      "learning_rate": 0.0002,
      "loss": 0.08216186165809632,
      "loss_d0": 0.08217487558722496,
      "step": 650
    },
    {
      "epoch": 16.5,
      "grad_norm": 0.578125,
      "learning_rate": 0.0002,
      "loss": 0.08583685159683227,
      "loss_d0": 0.08590926378965377,
      "step": 660
    },
    {
      "epoch": 16.75,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0002,
      "loss": 0.10179605484008789,
      "loss_d0": 0.10241870805621148,
      "step": 670
    },
    {
      "epoch": 17.0,
      "grad_norm": 0.75390625,
      "learning_rate": 0.0002,
      "loss": 0.11323529481887817,
      "loss_d0": 0.11270172968506813,
      "step": 680
    },
    {
      "epoch": 17.25,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0002,
      "loss": 0.07983130812644959,
      "loss_d0": 0.07931971848011017,
      "step": 690
    },
    {
      "epoch": 17.5,
      "grad_norm": 0.64453125,
      "learning_rate": 0.0002,
      "loss": 0.08101434707641601,
      "loss_d0": 0.0811703234910965,
      "step": 700
    },
    {
      "epoch": 17.75,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0002,
      "loss": 0.08358389735221863,
      "loss_d0": 0.08354934751987457,
      "step": 710
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.60546875,
      "learning_rate": 0.0002,
      "loss": 0.09092987775802612,
      "loss_d0": 0.09135034829378127,
      "step": 720
    },
    {
      "epoch": 18.25,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0002,
      "loss": 0.0751054584980011,
      "loss_d0": 0.07558580189943313,
      "step": 730
    },
    {
      "epoch": 18.5,
      "grad_norm": 0.625,
      "learning_rate": 0.0002,
      "loss": 0.08020662069320679,
      "loss_d0": 0.08090231120586396,
      "step": 740
    },
    {
      "epoch": 18.75,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0002,
      "loss": 0.08230118155479431,
      "loss_d0": 0.08202177286148071,
      "step": 750
    },
    {
      "epoch": 19.0,
      "grad_norm": 0.65625,
      "learning_rate": 0.0002,
      "loss": 0.0855340838432312,
      "loss_d0": 0.0853975273668766,
      "step": 760
    },
    {
      "epoch": 19.25,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0002,
      "loss": 0.07504370212554931,
      "loss_d0": 0.0755705066025257,
      "step": 770
    },
    {
      "epoch": 19.5,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0002,
      "loss": 0.07457044124603271,
      "loss_d0": 0.07485715299844742,
      "step": 780
    },
    {
      "epoch": 19.75,
      "grad_norm": 0.58203125,
      "learning_rate": 0.0002,
      "loss": 0.08841861486434936,
      "loss_d0": 0.08899671211838722,
      "step": 790
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0002,
      "loss": 0.08840059638023376,
      "loss_d0": 0.08874272927641869,
      "step": 800
    },
    {
      "epoch": 20.25,
      "grad_norm": 0.470703125,
      "learning_rate": 0.0002,
      "loss": 0.07279248833656311,
      "loss_d0": 0.0725507315248251,
      "step": 810
    },
    {
      "epoch": 20.5,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0002,
      "loss": 0.07490698099136353,
      "loss_d0": 0.07517094723880291,
      "step": 820
    },
    {
      "epoch": 20.75,
      "grad_norm": 0.466796875,
      "learning_rate": 0.0002,
      "loss": 0.07984906435012817,
      "loss_d0": 0.08029644414782525,
      "step": 830
    },
    {
      "epoch": 21.0,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0002,
      "loss": 0.08839969038963318,
      "loss_d0": 0.08827509433031082,
      "step": 840
    },
    {
      "epoch": 21.25,
      "grad_norm": 0.478515625,
      "learning_rate": 0.0002,
      "loss": 0.0686249315738678,
      "loss_d0": 0.06923733800649642,
      "step": 850
    },
    {
      "epoch": 21.5,
      "grad_norm": 0.466796875,
      "learning_rate": 0.0002,
      "loss": 0.08188758492469787,
      "loss_d0": 0.08173562847077846,
      "step": 860
    },
    {
      "epoch": 21.75,
      "grad_norm": 0.443359375,
      "learning_rate": 0.0002,
      "loss": 0.0769996166229248,
      "loss_d0": 0.07713011205196381,
      "step": 870
    },
    {
      "epoch": 22.0,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0002,
      "loss": 0.07695686221122741,
      "loss_d0": 0.07774162963032723,
      "step": 880
    },
    {
      "epoch": 22.25,
      "grad_norm": 0.478515625,
      "learning_rate": 0.0002,
      "loss": 0.07001073360443115,
      "loss_d0": 0.0701629150658846,
      "step": 890
    },
    {
      "epoch": 22.5,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0002,
      "loss": 0.07286003232002258,
      "loss_d0": 0.07322507128119468,
      "step": 900
    },
    {
      "epoch": 22.75,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0002,
      "loss": 0.07836406230926514,
      "loss_d0": 0.07865607067942619,
      "step": 910
    },
    {
      "epoch": 23.0,
      "grad_norm": 0.70703125,
      "learning_rate": 0.0002,
      "loss": 0.08143852353096008,
      "loss_d0": 0.08168832883238793,
      "step": 920
    },
    {
      "epoch": 23.25,
      "grad_norm": 0.59375,
      "learning_rate": 0.0002,
      "loss": 0.07572694420814514,
      "loss_d0": 0.07592412531375885,
      "step": 930
    },
    {
      "epoch": 23.5,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0002,
      "loss": 0.07522768378257752,
      "loss_d0": 0.07555339634418487,
      "step": 940
    },
    {
      "epoch": 23.75,
      "grad_norm": 0.5703125,
      "learning_rate": 0.0002,
      "loss": 0.07778022289276124,
      "loss_d0": 0.0778064027428627,
      "step": 950
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.65625,
      "learning_rate": 0.0002,
      "loss": 0.07971324920654296,
      "loss_d0": 0.08017545640468597,
      "step": 960
    },
    {
      "epoch": 24.25,
      "grad_norm": 0.7109375,
      "learning_rate": 0.0002,
      "loss": 0.06289821863174438,
      "loss_d0": 0.06302815079689025,
      "step": 970
    },
    {
      "epoch": 24.5,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0002,
      "loss": 0.0684018850326538,
      "loss_d0": 0.06837239749729633,
      "step": 980
    },
    {
      "epoch": 24.75,
      "grad_norm": 0.59375,
      "learning_rate": 0.0002,
      "loss": 0.0713754415512085,
      "loss_d0": 0.07187536023557187,
      "step": 990
    },
    {
      "epoch": 25.0,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0002,
      "loss": 0.07395302653312683,
      "loss_d0": 0.07426370158791543,
      "step": 1000
    },
    {
      "epoch": 25.0,
      "eval_loss": 9.472062110900879,
      "eval_runtime": 0.3448,
      "eval_samples_per_second": 725.044,
      "eval_steps_per_second": 72.504,
      "step": 1000
    },
    {
      "epoch": 25.25,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0002,
      "loss": 0.06430479884147644,
      "loss_d0": 0.06464392952620983,
      "step": 1010
    },
    {
      "epoch": 25.5,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0002,
      "loss": 0.0726861834526062,
      "loss_d0": 0.07402807176113128,
      "step": 1020
    },
    {
      "epoch": 25.75,
      "grad_norm": 0.609375,
      "learning_rate": 0.0002,
      "loss": 0.0761205494403839,
      "loss_d0": 0.07630706876516342,
      "step": 1030
    },
    {
      "epoch": 26.0,
      "grad_norm": 0.546875,
      "learning_rate": 0.0002,
      "loss": 0.07697510123252868,
      "loss_d0": 0.07700100690126419,
      "step": 1040
    },
    {
      "epoch": 26.25,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0002,
      "loss": 0.06419612169265747,
      "loss_d0": 0.06443683132529258,
      "step": 1050
    },
    {
      "epoch": 26.5,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0002,
      "loss": 0.0656801700592041,
      "loss_d0": 0.06635799780488014,
      "step": 1060
    },
    {
      "epoch": 26.75,
      "grad_norm": 0.546875,
      "learning_rate": 0.0002,
      "loss": 0.07432531118392945,
      "loss_d0": 0.07482590526342392,
      "step": 1070
    },
    {
      "epoch": 27.0,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0002,
      "loss": 0.07460427880287171,
      "loss_d0": 0.07486905790865421,
      "step": 1080
    },
    {
      "epoch": 27.25,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0002,
      "loss": 0.062262147665023804,
      "loss_d0": 0.06270815506577491,
      "step": 1090
    },
    {
      "epoch": 27.5,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0002,
      "loss": 0.07988451719284058,
      "loss_d0": 0.07888436727225781,
      "step": 1100
    },
    {
      "epoch": 27.75,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0002,
      "loss": 0.06685459613800049,
      "loss_d0": 0.06692123860120773,
      "step": 1110
    },
    {
      "epoch": 28.0,
      "grad_norm": 0.7265625,
      "learning_rate": 0.0002,
      "loss": 0.08263227939605713,
      "loss_d0": 0.08312211632728576,
      "step": 1120
    },
    {
      "epoch": 28.25,
      "grad_norm": 0.40234375,
      "learning_rate": 0.0002,
      "loss": 0.05786241292953491,
      "loss_d0": 0.05830040462315082,
      "step": 1130
    },
    {
      "epoch": 28.5,
      "grad_norm": 0.416015625,
      "learning_rate": 0.0002,
      "loss": 0.06246680021286011,
      "loss_d0": 0.06251397952437401,
      "step": 1140
    },
    {
      "epoch": 28.75,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0002,
      "loss": 0.06916601657867431,
      "loss_d0": 0.0696951575577259,
      "step": 1150
    },
    {
      "epoch": 29.0,
      "grad_norm": 0.5703125,
      "learning_rate": 0.0002,
      "loss": 0.07527051568031311,
      "loss_d0": 0.07468223795294762,
      "step": 1160
    },
    {
      "epoch": 29.25,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0002,
      "loss": 0.0581015408039093,
      "loss_d0": 0.05824718400835991,
      "step": 1170
    },
    {
      "epoch": 29.5,
      "grad_norm": 0.466796875,
      "learning_rate": 0.0002,
      "loss": 0.060128825902938846,
      "loss_d0": 0.06012755185365677,
      "step": 1180
    },
    {
      "epoch": 29.75,
      "grad_norm": 0.5,
      "learning_rate": 0.0002,
      "loss": 0.0673996388912201,
      "loss_d0": 0.06763093918561935,
      "step": 1190
    },
    {
      "epoch": 30.0,
      "grad_norm": 0.40625,
      "learning_rate": 0.0002,
      "loss": 0.07309567928314209,
      "loss_d0": 0.07359372191131115,
      "step": 1200
    },
    {
      "epoch": 30.25,
      "grad_norm": 0.38671875,
      "learning_rate": 0.0002,
      "loss": 0.05833629965782165,
      "loss_d0": 0.05870196744799614,
      "step": 1210
    },
    {
      "epoch": 30.5,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0002,
      "loss": 0.06625040769577026,
      "loss_d0": 0.06676065064966678,
      "step": 1220
    },
    {
      "epoch": 30.75,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0002,
      "loss": 0.06969956159591675,
      "loss_d0": 0.07000019364058971,
      "step": 1230
    },
    {
      "epoch": 31.0,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0002,
      "loss": 0.06795068383216858,
      "loss_d0": 0.06837498843669891,
      "step": 1240
    },
    {
      "epoch": 31.25,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0002,
      "loss": 0.058440542221069335,
      "loss_d0": 0.05858158580958843,
      "step": 1250
    },
    {
      "epoch": 31.5,
      "grad_norm": 0.4140625,
      "learning_rate": 0.0002,
      "loss": 0.06397214531898499,
      "loss_d0": 0.06426840759813786,
      "step": 1260
    },
    {
      "epoch": 31.75,
      "grad_norm": 0.423828125,
      "learning_rate": 0.0002,
      "loss": 0.06345818042755128,
      "loss_d0": 0.06368137337267399,
      "step": 1270
    },
    {
      "epoch": 32.0,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0002,
      "loss": 0.06745975017547608,
      "loss_d0": 0.0678976308554411,
      "step": 1280
    },
    {
      "epoch": 32.25,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0002,
      "loss": 0.059911799430847165,
      "loss_d0": 0.060089119151234625,
      "step": 1290
    },
    {
      "epoch": 32.5,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0002,
      "loss": 0.06502289175987244,
      "loss_d0": 0.06520598568022251,
      "step": 1300
    },
    {
      "epoch": 32.75,
      "grad_norm": 0.546875,
      "learning_rate": 0.0002,
      "loss": 0.06437732577323914,
      "loss_d0": 0.06469985544681549,
      "step": 1310
    },
    {
      "epoch": 33.0,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0002,
      "loss": 0.06933318376541138,
      "loss_d0": 0.06987486742436885,
      "step": 1320
    },
    {
      "epoch": 33.25,
      "grad_norm": 0.5,
      "learning_rate": 0.0002,
      "loss": 0.06083628535270691,
      "loss_d0": 0.06119122616946697,
      "step": 1330
    },
    {
      "epoch": 33.5,
      "grad_norm": 0.546875,
      "learning_rate": 0.0002,
      "loss": 0.06731333136558533,
      "loss_d0": 0.06779481768608094,
      "step": 1340
    },
    {
      "epoch": 33.75,
      "grad_norm": 0.4453125,
      "learning_rate": 0.0002,
      "loss": 0.06366108655929566,
      "loss_d0": 0.06400349549949169,
      "step": 1350
    },
    {
      "epoch": 34.0,
      "grad_norm": 0.53125,
      "learning_rate": 0.0002,
      "loss": 0.07252717018127441,
      "loss_d0": 0.07273228801786899,
      "step": 1360
    },
    {
      "epoch": 34.25,
      "grad_norm": 0.44921875,
      "learning_rate": 0.0002,
      "loss": 0.057601267099380495,
      "loss_d0": 0.057797104492783545,
      "step": 1370
    },
    {
      "epoch": 34.5,
      "grad_norm": 0.302734375,
      "learning_rate": 0.0002,
      "loss": 0.06590864658355713,
      "loss_d0": 0.06650126352906227,
      "step": 1380
    },
    {
      "epoch": 34.75,
      "grad_norm": 0.458984375,
      "learning_rate": 0.0002,
      "loss": 0.06279119253158569,
      "loss_d0": 0.06243448853492737,
      "step": 1390
    },
    {
      "epoch": 35.0,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0002,
      "loss": 0.06743443608283997,
      "loss_d0": 0.06724736131727696,
      "step": 1400
    },
    {
      "epoch": 35.25,
      "grad_norm": 0.5625,
      "learning_rate": 0.0002,
      "loss": 0.06006784439086914,
      "loss_d0": 0.06038782335817814,
      "step": 1410
    },
    {
      "epoch": 35.5,
      "grad_norm": 0.373046875,
      "learning_rate": 0.0002,
      "loss": 0.06705097556114196,
      "loss_d0": 0.06722290031611919,
      "step": 1420
    },
    {
      "epoch": 35.75,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0002,
      "loss": 0.0643904447555542,
      "loss_d0": 0.0645883984863758,
      "step": 1430
    },
    {
      "epoch": 36.0,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0002,
      "loss": 0.06919337511062622,
      "loss_d0": 0.06933849938213825,
      "step": 1440
    },
    {
      "epoch": 36.25,
      "grad_norm": 0.353515625,
      "learning_rate": 0.0002,
      "loss": 0.05321390628814697,
      "loss_d0": 0.0534587811678648,
      "step": 1450
    },
    {
      "epoch": 36.5,
      "grad_norm": 0.41796875,
      "learning_rate": 0.0002,
      "loss": 0.05641515254974365,
      "loss_d0": 0.05651510432362557,
      "step": 1460
    },
    {
      "epoch": 36.75,
      "grad_norm": 0.54296875,
      "learning_rate": 0.0002,
      "loss": 0.056172966957092285,
      "loss_d0": 0.05633020140230656,
      "step": 1470
    },
    {
      "epoch": 37.0,
      "grad_norm": 0.498046875,
      "learning_rate": 0.0002,
      "loss": 0.071669602394104,
      "loss_d0": 0.07104928344488144,
      "step": 1480
    },
    {
      "epoch": 37.25,
      "grad_norm": 0.337890625,
      "learning_rate": 0.0002,
      "loss": 0.057771342992782596,
      "loss_d0": 0.05772700011730194,
      "step": 1490
    },
    {
      "epoch": 37.5,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0002,
      "loss": 0.05799049735069275,
      "loss_d0": 0.058008070290088656,
      "step": 1500
    },
    {
      "epoch": 37.5,
      "eval_loss": 10.0471773147583,
      "eval_runtime": 0.3455,
      "eval_samples_per_second": 723.591,
      "eval_steps_per_second": 72.359,
      "step": 1500
    },
    {
      "epoch": 37.75,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0002,
      "loss": 0.060436797142028806,
      "loss_d0": 0.06095544844865799,
      "step": 1510
    },
    {
      "epoch": 38.0,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0002,
      "loss": 0.06353324055671691,
      "loss_d0": 0.06384937241673469,
      "step": 1520
    },
    {
      "epoch": 38.25,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0002,
      "loss": 0.05317062735557556,
      "loss_d0": 0.0535169318318367,
      "step": 1530
    },
    {
      "epoch": 38.5,
      "grad_norm": 0.49609375,
      "learning_rate": 0.0002,
      "loss": 0.0576020359992981,
      "loss_d0": 0.05768560767173767,
      "step": 1540
    },
    {
      "epoch": 38.75,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0002,
      "loss": 0.06362175345420837,
      "loss_d0": 0.06413243673741817,
      "step": 1550
    },
    {
      "epoch": 39.0,
      "grad_norm": 0.5,
      "learning_rate": 0.0002,
      "loss": 0.061510562896728516,
      "loss_d0": 0.061782880872488025,
      "step": 1560
    },
    {
      "epoch": 39.25,
      "grad_norm": 0.54296875,
      "learning_rate": 0.0002,
      "loss": 0.05666018724441528,
      "loss_d0": 0.05692802034318447,
      "step": 1570
    },
    {
      "epoch": 39.5,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0002,
      "loss": 0.06202887296676636,
      "loss_d0": 0.06217095740139485,
      "step": 1580
    },
    {
      "epoch": 39.75,
      "grad_norm": 0.71875,
      "learning_rate": 0.0002,
      "loss": 0.06877539157867432,
      "loss_d0": 0.0693345170468092,
      "step": 1590
    },
    {
      "epoch": 40.0,
      "grad_norm": 0.69921875,
      "learning_rate": 0.0002,
      "loss": 0.05825335383415222,
      "loss_d0": 0.0583782933652401,
      "step": 1600
    },
    {
      "epoch": 40.25,
      "grad_norm": 0.466796875,
      "learning_rate": 0.0002,
      "loss": 0.05497481226921082,
      "loss_d0": 0.0551537461578846,
      "step": 1610
    },
    {
      "epoch": 40.5,
      "grad_norm": 0.703125,
      "learning_rate": 0.0002,
      "loss": 0.060857725143432614,
      "loss_d0": 0.06165114678442478,
      "step": 1620
    },
    {
      "epoch": 40.75,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0002,
      "loss": 0.07385790944099427,
      "loss_d0": 0.07406756728887558,
      "step": 1630
    },
    {
      "epoch": 41.0,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0002,
      "loss": 0.0629690408706665,
      "loss_d0": 0.06308388896286488,
      "step": 1640
    },
    {
      "epoch": 41.25,
      "grad_norm": 0.4296875,
      "learning_rate": 0.0002,
      "loss": 0.054850274324417116,
      "loss_d0": 0.05506519712507725,
      "step": 1650
    },
    {
      "epoch": 41.5,
      "grad_norm": 0.478515625,
      "learning_rate": 0.0002,
      "loss": 0.054468965530395506,
      "loss_d0": 0.054471854120492935,
      "step": 1660
    },
    {
      "epoch": 41.75,
      "grad_norm": 0.41796875,
      "learning_rate": 0.0002,
      "loss": 0.05796172022819519,
      "loss_d0": 0.0579277478158474,
      "step": 1670
    },
    {
      "epoch": 42.0,
      "grad_norm": 0.41796875,
      "learning_rate": 0.0002,
      "loss": 0.05630323886871338,
      "loss_d0": 0.05661703534424305,
      "step": 1680
    },
    {
      "epoch": 42.25,
      "grad_norm": 0.416015625,
      "learning_rate": 0.0002,
      "loss": 0.054699838161468506,
      "loss_d0": 0.05516020208597183,
      "step": 1690
    },
    {
      "epoch": 42.5,
      "grad_norm": 0.39453125,
      "learning_rate": 0.0002,
      "loss": 0.058293914794921874,
      "loss_d0": 0.058815288543701175,
      "step": 1700
    },
    {
      "epoch": 42.75,
      "grad_norm": 0.42578125,
      "learning_rate": 0.0002,
      "loss": 0.057288384437561034,
      "loss_d0": 0.057978134229779245,
      "step": 1710
    },
    {
      "epoch": 43.0,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0002,
      "loss": 0.061647933721542356,
      "loss_d0": 0.061975281313061716,
      "step": 1720
    },
    {
      "epoch": 43.25,
      "grad_norm": 0.392578125,
      "learning_rate": 0.0002,
      "loss": 0.05214250087738037,
      "loss_d0": 0.05246074721217155,
      "step": 1730
    },
    {
      "epoch": 43.5,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0002,
      "loss": 0.05976005792617798,
      "loss_d0": 0.0598691787570715,
      "step": 1740
    },
    {
      "epoch": 43.75,
      "grad_norm": 0.44921875,
      "learning_rate": 0.0002,
      "loss": 0.0638785719871521,
      "loss_d0": 0.06381307877600192,
      "step": 1750
    },
    {
      "epoch": 44.0,
      "grad_norm": 0.435546875,
      "learning_rate": 0.0002,
      "loss": 0.06142584681510925,
      "loss_d0": 0.06180475577712059,
      "step": 1760
    },
    {
      "epoch": 44.25,
      "grad_norm": 0.39453125,
      "learning_rate": 0.0002,
      "loss": 0.05517058372497559,
      "loss_d0": 0.05530419759452343,
      "step": 1770
    },
    {
      "epoch": 44.5,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0002,
      "loss": 0.05579268932342529,
      "loss_d0": 0.056095103919506076,
      "step": 1780
    },
    {
      "epoch": 44.75,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0002,
      "loss": 0.06274228096008301,
      "loss_d0": 0.0628600399941206,
      "step": 1790
    },
    {
      "epoch": 45.0,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0002,
      "loss": 0.0648037850856781,
      "loss_d0": 0.06492549628019333,
      "step": 1800
    },
    {
      "epoch": 45.25,
      "grad_norm": 0.58203125,
      "learning_rate": 0.0002,
      "loss": 0.059372180700302125,
      "loss_d0": 0.059330419450998303,
      "step": 1810
    },
    {
      "epoch": 45.5,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0002,
      "loss": 0.05512714982032776,
      "loss_d0": 0.055520417913794515,
      "step": 1820
    },
    {
      "epoch": 45.75,
      "grad_norm": 0.419921875,
      "learning_rate": 0.0002,
      "loss": 0.05914598107337952,
      "loss_d0": 0.05946195349097252,
      "step": 1830
    },
    {
      "epoch": 46.0,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0002,
      "loss": 0.057545202970504764,
      "loss_d0": 0.057726585492491725,
      "step": 1840
    },
    {
      "epoch": 46.25,
      "grad_norm": 0.5,
      "learning_rate": 0.0002,
      "loss": 0.05496026873588562,
      "loss_d0": 0.0549175962805748,
      "step": 1850
    },
    {
      "epoch": 46.5,
      "grad_norm": 0.39453125,
      "learning_rate": 0.0002,
      "loss": 0.058978724479675296,
      "loss_d0": 0.05935865454375744,
      "step": 1860
    },
    {
      "epoch": 46.75,
      "grad_norm": 0.453125,
      "learning_rate": 0.0002,
      "loss": 0.05748326778411865,
      "loss_d0": 0.05784638747572899,
      "step": 1870
    },
    {
      "epoch": 47.0,
      "grad_norm": 0.38671875,
      "learning_rate": 0.0002,
      "loss": 0.06152337193489075,
      "loss_d0": 0.061656130105257036,
      "step": 1880
    },
    {
      "epoch": 47.25,
      "grad_norm": 0.44921875,
      "learning_rate": 0.0002,
      "loss": 0.05365397334098816,
      "loss_d0": 0.053687059134244916,
      "step": 1890
    },
    {
      "epoch": 47.5,
      "grad_norm": 0.31640625,
      "learning_rate": 0.0002,
      "loss": 0.05964898467063904,
      "loss_d0": 0.059862629324197766,
      "step": 1900
    },
    {
      "epoch": 47.75,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0002,
      "loss": 0.06160618662834168,
      "loss_d0": 0.061931657046079634,
      "step": 1910
    },
    {
      "epoch": 48.0,
      "grad_norm": 0.419921875,
      "learning_rate": 0.0002,
      "loss": 0.061159104108810425,
      "loss_d0": 0.06127244271337986,
      "step": 1920
    },
    {
      "epoch": 48.25,
      "grad_norm": 0.3828125,
      "learning_rate": 0.0002,
      "loss": 0.05568260550498962,
      "loss_d0": 0.056041232869029046,
      "step": 1930
    },
    {
      "epoch": 48.5,
      "grad_norm": 0.515625,
      "learning_rate": 0.0002,
      "loss": 0.05307339429855347,
      "loss_d0": 0.053248926252126697,
      "step": 1940
    },
    {
      "epoch": 48.75,
      "grad_norm": 0.44140625,
      "learning_rate": 0.0002,
      "loss": 0.05676477551460266,
      "loss_d0": 0.057009470835328105,
      "step": 1950
    },
    {
      "epoch": 49.0,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0002,
      "loss": 0.06322442889213561,
      "loss_d0": 0.0632315531373024,
      "step": 1960
    },
    {
      "epoch": 49.25,
      "grad_norm": 0.294921875,
      "learning_rate": 0.0002,
      "loss": 0.05208069682121277,
      "loss_d0": 0.05217769481241703,
      "step": 1970
    },
    {
      "epoch": 49.5,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0002,
      "loss": 0.05972009301185608,
      "loss_d0": 0.06001131311058998,
      "step": 1980
    },
    {
      "epoch": 49.75,
      "grad_norm": 0.431640625,
      "learning_rate": 0.0002,
      "loss": 0.06118760108947754,
      "loss_d0": 0.061306034401059153,
      "step": 1990
    },
    {
      "epoch": 50.0,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0002,
      "loss": 0.056265789270401004,
      "loss_d0": 0.056443619355559346,
      "step": 2000
    },
    {
      "epoch": 50.0,
      "eval_loss": 9.570272445678711,
      "eval_runtime": 0.3452,
      "eval_samples_per_second": 724.138,
      "eval_steps_per_second": 72.414,
      "step": 2000
    },
    {
      "epoch": 50.25,
      "grad_norm": 0.5625,
      "learning_rate": 0.0002,
      "loss": 0.05579528212547302,
      "loss_d0": 0.05573404803872108,
      "step": 2010
    },
    {
      "epoch": 50.5,
      "grad_norm": 0.42578125,
      "learning_rate": 0.0002,
      "loss": 0.05709593892097473,
      "loss_d0": 0.05716409236192703,
      "step": 2020
    },
    {
      "epoch": 50.75,
      "grad_norm": 0.4140625,
      "learning_rate": 0.0002,
      "loss": 0.05737549662590027,
      "loss_d0": 0.05729088075459003,
      "step": 2030
    },
    {
      "epoch": 51.0,
      "grad_norm": 0.5703125,
      "learning_rate": 0.0002,
      "loss": 0.05552052855491638,
      "loss_d0": 0.05591188482940197,
      "step": 2040
    },
    {
      "epoch": 51.25,
      "grad_norm": 0.349609375,
      "learning_rate": 0.0002,
      "loss": 0.050293844938278195,
      "loss_d0": 0.05039510801434517,
      "step": 2050
    },
    {
      "epoch": 51.5,
      "grad_norm": 0.453125,
      "learning_rate": 0.0002,
      "loss": 0.05217458605766297,
      "loss_d0": 0.05254131220281124,
      "step": 2060
    },
    {
      "epoch": 51.75,
      "grad_norm": 0.337890625,
      "learning_rate": 0.0002,
      "loss": 0.05404587984085083,
      "loss_d0": 0.05422148145735264,
      "step": 2070
    },
    {
      "epoch": 52.0,
      "grad_norm": 0.431640625,
      "learning_rate": 0.0002,
      "loss": 0.05949082374572754,
      "loss_d0": 0.05948565229773521,
      "step": 2080
    },
    {
      "epoch": 52.25,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0002,
      "loss": 0.057565832138061525,
      "loss_d0": 0.05814546495676041,
      "step": 2090
    },
    {
      "epoch": 52.5,
      "grad_norm": 0.388671875,
      "learning_rate": 0.0002,
      "loss": 0.052321404218673706,
      "loss_d0": 0.052815460786223414,
      "step": 2100
    },
    {
      "epoch": 52.75,
      "grad_norm": 0.4140625,
      "learning_rate": 0.0002,
      "loss": 0.058062076568603516,
      "loss_d0": 0.05846084877848625,
      "step": 2110
    },
    {
      "epoch": 53.0,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0002,
      "loss": 0.06171013712882996,
      "loss_d0": 0.06212322488427162,
      "step": 2120
    },
    {
      "epoch": 53.25,
      "grad_norm": 0.43359375,
      "learning_rate": 0.0002,
      "loss": 0.05460226535797119,
      "loss_d0": 0.05509419105947018,
      "step": 2130
    },
    {
      "epoch": 53.5,
      "grad_norm": 0.34765625,
      "learning_rate": 0.0002,
      "loss": 0.055398589372634886,
      "loss_d0": 0.055659336224198344,
      "step": 2140
    },
    {
      "epoch": 53.75,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0002,
      "loss": 0.05847206711769104,
      "loss_d0": 0.05927353985607624,
      "step": 2150
    },
    {
      "epoch": 54.0,
      "grad_norm": 0.392578125,
      "learning_rate": 0.0002,
      "loss": 0.06233756542205811,
      "loss_d0": 0.06288112588226795,
      "step": 2160
    },
    {
      "epoch": 54.25,
      "grad_norm": 0.46484375,
      "learning_rate": 0.0002,
      "loss": 0.05393359065055847,
      "loss_d0": 0.054285747557878496,
      "step": 2170
    },
    {
      "epoch": 54.5,
      "grad_norm": 0.37890625,
      "learning_rate": 0.0002,
      "loss": 0.05169063210487366,
      "loss_d0": 0.052193207666277885,
      "step": 2180
    },
    {
      "epoch": 54.75,
      "grad_norm": 0.37890625,
      "learning_rate": 0.0002,
      "loss": 0.0548134446144104,
      "loss_d0": 0.05492622479796409,
      "step": 2190
    },
    {
      "epoch": 55.0,
      "grad_norm": 0.4296875,
      "learning_rate": 0.0002,
      "loss": 0.05885629057884216,
      "loss_d0": 0.05889065600931644,
      "step": 2200
    },
    {
      "epoch": 55.25,
      "grad_norm": 0.287109375,
      "learning_rate": 0.0002,
      "loss": 0.0550631582736969,
      "loss_d0": 0.05513200797140598,
      "step": 2210
    },
    {
      "epoch": 55.5,
      "grad_norm": 0.40625,
      "learning_rate": 0.0002,
      "loss": 0.050622057914733884,
      "loss_d0": 0.051012787595391275,
      "step": 2220
    },
    {
      "epoch": 55.75,
      "grad_norm": 0.4140625,
      "learning_rate": 0.0002,
      "loss": 0.058926552534103394,
      "loss_d0": 0.05910953395068645,
      "step": 2230
    },
    {
      "epoch": 56.0,
      "grad_norm": 0.359375,
      "learning_rate": 0.0002,
      "loss": 0.05983954071998596,
      "loss_d0": 0.05987398028373718,
      "step": 2240
    },
    {
      "epoch": 56.25,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0002,
      "loss": 0.04893695116043091,
      "loss_d0": 0.04906199313700199,
      "step": 2250
    },
    {
      "epoch": 56.5,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0002,
      "loss": 0.05469831228256226,
      "loss_d0": 0.054801841452717784,
      "step": 2260
    },
    {
      "epoch": 56.75,
      "grad_norm": 0.453125,
      "learning_rate": 0.0002,
      "loss": 0.05877939462661743,
      "loss_d0": 0.058909759670495984,
      "step": 2270
    },
    {
      "epoch": 57.0,
      "grad_norm": 0.470703125,
      "learning_rate": 0.0002,
      "loss": 0.05709148049354553,
      "loss_d0": 0.057506448775529864,
      "step": 2280
    },
    {
      "epoch": 57.25,
      "grad_norm": 0.44140625,
      "learning_rate": 0.0002,
      "loss": 0.05298535823822022,
      "loss_d0": 0.05313003920018673,
      "step": 2290
    },
    {
      "epoch": 57.5,
      "grad_norm": 0.431640625,
      "learning_rate": 0.0002,
      "loss": 0.05260106325149536,
      "loss_d0": 0.052703161165118216,
      "step": 2300
    },
    {
      "epoch": 57.75,
      "grad_norm": 0.46875,
      "learning_rate": 0.0002,
      "loss": 0.05661575794219971,
      "loss_d0": 0.05684591718018055,
      "step": 2310
    },
    {
      "epoch": 58.0,
      "grad_norm": 0.41015625,
      "learning_rate": 0.0002,
      "loss": 0.05907288193702698,
      "loss_d0": 0.05924911350011826,
      "step": 2320
    },
    {
      "epoch": 58.25,
      "grad_norm": 0.41015625,
      "learning_rate": 0.0002,
      "loss": 0.05050989389419556,
      "loss_d0": 0.050834688171744344,
      "step": 2330
    },
    {
      "epoch": 58.5,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0002,
      "loss": 0.053637123107910155,
      "loss_d0": 0.05371711477637291,
      "step": 2340
    },
    {
      "epoch": 58.75,
      "grad_norm": 0.46484375,
      "learning_rate": 0.0002,
      "loss": 0.056911373138427736,
      "loss_d0": 0.05702054724097252,
      "step": 2350
    },
    {
      "epoch": 59.0,
      "grad_norm": 0.412109375,
      "learning_rate": 0.0002,
      "loss": 0.05872758030891419,
      "loss_d0": 0.0589618481695652,
      "step": 2360
    },
    {
      "epoch": 59.25,
      "grad_norm": 0.345703125,
      "learning_rate": 0.0002,
      "loss": 0.05019079446792603,
      "loss_d0": 0.050407799705862996,
      "step": 2370
    },
    {
      "epoch": 59.5,
      "grad_norm": 0.3515625,
      "learning_rate": 0.0002,
      "loss": 0.052620303630828855,
      "loss_d0": 0.05305602438747883,
      "step": 2380
    },
    {
      "epoch": 59.75,
      "grad_norm": 0.3046875,
      "learning_rate": 0.0002,
      "loss": 0.05450937747955322,
      "loss_d0": 0.05496942400932312,
      "step": 2390
    },
    {
      "epoch": 60.0,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0002,
      "loss": 0.05744432806968689,
      "loss_d0": 0.05768766738474369,
      "step": 2400
    },
    {
      "epoch": 60.25,
      "grad_norm": 0.32421875,
      "learning_rate": 0.0002,
      "loss": 0.051724094152450564,
      "loss_d0": 0.05191577039659023,
      "step": 2410
    },
    {
      "epoch": 60.5,
      "grad_norm": 0.423828125,
      "learning_rate": 0.0002,
      "loss": 0.053571826219558714,
      "loss_d0": 0.053789060562849045,
      "step": 2420
    },
    {
      "epoch": 60.75,
      "grad_norm": 0.3828125,
      "learning_rate": 0.0002,
      "loss": 0.0588623046875,
      "loss_d0": 0.059234515950083735,
      "step": 2430
    },
    {
      "epoch": 61.0,
      "grad_norm": 0.48828125,
      "learning_rate": 0.0002,
      "loss": 0.058709746599197386,
      "loss_d0": 0.059247952327132224,
      "step": 2440
    },
    {
      "epoch": 61.25,
      "grad_norm": 0.392578125,
      "learning_rate": 0.0002,
      "loss": 0.059135335683822635,
      "loss_d0": 0.0596803393214941,
      "step": 2450
    },
    {
      "epoch": 61.5,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0002,
      "loss": 0.05420001745223999,
      "loss_d0": 0.05439855307340622,
      "step": 2460
    },
    {
      "epoch": 61.75,
      "grad_norm": 0.36328125,
      "learning_rate": 0.0002,
      "loss": 0.051242280006408694,
      "loss_d0": 0.05156247094273567,
      "step": 2470
    },
    {
      "epoch": 62.0,
      "grad_norm": 0.330078125,
      "learning_rate": 0.0002,
      "loss": 0.05718609094619751,
      "loss_d0": 0.05745496265590191,
      "step": 2480
    },
    {
      "epoch": 62.25,
      "grad_norm": 0.35546875,
      "learning_rate": 0.0002,
      "loss": 0.04888695478439331,
      "loss_d0": 0.04925457388162613,
      "step": 2490
    },
    {
      "epoch": 62.5,
      "grad_norm": 0.38671875,
      "learning_rate": 0.0002,
      "loss": 0.051664423942565915,
      "loss_d0": 0.05186402052640915,
      "step": 2500
    },
    {
      "epoch": 62.5,
      "eval_loss": 8.208417892456055,
      "eval_runtime": 0.3448,
      "eval_samples_per_second": 725.048,
      "eval_steps_per_second": 72.505,
      "step": 2500
    },
    {
      "epoch": 62.75,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0002,
      "loss": 0.05614182949066162,
      "loss_d0": 0.05651816315948963,
      "step": 2510
    },
    {
      "epoch": 63.0,
      "grad_norm": 0.423828125,
      "learning_rate": 0.0002,
      "loss": 0.053869009017944336,
      "loss_d0": 0.05419423021376133,
      "step": 2520
    },
    {
      "epoch": 63.25,
      "grad_norm": 0.310546875,
      "learning_rate": 0.0002,
      "loss": 0.05018726587295532,
      "loss_d0": 0.05073915831744671,
      "step": 2530
    },
    {
      "epoch": 63.5,
      "grad_norm": 0.38671875,
      "learning_rate": 0.0002,
      "loss": 0.05394603013992309,
      "loss_d0": 0.05411759801208973,
      "step": 2540
    },
    {
      "epoch": 63.75,
      "grad_norm": 0.390625,
      "learning_rate": 0.0002,
      "loss": 0.05712904930114746,
      "loss_d0": 0.057449503988027575,
      "step": 2550
    },
    {
      "epoch": 64.0,
      "grad_norm": 0.416015625,
      "learning_rate": 0.0002,
      "loss": 0.05660791397094726,
      "loss_d0": 0.05653563365340233,
      "step": 2560
    },
    {
      "epoch": 64.25,
      "grad_norm": 0.392578125,
      "learning_rate": 0.0002,
      "loss": 0.053469395637512206,
      "loss_d0": 0.05360211282968521,
      "step": 2570
    },
    {
      "epoch": 64.5,
      "grad_norm": 0.361328125,
      "learning_rate": 0.0002,
      "loss": 0.05392343997955322,
      "loss_d0": 0.0542771577835083,
      "step": 2580
    },
    {
      "epoch": 64.75,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0002,
      "loss": 0.05510370135307312,
      "loss_d0": 0.0556726049631834,
      "step": 2590
    },
    {
      "epoch": 65.0,
      "grad_norm": 0.44921875,
      "learning_rate": 0.0002,
      "loss": 0.05375306606292725,
      "loss_d0": 0.054018044471740724,
      "step": 2600
    },
    {
      "epoch": 65.25,
      "grad_norm": 0.3671875,
      "learning_rate": 0.0002,
      "loss": 0.04939930737018585,
      "loss_d0": 0.04965954720973968,
      "step": 2610
    },
    {
      "epoch": 65.5,
      "grad_norm": 0.365234375,
      "learning_rate": 0.0002,
      "loss": 0.04847543239593506,
      "loss_d0": 0.04878879524767399,
      "step": 2620
    },
    {
      "epoch": 65.75,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0002,
      "loss": 0.05529288053512573,
      "loss_d0": 0.05569577738642693,
      "step": 2630
    },
    {
      "epoch": 66.0,
      "grad_norm": 0.5625,
      "learning_rate": 0.0002,
      "loss": 0.058045816421508786,
      "loss_d0": 0.058395038917660715,
      "step": 2640
    },
    {
      "epoch": 66.25,
      "grad_norm": 0.31640625,
      "learning_rate": 0.0002,
      "loss": 0.050916659832000735,
      "loss_d0": 0.05154414139688015,
      "step": 2650
    },
    {
      "epoch": 66.5,
      "grad_norm": 0.42578125,
      "learning_rate": 0.0002,
      "loss": 0.05039163827896118,
      "loss_d0": 0.050868920236825946,
      "step": 2660
    },
    {
      "epoch": 66.75,
      "grad_norm": 0.359375,
      "learning_rate": 0.0002,
      "loss": 0.0557156503200531,
      "loss_d0": 0.05598639473319054,
      "step": 2670
    },
    {
      "epoch": 67.0,
      "grad_norm": 0.435546875,
      "learning_rate": 0.0002,
      "loss": 0.05896541476249695,
      "loss_d0": 0.059347905591130255,
      "step": 2680
    },
    {
      "epoch": 67.25,
      "grad_norm": 0.46484375,
      "learning_rate": 0.0002,
      "loss": 0.04818243384361267,
      "loss_d0": 0.0484800923615694,
      "step": 2690
    },
    {
      "epoch": 67.5,
      "grad_norm": 0.39453125,
      "learning_rate": 0.0002,
      "loss": 0.055252212285995486,
      "loss_d0": 0.05612879283726215,
      "step": 2700
    },
    {
      "epoch": 67.75,
      "grad_norm": 0.37109375,
      "learning_rate": 0.0002,
      "loss": 0.05248673558235169,
      "loss_d0": 0.05283830314874649,
      "step": 2710
    },
    {
      "epoch": 68.0,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0002,
      "loss": 0.05527347922325134,
      "loss_d0": 0.055605581402778624,
      "step": 2720
    },
    {
      "epoch": 68.25,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0002,
      "loss": 0.05481758713722229,
      "loss_d0": 0.055027333274483684,
      "step": 2730
    },
    {
      "epoch": 68.5,
      "grad_norm": 0.380859375,
      "learning_rate": 0.0002,
      "loss": 0.05644075870513916,
      "loss_d0": 0.056725147366523745,
      "step": 2740
    },
    {
      "epoch": 68.75,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0002,
      "loss": 0.057813364267349246,
      "loss_d0": 0.05852428674697876,
      "step": 2750
    },
    {
      "epoch": 69.0,
      "grad_norm": 0.49609375,
      "learning_rate": 0.0002,
      "loss": 0.059580165147781375,
      "loss_d0": 0.06016777530312538,
      "step": 2760
    },
    {
      "epoch": 69.25,
      "grad_norm": 0.38671875,
      "learning_rate": 0.0002,
      "loss": 0.05708110332489014,
      "loss_d0": 0.05751705914735794,
      "step": 2770
    },
    {
      "epoch": 69.5,
      "grad_norm": 0.44140625,
      "learning_rate": 0.0002,
      "loss": 0.05195104479789734,
      "loss_d0": 0.052196959406137465,
      "step": 2780
    },
    {
      "epoch": 69.75,
      "grad_norm": 0.484375,
      "learning_rate": 0.0002,
      "loss": 0.05186697244644165,
      "loss_d0": 0.05212553851306438,
      "step": 2790
    },
    {
      "epoch": 70.0,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0002,
      "loss": 0.05758950710296631,
      "loss_d0": 0.05797795578837395,
      "step": 2800
    },
    {
      "epoch": 70.25,
      "grad_norm": 0.353515625,
      "learning_rate": 0.0002,
      "loss": 0.04725401401519776,
      "loss_d0": 0.047584396228194235,
      "step": 2810
    },
    {
      "epoch": 70.5,
      "grad_norm": 0.33203125,
      "learning_rate": 0.0002,
      "loss": 0.05146564841270447,
      "loss_d0": 0.05187108106911183,
      "step": 2820
    },
    {
      "epoch": 70.75,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0002,
      "loss": 0.05590651035308838,
      "loss_d0": 0.056247909739613536,
      "step": 2830
    },
    {
      "epoch": 71.0,
      "grad_norm": 0.43359375,
      "learning_rate": 0.0002,
      "loss": 0.055702883005142215,
      "loss_d0": 0.05585823506116867,
      "step": 2840
    },
    {
      "epoch": 71.25,
      "grad_norm": 0.4140625,
      "learning_rate": 0.0002,
      "loss": 0.0496380478143692,
      "loss_d0": 0.04990372993052006,
      "step": 2850
    },
    {
      "epoch": 71.5,
      "grad_norm": 0.384765625,
      "learning_rate": 0.0002,
      "loss": 0.0478007048368454,
      "loss_d0": 0.048089010268449785,
      "step": 2860
    },
    {
      "epoch": 71.75,
      "grad_norm": 0.421875,
      "learning_rate": 0.0002,
      "loss": 0.05283045768737793,
      "loss_d0": 0.05304308496415615,
      "step": 2870
    },
    {
      "epoch": 72.0,
      "grad_norm": 0.35546875,
      "learning_rate": 0.0002,
      "loss": 0.05530112385749817,
      "loss_d0": 0.055600708723068236,
      "step": 2880
    },
    {
      "epoch": 72.25,
      "grad_norm": 0.333984375,
      "learning_rate": 0.0002,
      "loss": 0.04873108565807342,
      "loss_d0": 0.049191296473145484,
      "step": 2890
    },
    {
      "epoch": 72.5,
      "grad_norm": 0.326171875,
      "learning_rate": 0.0002,
      "loss": 0.046818670630455014,
      "loss_d0": 0.04713470712304115,
      "step": 2900
    },
    {
      "epoch": 72.75,
      "grad_norm": 0.431640625,
      "learning_rate": 0.0002,
      "loss": 0.0500145673751831,
      "loss_d0": 0.05044983923435211,
      "step": 2910
    },
    {
      "epoch": 73.0,
      "grad_norm": 0.396484375,
      "learning_rate": 0.0002,
      "loss": 0.052328604459762576,
      "loss_d0": 0.052925263345241544,
      "step": 2920
    },
    {
      "epoch": 73.25,
      "grad_norm": 0.478515625,
      "learning_rate": 0.0002,
      "loss": 0.05180770754814148,
      "loss_d0": 0.05192505940794945,
      "step": 2930
    },
    {
      "epoch": 73.5,
      "grad_norm": 0.375,
      "learning_rate": 0.0002,
      "loss": 0.0494821310043335,
      "loss_d0": 0.049676749110221866,
      "step": 2940
    },
    {
      "epoch": 73.75,
      "grad_norm": 0.466796875,
      "learning_rate": 0.0002,
      "loss": 0.052517324686050415,
      "loss_d0": 0.05272014886140823,
      "step": 2950
    },
    {
      "epoch": 74.0,
      "grad_norm": 0.49609375,
      "learning_rate": 0.0002,
      "loss": 0.05500271916389465,
      "loss_d0": 0.0552448246628046,
      "step": 2960
    },
    {
      "epoch": 74.25,
      "grad_norm": 0.33203125,
      "learning_rate": 0.0002,
      "loss": 0.05111870765686035,
      "loss_d0": 0.05123648978769779,
      "step": 2970
    },
    {
      "epoch": 74.5,
      "grad_norm": 0.609375,
      "learning_rate": 0.0002,
      "loss": 0.05168828368186951,
      "loss_d0": 0.05195821598172188,
      "step": 2980
    },
    {
      "epoch": 74.75,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0002,
      "loss": 0.06273810863494873,
      "loss_d0": 0.06293985731899739,
      "step": 2990
    },
    {
      "epoch": 75.0,
      "grad_norm": 0.3515625,
      "learning_rate": 0.0002,
      "loss": 0.061145073175430296,
      "loss_d0": 0.06176314204931259,
      "step": 3000
    },
    {
      "epoch": 75.0,
      "eval_loss": 9.752898216247559,
      "eval_runtime": 0.3445,
      "eval_samples_per_second": 725.747,
      "eval_steps_per_second": 72.575,
      "step": 3000
    }
  ],
  "logging_steps": 10,
  "max_steps": 8000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 200,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.99653793349632e+16,
  "train_batch_size": 10,
  "trial_name": null,
  "trial_params": null
}