{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 500.0,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 5.453,
      "step": 1
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.349853515625,
      "learning_rate": 0.0001,
      "loss": 5.453,
      "step": 2
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.349853515625,
      "learning_rate": 0.0002,
      "loss": 5.453,
      "step": 3
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.8772072792053223,
      "learning_rate": 0.0001995983935742972,
      "loss": 5.1576,
      "step": 4
    },
    {
      "epoch": 5.0,
      "grad_norm": 5.726413726806641,
      "learning_rate": 0.0001991967871485944,
      "loss": 4.6773,
      "step": 5
    },
    {
      "epoch": 6.0,
      "grad_norm": 8.641162872314453,
      "learning_rate": 0.00019879518072289158,
      "loss": 4.2517,
      "step": 6
    },
    {
      "epoch": 7.0,
      "grad_norm": 11.281049728393555,
      "learning_rate": 0.00019839357429718877,
      "loss": 3.8401,
      "step": 7
    },
    {
      "epoch": 8.0,
      "grad_norm": 13.561233520507812,
      "learning_rate": 0.00019799196787148596,
      "loss": 3.428,
      "step": 8
    },
    {
      "epoch": 9.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00019799196787148596,
      "loss": 2.9945,
      "step": 9
    },
    {
      "epoch": 10.0,
      "grad_norm": 15.408284187316895,
      "learning_rate": 0.00019759036144578314,
      "loss": 2.9945,
      "step": 10
    },
    {
      "epoch": 11.0,
      "grad_norm": 16.737504959106445,
      "learning_rate": 0.00019718875502008033,
      "loss": 2.5293,
      "step": 11
    },
    {
      "epoch": 12.0,
      "grad_norm": 17.475238800048828,
      "learning_rate": 0.00019678714859437752,
      "loss": 2.0314,
      "step": 12
    },
    {
      "epoch": 13.0,
      "grad_norm": 17.607587814331055,
      "learning_rate": 0.0001963855421686747,
      "loss": 1.5013,
      "step": 13
    },
    {
      "epoch": 14.0,
      "grad_norm": 17.160503387451172,
      "learning_rate": 0.0001959839357429719,
      "loss": 0.9549,
      "step": 14
    },
    {
      "epoch": 15.0,
      "grad_norm": 16.20315933227539,
      "learning_rate": 0.00019558232931726906,
      "loss": 0.4056,
      "step": 15
    },
    {
      "epoch": 16.0,
      "grad_norm": 2.948519229888916,
      "learning_rate": 0.00019518072289156628,
      "loss": 0.1432,
      "step": 16
    },
    {
      "epoch": 17.0,
      "grad_norm": 3.228358507156372,
      "learning_rate": 0.00019477911646586347,
      "loss": 0.1246,
      "step": 17
    },
    {
      "epoch": 18.0,
      "grad_norm": 3.2282843589782715,
      "learning_rate": 0.00019437751004016066,
      "loss": 0.1034,
      "step": 18
    },
    {
      "epoch": 19.0,
      "grad_norm": 1.0867902040481567,
      "learning_rate": 0.00019397590361445782,
      "loss": 0.0786,
      "step": 19
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.5086488723754883,
      "learning_rate": 0.00019357429718875504,
      "loss": 0.0706,
      "step": 20
    },
    {
      "epoch": 21.0,
      "grad_norm": 0.4829910397529602,
      "learning_rate": 0.00019317269076305223,
      "loss": 0.063,
      "step": 21
    },
    {
      "epoch": 22.0,
      "grad_norm": 0.5215936899185181,
      "learning_rate": 0.00019277108433734942,
      "loss": 0.0565,
      "step": 22
    },
    {
      "epoch": 23.0,
      "grad_norm": 0.5226811766624451,
      "learning_rate": 0.00019236947791164658,
      "loss": 0.0445,
      "step": 23
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.5145213603973389,
      "learning_rate": 0.00019196787148594377,
      "loss": 0.0316,
      "step": 24
    },
    {
      "epoch": 25.0,
      "grad_norm": 0.50876384973526,
      "learning_rate": 0.00019156626506024098,
      "loss": 0.0197,
      "step": 25
    },
    {
      "epoch": 26.0,
      "grad_norm": 0.23416705429553986,
      "learning_rate": 0.00019116465863453817,
      "loss": 0.0081,
      "step": 26
    },
    {
      "epoch": 27.0,
      "grad_norm": 0.07123460620641708,
      "learning_rate": 0.00019076305220883533,
      "loss": 0.0055,
      "step": 27
    },
    {
      "epoch": 28.0,
      "grad_norm": 0.17463913559913635,
      "learning_rate": 0.00019036144578313252,
      "loss": 0.0056,
      "step": 28
    },
    {
      "epoch": 29.0,
      "grad_norm": 0.08099503815174103,
      "learning_rate": 0.00018995983935742974,
      "loss": 0.005,
      "step": 29
    },
    {
      "epoch": 30.0,
      "grad_norm": 0.10401125252246857,
      "learning_rate": 0.00018955823293172693,
      "loss": 0.005,
      "step": 30
    },
    {
      "epoch": 31.0,
      "grad_norm": 0.18095582723617554,
      "learning_rate": 0.0001891566265060241,
      "loss": 0.0057,
      "step": 31
    },
    {
      "epoch": 32.0,
      "grad_norm": 0.056214649230241776,
      "learning_rate": 0.00018875502008032128,
      "loss": 0.0048,
      "step": 32
    },
    {
      "epoch": 33.0,
      "grad_norm": 0.12311957776546478,
      "learning_rate": 0.0001883534136546185,
      "loss": 0.0051,
      "step": 33
    },
    {
      "epoch": 34.0,
      "grad_norm": 0.1186085045337677,
      "learning_rate": 0.00018795180722891569,
      "loss": 0.0052,
      "step": 34
    },
    {
      "epoch": 35.0,
      "grad_norm": 0.05622096359729767,
      "learning_rate": 0.00018755020080321285,
      "loss": 0.0048,
      "step": 35
    },
    {
      "epoch": 36.0,
      "grad_norm": 0.1179257333278656,
      "learning_rate": 0.00018714859437751004,
      "loss": 0.0051,
      "step": 36
    },
    {
      "epoch": 37.0,
      "grad_norm": 0.02104870229959488,
      "learning_rate": 0.00018674698795180723,
      "loss": 0.005,
      "step": 37
    },
    {
      "epoch": 38.0,
      "grad_norm": 0.09374430030584335,
      "learning_rate": 0.00018634538152610444,
      "loss": 0.005,
      "step": 38
    },
    {
      "epoch": 39.0,
      "grad_norm": 0.03337598219513893,
      "learning_rate": 0.0001859437751004016,
      "loss": 0.0048,
      "step": 39
    },
    {
      "epoch": 40.0,
      "grad_norm": 0.09240953624248505,
      "learning_rate": 0.0001855421686746988,
      "loss": 0.005,
      "step": 40
    },
    {
      "epoch": 41.0,
      "grad_norm": 0.02956731617450714,
      "learning_rate": 0.00018514056224899598,
      "loss": 0.0048,
      "step": 41
    },
    {
      "epoch": 42.0,
      "grad_norm": 0.08981137722730637,
      "learning_rate": 0.0001847389558232932,
      "loss": 0.005,
      "step": 42
    },
    {
      "epoch": 43.0,
      "grad_norm": 0.0320754237473011,
      "learning_rate": 0.00018433734939759036,
      "loss": 0.0048,
      "step": 43
    },
    {
      "epoch": 44.0,
      "grad_norm": 0.088747039437294,
      "learning_rate": 0.00018393574297188755,
      "loss": 0.005,
      "step": 44
    },
    {
      "epoch": 45.0,
      "grad_norm": 0.04862065240740776,
      "learning_rate": 0.00018353413654618474,
      "loss": 0.0048,
      "step": 45
    },
    {
      "epoch": 46.0,
      "grad_norm": 0.08655441552400589,
      "learning_rate": 0.00018313253012048193,
      "loss": 0.005,
      "step": 46
    },
    {
      "epoch": 47.0,
      "grad_norm": 0.06792537122964859,
      "learning_rate": 0.00018273092369477912,
      "loss": 0.0049,
      "step": 47
    },
    {
      "epoch": 48.0,
      "grad_norm": 0.08522295206785202,
      "learning_rate": 0.0001823293172690763,
      "loss": 0.005,
      "step": 48
    },
    {
      "epoch": 49.0,
      "grad_norm": 0.04663475975394249,
      "learning_rate": 0.0001819277108433735,
      "loss": 0.0048,
      "step": 49
    },
    {
      "epoch": 50.0,
      "grad_norm": 0.029744861647486687,
      "learning_rate": 0.0001815261044176707,
      "loss": 0.0048,
      "step": 50
    },
    {
      "epoch": 51.0,
      "grad_norm": 0.06534643471240997,
      "learning_rate": 0.0001811244979919679,
      "loss": 0.0049,
      "step": 51
    },
    {
      "epoch": 52.0,
      "grad_norm": 0.007808285299688578,
      "learning_rate": 0.00018072289156626507,
      "loss": 0.0047,
      "step": 52
    },
    {
      "epoch": 53.0,
      "grad_norm": 0.0546131432056427,
      "learning_rate": 0.00018032128514056225,
      "loss": 0.0051,
      "step": 53
    },
    {
      "epoch": 54.0,
      "grad_norm": 0.016065070405602455,
      "learning_rate": 0.00017991967871485944,
      "loss": 0.0045,
      "step": 54
    },
    {
      "epoch": 55.0,
      "grad_norm": 0.02857162430882454,
      "learning_rate": 0.00017951807228915663,
      "loss": 0.0048,
      "step": 55
    },
    {
      "epoch": 56.0,
      "grad_norm": 0.03628357872366905,
      "learning_rate": 0.00017911646586345382,
      "loss": 0.005,
      "step": 56
    },
    {
      "epoch": 57.0,
      "grad_norm": 0.0075448257848620415,
      "learning_rate": 0.000178714859437751,
      "loss": 0.0047,
      "step": 57
    },
    {
      "epoch": 58.0,
      "grad_norm": 0.04328390583395958,
      "learning_rate": 0.0001783132530120482,
      "loss": 0.0048,
      "step": 58
    },
    {
      "epoch": 59.0,
      "grad_norm": 0.0074650561437010765,
      "learning_rate": 0.0001779116465863454,
      "loss": 0.0047,
      "step": 59
    },
    {
      "epoch": 60.0,
      "grad_norm": 0.04476163163781166,
      "learning_rate": 0.00017751004016064258,
      "loss": 0.0048,
      "step": 60
    },
    {
      "epoch": 61.0,
      "grad_norm": 0.010204787366092205,
      "learning_rate": 0.00017710843373493977,
      "loss": 0.0047,
      "step": 61
    },
    {
      "epoch": 62.0,
      "grad_norm": 0.007382046431303024,
      "learning_rate": 0.00017670682730923696,
      "loss": 0.0047,
      "step": 62
    },
    {
      "epoch": 63.0,
      "grad_norm": 0.042141955345869064,
      "learning_rate": 0.00017630522088353415,
      "loss": 0.0048,
      "step": 63
    },
    {
      "epoch": 64.0,
      "grad_norm": 0.007281558588147163,
      "learning_rate": 0.00017590361445783134,
      "loss": 0.0047,
      "step": 64
    },
    {
      "epoch": 65.0,
      "grad_norm": 0.04382139816880226,
      "learning_rate": 0.00017550200803212853,
      "loss": 0.0048,
      "step": 65
    },
    {
      "epoch": 66.0,
      "grad_norm": 0.027046991512179375,
      "learning_rate": 0.00017510040160642571,
      "loss": 0.0048,
      "step": 66
    },
    {
      "epoch": 67.0,
      "grad_norm": 0.041668713092803955,
      "learning_rate": 0.0001746987951807229,
      "loss": 0.0048,
      "step": 67
    },
    {
      "epoch": 68.0,
      "grad_norm": 0.024397345259785652,
      "learning_rate": 0.0001742971887550201,
      "loss": 0.0048,
      "step": 68
    },
    {
      "epoch": 69.0,
      "grad_norm": 0.010038570500910282,
      "learning_rate": 0.00017389558232931728,
      "loss": 0.0047,
      "step": 69
    },
    {
      "epoch": 70.0,
      "grad_norm": 0.04328250139951706,
      "learning_rate": 0.00017349397590361447,
      "loss": 0.0048,
      "step": 70
    },
    {
      "epoch": 71.0,
      "grad_norm": 0.007169268559664488,
      "learning_rate": 0.00017309236947791166,
      "loss": 0.0047,
      "step": 71
    },
    {
      "epoch": 72.0,
      "grad_norm": 0.024196363985538483,
      "learning_rate": 0.00017269076305220885,
      "loss": 0.0048,
      "step": 72
    },
    {
      "epoch": 73.0,
      "grad_norm": 0.016377076506614685,
      "learning_rate": 0.00017228915662650604,
      "loss": 0.005,
      "step": 73
    },
    {
      "epoch": 74.0,
      "grad_norm": 0.009935774840414524,
      "learning_rate": 0.00017188755020080323,
      "loss": 0.0047,
      "step": 74
    },
    {
      "epoch": 75.0,
      "grad_norm": 0.026620058342814445,
      "learning_rate": 0.00017148594377510042,
      "loss": 0.0048,
      "step": 75
    },
    {
      "epoch": 76.0,
      "grad_norm": 0.00991370715200901,
      "learning_rate": 0.0001710843373493976,
      "loss": 0.0047,
      "step": 76
    },
    {
      "epoch": 77.0,
      "grad_norm": 0.041056908667087555,
      "learning_rate": 0.00017068273092369477,
      "loss": 0.0048,
      "step": 77
    },
    {
      "epoch": 78.0,
      "grad_norm": 0.0022562043741345406,
      "learning_rate": 0.00017028112449799199,
      "loss": 0.0045,
      "step": 78
    },
    {
      "epoch": 79.0,
      "grad_norm": 0.009892994537949562,
      "learning_rate": 0.00016987951807228917,
      "loss": 0.0047,
      "step": 79
    },
    {
      "epoch": 80.0,
      "grad_norm": 0.026598593220114708,
      "learning_rate": 0.00016947791164658636,
      "loss": 0.0048,
      "step": 80
    },
    {
      "epoch": 81.0,
      "grad_norm": 0.007094119675457478,
      "learning_rate": 0.00016907630522088353,
      "loss": 0.0047,
      "step": 81
    },
    {
      "epoch": 82.0,
      "grad_norm": 0.024074360728263855,
      "learning_rate": 0.00016867469879518074,
      "loss": 0.0048,
      "step": 82
    },
    {
      "epoch": 83.0,
      "grad_norm": 0.007143808528780937,
      "learning_rate": 0.00016827309236947793,
      "loss": 0.0047,
      "step": 83
    },
    {
      "epoch": 84.0,
      "grad_norm": 0.00991752091795206,
      "learning_rate": 0.00016787148594377512,
      "loss": 0.0047,
      "step": 84
    },
    {
      "epoch": 85.0,
      "grad_norm": 0.026643570512533188,
      "learning_rate": 0.00016746987951807228,
      "loss": 0.0048,
      "step": 85
    },
    {
      "epoch": 86.0,
      "grad_norm": 0.0071501159109175205,
      "learning_rate": 0.00016706827309236947,
      "loss": 0.0047,
      "step": 86
    },
    {
      "epoch": 87.0,
      "grad_norm": 0.041228219866752625,
      "learning_rate": 0.0001666666666666667,
      "loss": 0.0048,
      "step": 87
    },
    {
      "epoch": 88.0,
      "grad_norm": 0.026733651757240295,
      "learning_rate": 0.00016626506024096388,
      "loss": 0.0048,
      "step": 88
    },
    {
      "epoch": 89.0,
      "grad_norm": 0.01906830444931984,
      "learning_rate": 0.00016586345381526104,
      "loss": 0.0045,
      "step": 89
    },
    {
      "epoch": 90.0,
      "grad_norm": 0.016484271734952927,
      "learning_rate": 0.00016546184738955823,
      "loss": 0.005,
      "step": 90
    },
    {
      "epoch": 91.0,
      "grad_norm": 0.007174884434789419,
      "learning_rate": 0.00016506024096385545,
      "loss": 0.0047,
      "step": 91
    },
    {
      "epoch": 92.0,
      "grad_norm": 0.007238594815135002,
      "learning_rate": 0.00016465863453815263,
      "loss": 0.0047,
      "step": 92
    },
    {
      "epoch": 93.0,
      "grad_norm": 0.010034309700131416,
      "learning_rate": 0.0001642570281124498,
      "loss": 0.0047,
      "step": 93
    },
    {
      "epoch": 94.0,
      "grad_norm": 0.02700984664261341,
      "learning_rate": 0.00016385542168674699,
      "loss": 0.0048,
      "step": 94
    },
    {
      "epoch": 95.0,
      "grad_norm": 0.015189659781754017,
      "learning_rate": 0.00016345381526104417,
      "loss": 0.0045,
      "step": 95
    },
    {
      "epoch": 96.0,
      "grad_norm": 0.024589456617832184,
      "learning_rate": 0.0001630522088353414,
      "loss": 0.0048,
      "step": 96
    },
    {
      "epoch": 97.0,
      "grad_norm": 0.001375726773403585,
      "learning_rate": 0.00016265060240963855,
      "loss": 0.0049,
      "step": 97
    },
    {
      "epoch": 98.0,
      "grad_norm": 0.04419811815023422,
      "learning_rate": 0.00016224899598393574,
      "loss": 0.0048,
      "step": 98
    },
    {
      "epoch": 99.0,
      "grad_norm": 0.007295660208910704,
      "learning_rate": 0.00016184738955823293,
      "loss": 0.0047,
      "step": 99
    },
    {
      "epoch": 100.0,
      "grad_norm": 0.0422837920486927,
      "learning_rate": 0.00016144578313253015,
      "loss": 0.0048,
      "step": 100
    },
    {
      "epoch": 101.0,
      "grad_norm": 0.007468740921467543,
      "learning_rate": 0.0001610441767068273,
      "loss": 0.0047,
      "step": 101
    },
    {
      "epoch": 102.0,
      "grad_norm": 0.052220698446035385,
      "learning_rate": 0.0001606425702811245,
      "loss": 0.0051,
      "step": 102
    },
    {
      "epoch": 103.0,
      "grad_norm": 0.010211293585598469,
      "learning_rate": 0.0001602409638554217,
      "loss": 0.0047,
      "step": 103
    },
    {
      "epoch": 104.0,
      "grad_norm": 0.0423286072909832,
      "learning_rate": 0.00015983935742971888,
      "loss": 0.0048,
      "step": 104
    },
    {
      "epoch": 105.0,
      "grad_norm": 0.02506270445883274,
      "learning_rate": 0.00015943775100401607,
      "loss": 0.0048,
      "step": 105
    },
    {
      "epoch": 106.0,
      "grad_norm": 0.04458456113934517,
      "learning_rate": 0.00015903614457831326,
      "loss": 0.0048,
      "step": 106
    },
    {
      "epoch": 107.0,
      "grad_norm": 0.02753458172082901,
      "learning_rate": 0.00015863453815261045,
      "loss": 0.0048,
      "step": 107
    },
    {
      "epoch": 108.0,
      "grad_norm": 0.04234497249126434,
      "learning_rate": 0.00015823293172690763,
      "loss": 0.0048,
      "step": 108
    },
    {
      "epoch": 109.0,
      "grad_norm": 0.03287286311388016,
      "learning_rate": 0.00015783132530120482,
      "loss": 0.0046,
      "step": 109
    },
    {
      "epoch": 110.0,
      "grad_norm": 0.03671009838581085,
      "learning_rate": 0.000157429718875502,
      "loss": 0.0046,
      "step": 110
    },
    {
      "epoch": 111.0,
      "grad_norm": 0.044372282922267914,
      "learning_rate": 0.0001570281124497992,
      "loss": 0.0048,
      "step": 111
    },
    {
      "epoch": 112.0,
      "grad_norm": 0.03433435037732124,
      "learning_rate": 0.0001566265060240964,
      "loss": 0.005,
      "step": 112
    },
    {
      "epoch": 113.0,
      "grad_norm": 0.04201051965355873,
      "learning_rate": 0.00015622489959839358,
      "loss": 0.0048,
      "step": 113
    },
    {
      "epoch": 114.0,
      "grad_norm": 0.027228495106101036,
      "learning_rate": 0.00015582329317269077,
      "loss": 0.0048,
      "step": 114
    },
    {
      "epoch": 115.0,
      "grad_norm": 0.04395684599876404,
      "learning_rate": 0.00015542168674698796,
      "loss": 0.0048,
      "step": 115
    },
    {
      "epoch": 116.0,
      "grad_norm": 0.0418144129216671,
      "learning_rate": 0.00015502008032128515,
      "loss": 0.0048,
      "step": 116
    },
    {
      "epoch": 117.0,
      "grad_norm": 0.015051459893584251,
      "learning_rate": 0.00015461847389558234,
      "loss": 0.0045,
      "step": 117
    },
    {
      "epoch": 118.0,
      "grad_norm": 0.010102898813784122,
      "learning_rate": 0.00015421686746987953,
      "loss": 0.0047,
      "step": 118
    },
    {
      "epoch": 119.0,
      "grad_norm": 0.026845330372452736,
      "learning_rate": 0.00015381526104417672,
      "loss": 0.0048,
      "step": 119
    },
    {
      "epoch": 120.0,
      "grad_norm": 0.010000316426157951,
      "learning_rate": 0.0001534136546184739,
      "loss": 0.0047,
      "step": 120
    },
    {
      "epoch": 121.0,
      "grad_norm": 0.041023530066013336,
      "learning_rate": 0.0001530120481927711,
      "loss": 0.0048,
      "step": 121
    },
    {
      "epoch": 122.0,
      "grad_norm": 0.0072840056382119656,
      "learning_rate": 0.00015261044176706828,
      "loss": 0.0047,
      "step": 122
    },
    {
      "epoch": 123.0,
      "grad_norm": 0.02653086557984352,
      "learning_rate": 0.00015220883534136547,
      "loss": 0.0048,
      "step": 123
    },
    {
      "epoch": 124.0,
      "grad_norm": 0.01739220879971981,
      "learning_rate": 0.00015180722891566266,
      "loss": 0.005,
      "step": 124
    },
    {
      "epoch": 125.0,
      "grad_norm": 0.007263750769197941,
      "learning_rate": 0.00015140562248995985,
      "loss": 0.0047,
      "step": 125
    },
    {
      "epoch": 126.0,
      "grad_norm": 0.04079929739236832,
      "learning_rate": 0.00015100401606425701,
      "loss": 0.0048,
      "step": 126
    },
    {
      "epoch": 127.0,
      "grad_norm": 0.009831869974732399,
      "learning_rate": 0.00015060240963855423,
      "loss": 0.0047,
      "step": 127
    },
    {
      "epoch": 128.0,
      "grad_norm": 0.026599382981657982,
      "learning_rate": 0.00015020080321285142,
      "loss": 0.0048,
      "step": 128
    },
    {
      "epoch": 129.0,
      "grad_norm": 0.018224092200398445,
      "learning_rate": 0.0001497991967871486,
      "loss": 0.0047,
      "step": 129
    },
    {
      "epoch": 130.0,
      "grad_norm": 0.02792244777083397,
      "learning_rate": 0.00014939759036144577,
      "loss": 0.0047,
      "step": 130
    },
    {
      "epoch": 131.0,
      "grad_norm": 0.024047773331403732,
      "learning_rate": 0.000148995983935743,
      "loss": 0.0048,
      "step": 131
    },
    {
      "epoch": 132.0,
      "grad_norm": 0.018245236948132515,
      "learning_rate": 0.00014859437751004018,
      "loss": 0.0047,
      "step": 132
    },
    {
      "epoch": 133.0,
      "grad_norm": 0.026581475511193275,
      "learning_rate": 0.00014819277108433737,
      "loss": 0.0048,
      "step": 133
    },
    {
      "epoch": 134.0,
      "grad_norm": 0.007153503131121397,
      "learning_rate": 0.00014779116465863453,
      "loss": 0.0047,
      "step": 134
    },
    {
      "epoch": 135.0,
      "grad_norm": 0.0323675237596035,
      "learning_rate": 0.00014738955823293172,
      "loss": 0.0048,
      "step": 135
    },
    {
      "epoch": 136.0,
      "grad_norm": 0.005924216937273741,
      "learning_rate": 0.00014698795180722893,
      "loss": 0.0046,
      "step": 136
    },
    {
      "epoch": 137.0,
      "grad_norm": 0.026371264830231667,
      "learning_rate": 0.00014658634538152612,
      "loss": 0.0048,
      "step": 137
    },
    {
      "epoch": 138.0,
      "grad_norm": 0.002094594296067953,
      "learning_rate": 0.00014618473895582328,
      "loss": 0.0047,
      "step": 138
    },
    {
      "epoch": 139.0,
      "grad_norm": 0.023835647851228714,
      "learning_rate": 0.00014578313253012047,
      "loss": 0.0048,
      "step": 139
    },
    {
      "epoch": 140.0,
      "grad_norm": 0.007189361844211817,
      "learning_rate": 0.0001453815261044177,
      "loss": 0.0047,
      "step": 140
    },
    {
      "epoch": 141.0,
      "grad_norm": 0.021903127431869507,
      "learning_rate": 0.00014497991967871488,
      "loss": 0.0049,
      "step": 141
    },
    {
      "epoch": 142.0,
      "grad_norm": 0.009862157516181469,
      "learning_rate": 0.00014457831325301204,
      "loss": 0.0047,
      "step": 142
    },
    {
      "epoch": 143.0,
      "grad_norm": 0.01557657215744257,
      "learning_rate": 0.00014417670682730923,
      "loss": 0.0047,
      "step": 143
    },
    {
      "epoch": 144.0,
      "grad_norm": 0.01178675051778555,
      "learning_rate": 0.00014377510040160642,
      "loss": 0.0048,
      "step": 144
    },
    {
      "epoch": 145.0,
      "grad_norm": 0.009945346042513847,
      "learning_rate": 0.00014337349397590364,
      "loss": 0.0047,
      "step": 145
    },
    {
      "epoch": 146.0,
      "grad_norm": 0.013681800104677677,
      "learning_rate": 0.0001429718875502008,
      "loss": 0.0048,
      "step": 146
    },
    {
      "epoch": 147.0,
      "grad_norm": 0.0072199697606265545,
      "learning_rate": 0.000142570281124498,
      "loss": 0.0047,
      "step": 147
    },
    {
      "epoch": 148.0,
      "grad_norm": 0.015607825480401516,
      "learning_rate": 0.00014216867469879518,
      "loss": 0.0047,
      "step": 148
    },
    {
      "epoch": 149.0,
      "grad_norm": 0.00990898534655571,
      "learning_rate": 0.0001417670682730924,
      "loss": 0.0047,
      "step": 149
    },
    {
      "epoch": 150.0,
      "grad_norm": 0.009972168132662773,
      "learning_rate": 0.00014136546184738956,
      "loss": 0.0047,
      "step": 150
    },
    {
      "epoch": 151.0,
      "grad_norm": 0.007324350066483021,
      "learning_rate": 0.00014096385542168674,
      "loss": 0.0047,
      "step": 151
    },
    {
      "epoch": 152.0,
      "grad_norm": 0.007397031411528587,
      "learning_rate": 0.00014056224899598393,
      "loss": 0.0047,
      "step": 152
    },
    {
      "epoch": 153.0,
      "grad_norm": 0.00632756482809782,
      "learning_rate": 0.00014016064257028115,
      "loss": 0.0046,
      "step": 153
    },
    {
      "epoch": 154.0,
      "grad_norm": 0.0021140226162970066,
      "learning_rate": 0.00013975903614457834,
      "loss": 0.0047,
      "step": 154
    },
    {
      "epoch": 155.0,
      "grad_norm": 0.0021610369440168142,
      "learning_rate": 0.0001393574297188755,
      "loss": 0.0047,
      "step": 155
    },
    {
      "epoch": 156.0,
      "grad_norm": 0.004180264193564653,
      "learning_rate": 0.0001389558232931727,
      "loss": 0.0048,
      "step": 156
    },
    {
      "epoch": 157.0,
      "grad_norm": 0.002081150421872735,
      "learning_rate": 0.00013855421686746988,
      "loss": 0.0047,
      "step": 157
    },
    {
      "epoch": 158.0,
      "grad_norm": 0.004214874934405088,
      "learning_rate": 0.0001381526104417671,
      "loss": 0.0048,
      "step": 158
    },
    {
      "epoch": 159.0,
      "grad_norm": 0.002199581591412425,
      "learning_rate": 0.00013775100401606426,
      "loss": 0.0047,
      "step": 159
    },
    {
      "epoch": 160.0,
      "grad_norm": 0.010448895394802094,
      "learning_rate": 0.00013734939759036145,
      "loss": 0.0047,
      "step": 160
    },
    {
      "epoch": 161.0,
      "grad_norm": 0.007638930808752775,
      "learning_rate": 0.00013694779116465864,
      "loss": 0.0047,
      "step": 161
    },
    {
      "epoch": 162.0,
      "grad_norm": 0.007661975454539061,
      "learning_rate": 0.00013654618473895585,
      "loss": 0.0047,
      "step": 162
    },
    {
      "epoch": 163.0,
      "grad_norm": 0.0014610282378271222,
      "learning_rate": 0.00013614457831325302,
      "loss": 0.0047,
      "step": 163
    },
    {
      "epoch": 164.0,
      "grad_norm": 0.01049406360834837,
      "learning_rate": 0.0001357429718875502,
      "loss": 0.0047,
      "step": 164
    },
    {
      "epoch": 165.0,
      "grad_norm": 0.0021653317380696535,
      "learning_rate": 0.0001353413654618474,
      "loss": 0.0047,
      "step": 165
    },
    {
      "epoch": 166.0,
      "grad_norm": 0.016612282022833824,
      "learning_rate": 0.00013493975903614458,
      "loss": 0.0047,
      "step": 166
    },
    {
      "epoch": 167.0,
      "grad_norm": 0.005656089633703232,
      "learning_rate": 0.00013453815261044177,
      "loss": 0.0048,
      "step": 167
    },
    {
      "epoch": 168.0,
      "grad_norm": 0.015557671897113323,
      "learning_rate": 0.00013413654618473896,
      "loss": 0.0046,
      "step": 168
    },
    {
      "epoch": 169.0,
      "grad_norm": 0.00793896708637476,
      "learning_rate": 0.00013373493975903615,
      "loss": 0.0047,
      "step": 169
    },
    {
      "epoch": 170.0,
      "grad_norm": 0.016965791583061218,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.0047,
      "step": 170
    },
    {
      "epoch": 171.0,
      "grad_norm": 0.010896142572164536,
      "learning_rate": 0.00013293172690763053,
      "loss": 0.0047,
      "step": 171
    },
    {
      "epoch": 172.0,
      "grad_norm": 0.014977889135479927,
      "learning_rate": 0.00013253012048192772,
      "loss": 0.0048,
      "step": 172
    },
    {
      "epoch": 173.0,
      "grad_norm": 0.007990415208041668,
      "learning_rate": 0.0001321285140562249,
      "loss": 0.0047,
      "step": 173
    },
    {
      "epoch": 174.0,
      "grad_norm": 0.01746082492172718,
      "learning_rate": 0.0001317269076305221,
      "loss": 0.0047,
      "step": 174
    },
    {
      "epoch": 175.0,
      "grad_norm": 0.011155808344483376,
      "learning_rate": 0.00013132530120481929,
      "loss": 0.0047,
      "step": 175
    },
    {
      "epoch": 176.0,
      "grad_norm": 0.020374253392219543,
      "learning_rate": 0.00013092369477911648,
      "loss": 0.0047,
      "step": 176
    },
    {
      "epoch": 177.0,
      "grad_norm": 0.027189314365386963,
      "learning_rate": 0.00013052208835341366,
      "loss": 0.0048,
      "step": 177
    },
    {
      "epoch": 178.0,
      "grad_norm": 0.0035853274166584015,
      "learning_rate": 0.00013012048192771085,
      "loss": 0.0046,
      "step": 178
    },
    {
      "epoch": 179.0,
      "grad_norm": 0.02057839184999466,
      "learning_rate": 0.00012971887550200804,
      "loss": 0.0047,
      "step": 179
    },
    {
      "epoch": 180.0,
      "grad_norm": 0.004144900944083929,
      "learning_rate": 0.00012931726907630523,
      "loss": 0.0047,
      "step": 180
    },
    {
      "epoch": 181.0,
      "grad_norm": 0.027396870777010918,
      "learning_rate": 0.00012891566265060242,
      "loss": 0.0048,
      "step": 181
    },
    {
      "epoch": 182.0,
      "grad_norm": 0.004257265478372574,
      "learning_rate": 0.0001285140562248996,
      "loss": 0.0048,
      "step": 182
    },
    {
      "epoch": 183.0,
      "grad_norm": 0.029999705031514168,
      "learning_rate": 0.0001281124497991968,
      "loss": 0.0048,
      "step": 183
    },
    {
      "epoch": 184.0,
      "grad_norm": 0.012998619116842747,
      "learning_rate": 0.00012771084337349396,
      "loss": 0.0047,
      "step": 184
    },
    {
      "epoch": 185.0,
      "grad_norm": 0.027277518063783646,
      "learning_rate": 0.00012730923694779118,
      "loss": 0.0048,
      "step": 185
    },
    {
      "epoch": 186.0,
      "grad_norm": 0.020436229184269905,
      "learning_rate": 0.00012690763052208837,
      "loss": 0.0047,
      "step": 186
    },
    {
      "epoch": 187.0,
      "grad_norm": 0.017912449315190315,
      "learning_rate": 0.00012650602409638556,
      "loss": 0.0048,
      "step": 187
    },
    {
      "epoch": 188.0,
      "grad_norm": 0.022624023258686066,
      "learning_rate": 0.00012610441767068272,
      "loss": 0.0047,
      "step": 188
    },
    {
      "epoch": 189.0,
      "grad_norm": 0.00817954819649458,
      "learning_rate": 0.00012570281124497994,
      "loss": 0.0047,
      "step": 189
    },
    {
      "epoch": 190.0,
      "grad_norm": 0.01837238110601902,
      "learning_rate": 0.00012530120481927712,
      "loss": 0.0047,
      "step": 190
    },
    {
      "epoch": 191.0,
      "grad_norm": 0.006303516216576099,
      "learning_rate": 0.0001248995983935743,
      "loss": 0.0047,
      "step": 191
    },
    {
      "epoch": 192.0,
      "grad_norm": 0.02228759415447712,
      "learning_rate": 0.00012449799196787148,
      "loss": 0.0047,
      "step": 192
    },
    {
      "epoch": 193.0,
      "grad_norm": 0.004886090289801359,
      "learning_rate": 0.0001240963855421687,
      "loss": 0.0047,
      "step": 193
    },
    {
      "epoch": 194.0,
      "grad_norm": 0.017957722768187523,
      "learning_rate": 0.00012369477911646588,
      "loss": 0.0048,
      "step": 194
    },
    {
      "epoch": 195.0,
      "grad_norm": 0.008203946985304356,
      "learning_rate": 0.00012329317269076307,
      "loss": 0.0047,
      "step": 195
    },
    {
      "epoch": 196.0,
      "grad_norm": 0.013346477411687374,
      "learning_rate": 0.00012289156626506023,
      "loss": 0.0047,
      "step": 196
    },
    {
      "epoch": 197.0,
      "grad_norm": 0.008646669797599316,
      "learning_rate": 0.00012248995983935742,
      "loss": 0.0048,
      "step": 197
    },
    {
      "epoch": 198.0,
      "grad_norm": 0.011337646283209324,
      "learning_rate": 0.00012208835341365464,
      "loss": 0.0047,
      "step": 198
    },
    {
      "epoch": 199.0,
      "grad_norm": 0.01342825498431921,
      "learning_rate": 0.00012168674698795181,
      "loss": 0.0047,
      "step": 199
    },
    {
      "epoch": 200.0,
      "grad_norm": 0.006050780415534973,
      "learning_rate": 0.000121285140562249,
      "loss": 0.0048,
      "step": 200
    },
    {
      "epoch": 201.0,
      "grad_norm": 0.01836731843650341,
      "learning_rate": 0.00012088353413654618,
      "loss": 0.0048,
      "step": 201
    },
    {
      "epoch": 202.0,
      "grad_norm": 0.008505699224770069,
      "learning_rate": 0.0001204819277108434,
      "loss": 0.0047,
      "step": 202
    },
    {
      "epoch": 203.0,
      "grad_norm": 0.008753190748393536,
      "learning_rate": 0.00012008032128514057,
      "loss": 0.0047,
      "step": 203
    },
    {
      "epoch": 204.0,
      "grad_norm": 0.009133870713412762,
      "learning_rate": 0.00011967871485943776,
      "loss": 0.0048,
      "step": 204
    },
    {
      "epoch": 205.0,
      "grad_norm": 0.0029124633874744177,
      "learning_rate": 0.00011927710843373494,
      "loss": 0.0048,
      "step": 205
    },
    {
      "epoch": 206.0,
      "grad_norm": 0.00847614649683237,
      "learning_rate": 0.00011887550200803212,
      "loss": 0.0047,
      "step": 206
    },
    {
      "epoch": 207.0,
      "grad_norm": 0.0028652322944253683,
      "learning_rate": 0.00011847389558232933,
      "loss": 0.0047,
      "step": 207
    },
    {
      "epoch": 208.0,
      "grad_norm": 0.009550940245389938,
      "learning_rate": 0.00011807228915662652,
      "loss": 0.0047,
      "step": 208
    },
    {
      "epoch": 209.0,
      "grad_norm": 0.008687314577400684,
      "learning_rate": 0.00011767068273092369,
      "loss": 0.0047,
      "step": 209
    },
    {
      "epoch": 210.0,
      "grad_norm": 0.004487687721848488,
      "learning_rate": 0.00011726907630522088,
      "loss": 0.0047,
      "step": 210
    },
    {
      "epoch": 211.0,
      "grad_norm": 0.009615003131330013,
      "learning_rate": 0.00011686746987951808,
      "loss": 0.0047,
      "step": 211
    },
    {
      "epoch": 212.0,
      "grad_norm": 0.001644686795771122,
      "learning_rate": 0.00011646586345381527,
      "loss": 0.0047,
      "step": 212
    },
    {
      "epoch": 213.0,
      "grad_norm": 0.012544874101877213,
      "learning_rate": 0.00011606425702811245,
      "loss": 0.0047,
      "step": 213
    },
    {
      "epoch": 214.0,
      "grad_norm": 0.009784480556845665,
      "learning_rate": 0.00011566265060240964,
      "loss": 0.0047,
      "step": 214
    },
    {
      "epoch": 215.0,
      "grad_norm": 0.008885402232408524,
      "learning_rate": 0.00011526104417670683,
      "loss": 0.0047,
      "step": 215
    },
    {
      "epoch": 216.0,
      "grad_norm": 0.015116319991648197,
      "learning_rate": 0.00011485943775100403,
      "loss": 0.0047,
      "step": 216
    },
    {
      "epoch": 217.0,
      "grad_norm": 0.0030038722325116396,
      "learning_rate": 0.0001144578313253012,
      "loss": 0.0047,
      "step": 217
    },
    {
      "epoch": 218.0,
      "grad_norm": 0.014432215131819248,
      "learning_rate": 0.0001140562248995984,
      "loss": 0.0047,
      "step": 218
    },
    {
      "epoch": 219.0,
      "grad_norm": 0.012555493041872978,
      "learning_rate": 0.00011365461847389558,
      "loss": 0.0047,
      "step": 219
    },
    {
      "epoch": 220.0,
      "grad_norm": 0.004350865725427866,
      "learning_rate": 0.00011325301204819279,
      "loss": 0.0047,
      "step": 220
    },
    {
      "epoch": 221.0,
      "grad_norm": 0.011415142565965652,
      "learning_rate": 0.00011285140562248996,
      "loss": 0.0047,
      "step": 221
    },
    {
      "epoch": 222.0,
      "grad_norm": 0.002292638411745429,
      "learning_rate": 0.00011244979919678715,
      "loss": 0.0047,
      "step": 222
    },
    {
      "epoch": 223.0,
      "grad_norm": 0.011944664642214775,
      "learning_rate": 0.00011204819277108434,
      "loss": 0.0047,
      "step": 223
    },
    {
      "epoch": 224.0,
      "grad_norm": 0.009497747756540775,
      "learning_rate": 0.00011164658634538152,
      "loss": 0.0047,
      "step": 224
    },
    {
      "epoch": 225.0,
      "grad_norm": 0.004834707360714674,
      "learning_rate": 0.00011124497991967872,
      "loss": 0.0047,
      "step": 225
    },
    {
      "epoch": 226.0,
      "grad_norm": 0.012152622453868389,
      "learning_rate": 0.00011084337349397591,
      "loss": 0.0048,
      "step": 226
    },
    {
      "epoch": 227.0,
      "grad_norm": 0.005508288741111755,
      "learning_rate": 0.0001104417670682731,
      "loss": 0.0047,
      "step": 227
    },
    {
      "epoch": 228.0,
      "grad_norm": 0.004662630148231983,
      "learning_rate": 0.00011004016064257027,
      "loss": 0.0047,
      "step": 228
    },
    {
      "epoch": 229.0,
      "grad_norm": 0.0046459161676466465,
      "learning_rate": 0.00010963855421686749,
      "loss": 0.0047,
      "step": 229
    },
    {
      "epoch": 230.0,
      "grad_norm": 0.003354718443006277,
      "learning_rate": 0.00010923694779116467,
      "loss": 0.0047,
      "step": 230
    },
    {
      "epoch": 231.0,
      "grad_norm": 0.005278999917209148,
      "learning_rate": 0.00010883534136546186,
      "loss": 0.0048,
      "step": 231
    },
    {
      "epoch": 232.0,
      "grad_norm": 0.006012341473251581,
      "learning_rate": 0.00010843373493975903,
      "loss": 0.0047,
      "step": 232
    },
    {
      "epoch": 233.0,
      "grad_norm": 0.0016911854036152363,
      "learning_rate": 0.00010803212851405625,
      "loss": 0.0047,
      "step": 233
    },
    {
      "epoch": 234.0,
      "grad_norm": 0.004478626884520054,
      "learning_rate": 0.00010763052208835342,
      "loss": 0.0047,
      "step": 234
    },
    {
      "epoch": 235.0,
      "grad_norm": 0.005508603993803263,
      "learning_rate": 0.00010722891566265061,
      "loss": 0.0047,
      "step": 235
    },
    {
      "epoch": 236.0,
      "grad_norm": 0.0032062928657978773,
      "learning_rate": 0.00010682730923694779,
      "loss": 0.0047,
      "step": 236
    },
    {
      "epoch": 237.0,
      "grad_norm": 0.006789966020733118,
      "learning_rate": 0.00010642570281124498,
      "loss": 0.0048,
      "step": 237
    },
    {
      "epoch": 238.0,
      "grad_norm": 0.005682968068867922,
      "learning_rate": 0.00010602409638554218,
      "loss": 0.0047,
      "step": 238
    },
    {
      "epoch": 239.0,
      "grad_norm": 0.00296304514631629,
      "learning_rate": 0.00010562248995983937,
      "loss": 0.0047,
      "step": 239
    },
    {
      "epoch": 240.0,
      "grad_norm": 0.006762088742107153,
      "learning_rate": 0.00010522088353413654,
      "loss": 0.0047,
      "step": 240
    },
    {
      "epoch": 241.0,
      "grad_norm": 0.0028762409929186106,
      "learning_rate": 0.00010481927710843373,
      "loss": 0.0047,
      "step": 241
    },
    {
      "epoch": 242.0,
      "grad_norm": 0.009846026077866554,
      "learning_rate": 0.00010441767068273094,
      "loss": 0.0047,
      "step": 242
    },
    {
      "epoch": 243.0,
      "grad_norm": 0.01071733795106411,
      "learning_rate": 0.00010401606425702813,
      "loss": 0.0047,
      "step": 243
    },
    {
      "epoch": 244.0,
      "grad_norm": 0.001721803448162973,
      "learning_rate": 0.0001036144578313253,
      "loss": 0.0047,
      "step": 244
    },
    {
      "epoch": 245.0,
      "grad_norm": 0.015101822093129158,
      "learning_rate": 0.00010321285140562249,
      "loss": 0.0047,
      "step": 245
    },
    {
      "epoch": 246.0,
      "grad_norm": 0.012098951265215874,
      "learning_rate": 0.00010281124497991968,
      "loss": 0.0047,
      "step": 246
    },
    {
      "epoch": 247.0,
      "grad_norm": 0.006853122264146805,
      "learning_rate": 0.00010240963855421688,
      "loss": 0.0047,
      "step": 247
    },
    {
      "epoch": 248.0,
      "grad_norm": 0.016420654952526093,
      "learning_rate": 0.00010200803212851406,
      "loss": 0.0047,
      "step": 248
    },
    {
      "epoch": 249.0,
      "grad_norm": 0.00433447165414691,
      "learning_rate": 0.00010160642570281125,
      "loss": 0.0047,
      "step": 249
    },
    {
      "epoch": 250.0,
      "grad_norm": 0.012108503840863705,
      "learning_rate": 0.00010120481927710844,
      "loss": 0.0047,
      "step": 250
    },
    {
      "epoch": 251.0,
      "grad_norm": 0.008616012521088123,
      "learning_rate": 0.00010080321285140564,
      "loss": 0.0047,
      "step": 251
    },
    {
      "epoch": 252.0,
      "grad_norm": 0.008652638643980026,
      "learning_rate": 0.00010040160642570282,
      "loss": 0.0047,
      "step": 252
    },
    {
      "epoch": 253.0,
      "grad_norm": 0.01692233793437481,
      "learning_rate": 0.0001,
      "loss": 0.0047,
      "step": 253
    },
    {
      "epoch": 254.0,
      "grad_norm": 0.006091867107897997,
      "learning_rate": 9.95983935742972e-05,
      "loss": 0.0047,
      "step": 254
    },
    {
      "epoch": 255.0,
      "grad_norm": 0.011222707107663155,
      "learning_rate": 9.919678714859438e-05,
      "loss": 0.0047,
      "step": 255
    },
    {
      "epoch": 256.0,
      "grad_norm": 0.011931284330785275,
      "learning_rate": 9.879518072289157e-05,
      "loss": 0.0047,
      "step": 256
    },
    {
      "epoch": 257.0,
      "grad_norm": 0.004332751967012882,
      "learning_rate": 9.839357429718876e-05,
      "loss": 0.0048,
      "step": 257
    },
    {
      "epoch": 258.0,
      "grad_norm": 0.013347852043807507,
      "learning_rate": 9.799196787148595e-05,
      "loss": 0.0047,
      "step": 258
    },
    {
      "epoch": 259.0,
      "grad_norm": 0.006608230993151665,
      "learning_rate": 9.759036144578314e-05,
      "loss": 0.0047,
      "step": 259
    },
    {
      "epoch": 260.0,
      "grad_norm": 0.004314239602535963,
      "learning_rate": 9.718875502008033e-05,
      "loss": 0.0048,
      "step": 260
    },
    {
      "epoch": 261.0,
      "grad_norm": 0.0060504162684082985,
      "learning_rate": 9.678714859437752e-05,
      "loss": 0.0048,
      "step": 261
    },
    {
      "epoch": 262.0,
      "grad_norm": 0.0033889045007526875,
      "learning_rate": 9.638554216867471e-05,
      "loss": 0.0047,
      "step": 262
    },
    {
      "epoch": 263.0,
      "grad_norm": 0.008607766591012478,
      "learning_rate": 9.598393574297188e-05,
      "loss": 0.0048,
      "step": 263
    },
    {
      "epoch": 264.0,
      "grad_norm": 0.001759338192641735,
      "learning_rate": 9.558232931726909e-05,
      "loss": 0.0047,
      "step": 264
    },
    {
      "epoch": 265.0,
      "grad_norm": 0.008092672564089298,
      "learning_rate": 9.518072289156626e-05,
      "loss": 0.0047,
      "step": 265
    },
    {
      "epoch": 266.0,
      "grad_norm": 0.0017322164494544268,
      "learning_rate": 9.477911646586346e-05,
      "loss": 0.0047,
      "step": 266
    },
    {
      "epoch": 267.0,
      "grad_norm": 0.0017360730562359095,
      "learning_rate": 9.437751004016064e-05,
      "loss": 0.0047,
      "step": 267
    },
    {
      "epoch": 268.0,
      "grad_norm": 0.0017291605472564697,
      "learning_rate": 9.397590361445784e-05,
      "loss": 0.0047,
      "step": 268
    },
    {
      "epoch": 269.0,
      "grad_norm": 0.0016873552231118083,
      "learning_rate": 9.357429718875502e-05,
      "loss": 0.0047,
      "step": 269
    },
    {
      "epoch": 270.0,
      "grad_norm": 0.007920237258076668,
      "learning_rate": 9.317269076305222e-05,
      "loss": 0.0047,
      "step": 270
    },
    {
      "epoch": 271.0,
      "grad_norm": 0.0036651096306741238,
      "learning_rate": 9.27710843373494e-05,
      "loss": 0.0048,
      "step": 271
    },
    {
      "epoch": 272.0,
      "grad_norm": 0.01999688521027565,
      "learning_rate": 9.23694779116466e-05,
      "loss": 0.0047,
      "step": 272
    },
    {
      "epoch": 273.0,
      "grad_norm": 0.007801530417054892,
      "learning_rate": 9.196787148594378e-05,
      "loss": 0.0047,
      "step": 273
    },
    {
      "epoch": 274.0,
      "grad_norm": 0.012862344272434711,
      "learning_rate": 9.156626506024096e-05,
      "loss": 0.0048,
      "step": 274
    },
    {
      "epoch": 275.0,
      "grad_norm": 0.0016800492303445935,
      "learning_rate": 9.116465863453815e-05,
      "loss": 0.0047,
      "step": 275
    },
    {
      "epoch": 276.0,
      "grad_norm": 0.019683390855789185,
      "learning_rate": 9.076305220883534e-05,
      "loss": 0.0047,
      "step": 276
    },
    {
      "epoch": 277.0,
      "grad_norm": 0.0017172418301925063,
      "learning_rate": 9.036144578313253e-05,
      "loss": 0.0047,
      "step": 277
    },
    {
      "epoch": 278.0,
      "grad_norm": 0.030774248763918877,
      "learning_rate": 8.995983935742972e-05,
      "loss": 0.0049,
      "step": 278
    },
    {
      "epoch": 279.0,
      "grad_norm": 0.0035772966220974922,
      "learning_rate": 8.955823293172691e-05,
      "loss": 0.0048,
      "step": 279
    },
    {
      "epoch": 280.0,
      "grad_norm": 0.03694264590740204,
      "learning_rate": 8.91566265060241e-05,
      "loss": 0.0048,
      "step": 280
    },
    {
      "epoch": 281.0,
      "grad_norm": 0.0031574727036058903,
      "learning_rate": 8.875502008032129e-05,
      "loss": 0.0045,
      "step": 281
    },
    {
      "epoch": 282.0,
      "grad_norm": 0.03539184108376503,
      "learning_rate": 8.835341365461848e-05,
      "loss": 0.005,
      "step": 282
    },
    {
      "epoch": 283.0,
      "grad_norm": 0.010516048409044743,
      "learning_rate": 8.795180722891567e-05,
      "loss": 0.0047,
      "step": 283
    },
    {
      "epoch": 284.0,
      "grad_norm": 0.010487216524779797,
      "learning_rate": 8.755020080321286e-05,
      "loss": 0.0047,
      "step": 284
    },
    {
      "epoch": 285.0,
      "grad_norm": 0.010434217751026154,
      "learning_rate": 8.714859437751005e-05,
      "loss": 0.0047,
      "step": 285
    },
    {
      "epoch": 286.0,
      "grad_norm": 0.007612716872245073,
      "learning_rate": 8.674698795180724e-05,
      "loss": 0.0047,
      "step": 286
    },
    {
      "epoch": 287.0,
      "grad_norm": 0.025456130504608154,
      "learning_rate": 8.634538152610442e-05,
      "loss": 0.0048,
      "step": 287
    },
    {
      "epoch": 288.0,
      "grad_norm": 0.010371106676757336,
      "learning_rate": 8.594377510040161e-05,
      "loss": 0.0047,
      "step": 288
    },
    {
      "epoch": 289.0,
      "grad_norm": 0.028378885239362717,
      "learning_rate": 8.55421686746988e-05,
      "loss": 0.0048,
      "step": 289
    },
    {
      "epoch": 290.0,
      "grad_norm": 0.007674542721360922,
      "learning_rate": 8.514056224899599e-05,
      "loss": 0.0047,
      "step": 290
    },
    {
      "epoch": 291.0,
      "grad_norm": 0.025507695972919464,
      "learning_rate": 8.473895582329318e-05,
      "loss": 0.0048,
      "step": 291
    },
    {
      "epoch": 292.0,
      "grad_norm": 0.0009824644075706601,
      "learning_rate": 8.433734939759037e-05,
      "loss": 0.0049,
      "step": 292
    },
    {
      "epoch": 293.0,
      "grad_norm": 0.03781874477863312,
      "learning_rate": 8.393574297188756e-05,
      "loss": 0.0046,
      "step": 293
    },
    {
      "epoch": 294.0,
      "grad_norm": 0.007710340432822704,
      "learning_rate": 8.353413654618474e-05,
      "loss": 0.0047,
      "step": 294
    },
    {
      "epoch": 295.0,
      "grad_norm": 0.03508257865905762,
      "learning_rate": 8.313253012048194e-05,
      "loss": 0.005,
      "step": 295
    },
    {
      "epoch": 296.0,
      "grad_norm": 0.007686258759349585,
      "learning_rate": 8.273092369477911e-05,
      "loss": 0.0047,
      "step": 296
    },
    {
      "epoch": 297.0,
      "grad_norm": 0.04559750109910965,
      "learning_rate": 8.232931726907632e-05,
      "loss": 0.0048,
      "step": 297
    },
    {
      "epoch": 298.0,
      "grad_norm": 0.010225856676697731,
      "learning_rate": 8.192771084337349e-05,
      "loss": 0.0047,
      "step": 298
    },
    {
      "epoch": 299.0,
      "grad_norm": 0.02517218515276909,
      "learning_rate": 8.15261044176707e-05,
      "loss": 0.0048,
      "step": 299
    },
    {
      "epoch": 300.0,
      "grad_norm": 0.025102809071540833,
      "learning_rate": 8.112449799196787e-05,
      "loss": 0.0048,
      "step": 300
    },
    {
      "epoch": 301.0,
      "grad_norm": 0.0025317783001810312,
      "learning_rate": 8.072289156626507e-05,
      "loss": 0.0045,
      "step": 301
    },
    {
      "epoch": 302.0,
      "grad_norm": 0.0447322279214859,
      "learning_rate": 8.032128514056225e-05,
      "loss": 0.0048,
      "step": 302
    },
    {
      "epoch": 303.0,
      "grad_norm": 0.0074921357445418835,
      "learning_rate": 7.991967871485944e-05,
      "loss": 0.0047,
      "step": 303
    },
    {
      "epoch": 304.0,
      "grad_norm": 0.02473229542374611,
      "learning_rate": 7.951807228915663e-05,
      "loss": 0.0048,
      "step": 304
    },
    {
      "epoch": 305.0,
      "grad_norm": 0.007424222771078348,
      "learning_rate": 7.911646586345382e-05,
      "loss": 0.0047,
      "step": 305
    },
    {
      "epoch": 306.0,
      "grad_norm": 0.027306661009788513,
      "learning_rate": 7.8714859437751e-05,
      "loss": 0.0048,
      "step": 306
    },
    {
      "epoch": 307.0,
      "grad_norm": 0.007345912978053093,
      "learning_rate": 7.83132530120482e-05,
      "loss": 0.0047,
      "step": 307
    },
    {
      "epoch": 308.0,
      "grad_norm": 0.02444930374622345,
      "learning_rate": 7.791164658634539e-05,
      "loss": 0.0048,
      "step": 308
    },
    {
      "epoch": 309.0,
      "grad_norm": 0.02708006091415882,
      "learning_rate": 7.751004016064257e-05,
      "loss": 0.0048,
      "step": 309
    },
    {
      "epoch": 310.0,
      "grad_norm": 0.007334452122449875,
      "learning_rate": 7.710843373493976e-05,
      "loss": 0.0047,
      "step": 310
    },
    {
      "epoch": 311.0,
      "grad_norm": 0.009786078706383705,
      "learning_rate": 7.670682730923695e-05,
      "loss": 0.0047,
      "step": 311
    },
    {
      "epoch": 312.0,
      "grad_norm": 0.007255980744957924,
      "learning_rate": 7.630522088353414e-05,
      "loss": 0.0047,
      "step": 312
    },
    {
      "epoch": 313.0,
      "grad_norm": 0.014863966032862663,
      "learning_rate": 7.590361445783133e-05,
      "loss": 0.0045,
      "step": 313
    },
    {
      "epoch": 314.0,
      "grad_norm": 0.009730237536132336,
      "learning_rate": 7.550200803212851e-05,
      "loss": 0.0047,
      "step": 314
    },
    {
      "epoch": 315.0,
      "grad_norm": 0.00974965188652277,
      "learning_rate": 7.510040160642571e-05,
      "loss": 0.0047,
      "step": 315
    },
    {
      "epoch": 316.0,
      "grad_norm": 0.0008471576729789376,
      "learning_rate": 7.469879518072289e-05,
      "loss": 0.0049,
      "step": 316
    },
    {
      "epoch": 317.0,
      "grad_norm": 0.007247288711369038,
      "learning_rate": 7.429718875502009e-05,
      "loss": 0.0047,
      "step": 317
    },
    {
      "epoch": 318.0,
      "grad_norm": 0.00724770175293088,
      "learning_rate": 7.389558232931726e-05,
      "loss": 0.0047,
      "step": 318
    },
    {
      "epoch": 319.0,
      "grad_norm": 0.009726744145154953,
      "learning_rate": 7.349397590361447e-05,
      "loss": 0.0047,
      "step": 319
    },
    {
      "epoch": 320.0,
      "grad_norm": 0.002370405476540327,
      "learning_rate": 7.309236947791164e-05,
      "loss": 0.0045,
      "step": 320
    },
    {
      "epoch": 321.0,
      "grad_norm": 0.007258490659296513,
      "learning_rate": 7.269076305220885e-05,
      "loss": 0.0047,
      "step": 321
    },
    {
      "epoch": 322.0,
      "grad_norm": 0.009763582609593868,
      "learning_rate": 7.228915662650602e-05,
      "loss": 0.0047,
      "step": 322
    },
    {
      "epoch": 323.0,
      "grad_norm": 0.0072934250347316265,
      "learning_rate": 7.188755020080321e-05,
      "loss": 0.0047,
      "step": 323
    },
    {
      "epoch": 324.0,
      "grad_norm": 0.007325456012040377,
      "learning_rate": 7.14859437751004e-05,
      "loss": 0.0047,
      "step": 324
    },
    {
      "epoch": 325.0,
      "grad_norm": 0.02716483175754547,
      "learning_rate": 7.108433734939759e-05,
      "loss": 0.0048,
      "step": 325
    },
    {
      "epoch": 326.0,
      "grad_norm": 0.007358341012150049,
      "learning_rate": 7.068273092369478e-05,
      "loss": 0.0047,
      "step": 326
    },
    {
      "epoch": 327.0,
      "grad_norm": 0.04177234321832657,
      "learning_rate": 7.028112449799197e-05,
      "loss": 0.0048,
      "step": 327
    },
    {
      "epoch": 328.0,
      "grad_norm": 0.027355682104825974,
      "learning_rate": 6.987951807228917e-05,
      "loss": 0.0048,
      "step": 328
    },
    {
      "epoch": 329.0,
      "grad_norm": 0.02739114686846733,
      "learning_rate": 6.947791164658635e-05,
      "loss": 0.0048,
      "step": 329
    },
    {
      "epoch": 330.0,
      "grad_norm": 0.024521242827177048,
      "learning_rate": 6.907630522088355e-05,
      "loss": 0.0048,
      "step": 330
    },
    {
      "epoch": 331.0,
      "grad_norm": 0.007352576591074467,
      "learning_rate": 6.867469879518072e-05,
      "loss": 0.0047,
      "step": 331
    },
    {
      "epoch": 332.0,
      "grad_norm": 0.0023073432967066765,
      "learning_rate": 6.827309236947793e-05,
      "loss": 0.0045,
      "step": 332
    },
    {
      "epoch": 333.0,
      "grad_norm": 0.009870602749288082,
      "learning_rate": 6.78714859437751e-05,
      "loss": 0.0047,
      "step": 333
    },
    {
      "epoch": 334.0,
      "grad_norm": 0.00988033413887024,
      "learning_rate": 6.746987951807229e-05,
      "loss": 0.0047,
      "step": 334
    },
    {
      "epoch": 335.0,
      "grad_norm": 0.024536525830626488,
      "learning_rate": 6.706827309236948e-05,
      "loss": 0.0048,
      "step": 335
    },
    {
      "epoch": 336.0,
      "grad_norm": 0.009867388755083084,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.0047,
      "step": 336
    },
    {
      "epoch": 337.0,
      "grad_norm": 0.00988020095974207,
      "learning_rate": 6.626506024096386e-05,
      "loss": 0.0047,
      "step": 337
    },
    {
      "epoch": 338.0,
      "grad_norm": 0.007355022244155407,
      "learning_rate": 6.586345381526105e-05,
      "loss": 0.0047,
      "step": 338
    },
    {
      "epoch": 339.0,
      "grad_norm": 0.007368543650954962,
      "learning_rate": 6.546184738955824e-05,
      "loss": 0.0047,
      "step": 339
    },
    {
      "epoch": 340.0,
      "grad_norm": 0.0073932805098593235,
      "learning_rate": 6.506024096385543e-05,
      "loss": 0.0047,
      "step": 340
    },
    {
      "epoch": 341.0,
      "grad_norm": 0.027419747784733772,
      "learning_rate": 6.465863453815262e-05,
      "loss": 0.0048,
      "step": 341
    },
    {
      "epoch": 342.0,
      "grad_norm": 0.009928101673722267,
      "learning_rate": 6.42570281124498e-05,
      "loss": 0.0047,
      "step": 342
    },
    {
      "epoch": 343.0,
      "grad_norm": 0.03391006961464882,
      "learning_rate": 6.385542168674698e-05,
      "loss": 0.005,
      "step": 343
    },
    {
      "epoch": 344.0,
      "grad_norm": 0.024691926315426826,
      "learning_rate": 6.345381526104418e-05,
      "loss": 0.0048,
      "step": 344
    },
    {
      "epoch": 345.0,
      "grad_norm": 0.036551907658576965,
      "learning_rate": 6.305220883534136e-05,
      "loss": 0.0046,
      "step": 345
    },
    {
      "epoch": 346.0,
      "grad_norm": 0.027336876839399338,
      "learning_rate": 6.265060240963856e-05,
      "loss": 0.0048,
      "step": 346
    },
    {
      "epoch": 347.0,
      "grad_norm": 0.016788320615887642,
      "learning_rate": 6.224899598393574e-05,
      "loss": 0.005,
      "step": 347
    },
    {
      "epoch": 348.0,
      "grad_norm": 0.0243778508156538,
      "learning_rate": 6.184738955823294e-05,
      "loss": 0.0048,
      "step": 348
    },
    {
      "epoch": 349.0,
      "grad_norm": 0.007300014141947031,
      "learning_rate": 6.144578313253012e-05,
      "loss": 0.0047,
      "step": 349
    },
    {
      "epoch": 350.0,
      "grad_norm": 0.01766958087682724,
      "learning_rate": 6.104417670682732e-05,
      "loss": 0.005,
      "step": 350
    },
    {
      "epoch": 351.0,
      "grad_norm": 0.026927420869469643,
      "learning_rate": 6.06425702811245e-05,
      "loss": 0.0048,
      "step": 351
    },
    {
      "epoch": 352.0,
      "grad_norm": 0.016552148386836052,
      "learning_rate": 6.02409638554217e-05,
      "loss": 0.005,
      "step": 352
    },
    {
      "epoch": 353.0,
      "grad_norm": 0.02408471331000328,
      "learning_rate": 5.983935742971888e-05,
      "loss": 0.0048,
      "step": 353
    },
    {
      "epoch": 354.0,
      "grad_norm": 0.0007216089288704097,
      "learning_rate": 5.943775100401606e-05,
      "loss": 0.0049,
      "step": 354
    },
    {
      "epoch": 355.0,
      "grad_norm": 0.009755726903676987,
      "learning_rate": 5.903614457831326e-05,
      "loss": 0.0047,
      "step": 355
    },
    {
      "epoch": 356.0,
      "grad_norm": 0.026764798909425735,
      "learning_rate": 5.863453815261044e-05,
      "loss": 0.0048,
      "step": 356
    },
    {
      "epoch": 357.0,
      "grad_norm": 0.007179186213761568,
      "learning_rate": 5.823293172690764e-05,
      "loss": 0.0047,
      "step": 357
    },
    {
      "epoch": 358.0,
      "grad_norm": 0.023883482441306114,
      "learning_rate": 5.783132530120482e-05,
      "loss": 0.0048,
      "step": 358
    },
    {
      "epoch": 359.0,
      "grad_norm": 0.007125131320208311,
      "learning_rate": 5.7429718875502015e-05,
      "loss": 0.0047,
      "step": 359
    },
    {
      "epoch": 360.0,
      "grad_norm": 0.026448102667927742,
      "learning_rate": 5.70281124497992e-05,
      "loss": 0.0048,
      "step": 360
    },
    {
      "epoch": 361.0,
      "grad_norm": 0.009603764861822128,
      "learning_rate": 5.6626506024096394e-05,
      "loss": 0.0047,
      "step": 361
    },
    {
      "epoch": 362.0,
      "grad_norm": 0.023749757558107376,
      "learning_rate": 5.6224899598393576e-05,
      "loss": 0.0048,
      "step": 362
    },
    {
      "epoch": 363.0,
      "grad_norm": 0.007082722615450621,
      "learning_rate": 5.582329317269076e-05,
      "loss": 0.0047,
      "step": 363
    },
    {
      "epoch": 364.0,
      "grad_norm": 0.0006819483824074268,
      "learning_rate": 5.5421686746987955e-05,
      "loss": 0.0049,
      "step": 364
    },
    {
      "epoch": 365.0,
      "grad_norm": 0.026299143210053444,
      "learning_rate": 5.502008032128514e-05,
      "loss": 0.0048,
      "step": 365
    },
    {
      "epoch": 366.0,
      "grad_norm": 0.007077427115291357,
      "learning_rate": 5.461847389558233e-05,
      "loss": 0.0047,
      "step": 366
    },
    {
      "epoch": 367.0,
      "grad_norm": 0.0070725963450968266,
      "learning_rate": 5.4216867469879516e-05,
      "loss": 0.0047,
      "step": 367
    },
    {
      "epoch": 368.0,
      "grad_norm": 0.007042643614113331,
      "learning_rate": 5.381526104417671e-05,
      "loss": 0.0047,
      "step": 368
    },
    {
      "epoch": 369.0,
      "grad_norm": 0.009521468542516232,
      "learning_rate": 5.3413654618473894e-05,
      "loss": 0.0047,
      "step": 369
    },
    {
      "epoch": 370.0,
      "grad_norm": 0.0022259766701608896,
      "learning_rate": 5.301204819277109e-05,
      "loss": 0.0045,
      "step": 370
    },
    {
      "epoch": 371.0,
      "grad_norm": 0.023667145520448685,
      "learning_rate": 5.261044176706827e-05,
      "loss": 0.0048,
      "step": 371
    },
    {
      "epoch": 372.0,
      "grad_norm": 0.009567582048475742,
      "learning_rate": 5.220883534136547e-05,
      "loss": 0.0047,
      "step": 372
    },
    {
      "epoch": 373.0,
      "grad_norm": 0.00958697684109211,
      "learning_rate": 5.180722891566265e-05,
      "loss": 0.0047,
      "step": 373
    },
    {
      "epoch": 374.0,
      "grad_norm": 0.00959386583417654,
      "learning_rate": 5.140562248995984e-05,
      "loss": 0.0047,
      "step": 374
    },
    {
      "epoch": 375.0,
      "grad_norm": 0.00957813672721386,
      "learning_rate": 5.100401606425703e-05,
      "loss": 0.0047,
      "step": 375
    },
    {
      "epoch": 376.0,
      "grad_norm": 0.016183258965611458,
      "learning_rate": 5.060240963855422e-05,
      "loss": 0.005,
      "step": 376
    },
    {
      "epoch": 377.0,
      "grad_norm": 0.007078688126057386,
      "learning_rate": 5.020080321285141e-05,
      "loss": 0.0047,
      "step": 377
    },
    {
      "epoch": 378.0,
      "grad_norm": 0.007077342830598354,
      "learning_rate": 4.97991967871486e-05,
      "loss": 0.0047,
      "step": 378
    },
    {
      "epoch": 379.0,
      "grad_norm": 0.002201406517997384,
      "learning_rate": 4.9397590361445786e-05,
      "loss": 0.0045,
      "step": 379
    },
    {
      "epoch": 380.0,
      "grad_norm": 0.009655999019742012,
      "learning_rate": 4.8995983935742975e-05,
      "loss": 0.0047,
      "step": 380
    },
    {
      "epoch": 381.0,
      "grad_norm": 0.017296917736530304,
      "learning_rate": 4.8594377510040165e-05,
      "loss": 0.005,
      "step": 381
    },
    {
      "epoch": 382.0,
      "grad_norm": 0.007125664968043566,
      "learning_rate": 4.8192771084337354e-05,
      "loss": 0.0047,
      "step": 382
    },
    {
      "epoch": 383.0,
      "grad_norm": 0.0238560251891613,
      "learning_rate": 4.779116465863454e-05,
      "loss": 0.0048,
      "step": 383
    },
    {
      "epoch": 384.0,
      "grad_norm": 0.009672388434410095,
      "learning_rate": 4.738955823293173e-05,
      "loss": 0.0047,
      "step": 384
    },
    {
      "epoch": 385.0,
      "grad_norm": 0.0022134389728307724,
      "learning_rate": 4.698795180722892e-05,
      "loss": 0.0045,
      "step": 385
    },
    {
      "epoch": 386.0,
      "grad_norm": 0.00969780795276165,
      "learning_rate": 4.658634538152611e-05,
      "loss": 0.0047,
      "step": 386
    },
    {
      "epoch": 387.0,
      "grad_norm": 0.007160300388932228,
      "learning_rate": 4.61847389558233e-05,
      "loss": 0.0047,
      "step": 387
    },
    {
      "epoch": 388.0,
      "grad_norm": 0.007146279327571392,
      "learning_rate": 4.578313253012048e-05,
      "loss": 0.0047,
      "step": 388
    },
    {
      "epoch": 389.0,
      "grad_norm": 0.002192781073972583,
      "learning_rate": 4.538152610441767e-05,
      "loss": 0.0045,
      "step": 389
    },
    {
      "epoch": 390.0,
      "grad_norm": 0.009754459373652935,
      "learning_rate": 4.497991967871486e-05,
      "loss": 0.0047,
      "step": 390
    },
    {
      "epoch": 391.0,
      "grad_norm": 0.00970425084233284,
      "learning_rate": 4.457831325301205e-05,
      "loss": 0.0047,
      "step": 391
    },
    {
      "epoch": 392.0,
      "grad_norm": 0.007177860010415316,
      "learning_rate": 4.417670682730924e-05,
      "loss": 0.0047,
      "step": 392
    },
    {
      "epoch": 393.0,
      "grad_norm": 0.024075862020254135,
      "learning_rate": 4.377510040160643e-05,
      "loss": 0.0048,
      "step": 393
    },
    {
      "epoch": 394.0,
      "grad_norm": 0.009754209779202938,
      "learning_rate": 4.337349397590362e-05,
      "loss": 0.0047,
      "step": 394
    },
    {
      "epoch": 395.0,
      "grad_norm": 0.009791336953639984,
      "learning_rate": 4.297188755020081e-05,
      "loss": 0.0047,
      "step": 395
    },
    {
      "epoch": 396.0,
      "grad_norm": 0.009780234657227993,
      "learning_rate": 4.2570281124497996e-05,
      "loss": 0.0047,
      "step": 396
    },
    {
      "epoch": 397.0,
      "grad_norm": 0.007180201821029186,
      "learning_rate": 4.2168674698795186e-05,
      "loss": 0.0047,
      "step": 397
    },
    {
      "epoch": 398.0,
      "grad_norm": 0.007196042221039534,
      "learning_rate": 4.176706827309237e-05,
      "loss": 0.0047,
      "step": 398
    },
    {
      "epoch": 399.0,
      "grad_norm": 0.00722030783072114,
      "learning_rate": 4.136546184738956e-05,
      "loss": 0.0047,
      "step": 399
    },
    {
      "epoch": 400.0,
      "grad_norm": 0.009809617884457111,
      "learning_rate": 4.0963855421686746e-05,
      "loss": 0.0047,
      "step": 400
    },
    {
      "epoch": 401.0,
      "grad_norm": 0.007194210775196552,
      "learning_rate": 4.0562248995983936e-05,
      "loss": 0.0047,
      "step": 401
    },
    {
      "epoch": 402.0,
      "grad_norm": 0.00984671525657177,
      "learning_rate": 4.0160642570281125e-05,
      "loss": 0.0047,
      "step": 402
    },
    {
      "epoch": 403.0,
      "grad_norm": 0.00983081478625536,
      "learning_rate": 3.9759036144578314e-05,
      "loss": 0.0047,
      "step": 403
    },
    {
      "epoch": 404.0,
      "grad_norm": 0.007195820100605488,
      "learning_rate": 3.93574297188755e-05,
      "loss": 0.0047,
      "step": 404
    },
    {
      "epoch": 405.0,
      "grad_norm": 0.007182563189417124,
      "learning_rate": 3.895582329317269e-05,
      "loss": 0.0047,
      "step": 405
    },
    {
      "epoch": 406.0,
      "grad_norm": 0.007222824264317751,
      "learning_rate": 3.855421686746988e-05,
      "loss": 0.0047,
      "step": 406
    },
    {
      "epoch": 407.0,
      "grad_norm": 0.007211462128907442,
      "learning_rate": 3.815261044176707e-05,
      "loss": 0.0047,
      "step": 407
    },
    {
      "epoch": 408.0,
      "grad_norm": 0.009921679273247719,
      "learning_rate": 3.7751004016064253e-05,
      "loss": 0.0047,
      "step": 408
    },
    {
      "epoch": 409.0,
      "grad_norm": 0.02687370777130127,
      "learning_rate": 3.734939759036144e-05,
      "loss": 0.0048,
      "step": 409
    },
    {
      "epoch": 410.0,
      "grad_norm": 0.009853348135948181,
      "learning_rate": 3.694779116465863e-05,
      "loss": 0.0047,
      "step": 410
    },
    {
      "epoch": 411.0,
      "grad_norm": 0.024240443482995033,
      "learning_rate": 3.654618473895582e-05,
      "loss": 0.0048,
      "step": 411
    },
    {
      "epoch": 412.0,
      "grad_norm": 0.04131508618593216,
      "learning_rate": 3.614457831325301e-05,
      "loss": 0.0048,
      "step": 412
    },
    {
      "epoch": 413.0,
      "grad_norm": 0.009910643100738525,
      "learning_rate": 3.57429718875502e-05,
      "loss": 0.0047,
      "step": 413
    },
    {
      "epoch": 414.0,
      "grad_norm": 0.03583821654319763,
      "learning_rate": 3.534136546184739e-05,
      "loss": 0.0046,
      "step": 414
    },
    {
      "epoch": 415.0,
      "grad_norm": 0.017576098442077637,
      "learning_rate": 3.4939759036144585e-05,
      "loss": 0.005,
      "step": 415
    },
    {
      "epoch": 416.0,
      "grad_norm": 0.01653764583170414,
      "learning_rate": 3.4538152610441774e-05,
      "loss": 0.005,
      "step": 416
    },
    {
      "epoch": 417.0,
      "grad_norm": 0.024247560650110245,
      "learning_rate": 3.413654618473896e-05,
      "loss": 0.0048,
      "step": 417
    },
    {
      "epoch": 418.0,
      "grad_norm": 0.007164428010582924,
      "learning_rate": 3.3734939759036146e-05,
      "loss": 0.0047,
      "step": 418
    },
    {
      "epoch": 419.0,
      "grad_norm": 0.0098257539793849,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.0047,
      "step": 419
    },
    {
      "epoch": 420.0,
      "grad_norm": 0.017520597204566002,
      "learning_rate": 3.2931726907630524e-05,
      "loss": 0.005,
      "step": 420
    },
    {
      "epoch": 421.0,
      "grad_norm": 0.018933523446321487,
      "learning_rate": 3.253012048192771e-05,
      "loss": 0.0045,
      "step": 421
    },
    {
      "epoch": 422.0,
      "grad_norm": 0.007097120396792889,
      "learning_rate": 3.21285140562249e-05,
      "loss": 0.0047,
      "step": 422
    },
    {
      "epoch": 423.0,
      "grad_norm": 0.024037552997469902,
      "learning_rate": 3.172690763052209e-05,
      "loss": 0.0048,
      "step": 423
    },
    {
      "epoch": 424.0,
      "grad_norm": 0.024066118523478508,
      "learning_rate": 3.132530120481928e-05,
      "loss": 0.0048,
      "step": 424
    },
    {
      "epoch": 425.0,
      "grad_norm": 0.0006517537985928357,
      "learning_rate": 3.092369477911647e-05,
      "loss": 0.0049,
      "step": 425
    },
    {
      "epoch": 426.0,
      "grad_norm": 0.034052085131406784,
      "learning_rate": 3.052208835341366e-05,
      "loss": 0.005,
      "step": 426
    },
    {
      "epoch": 427.0,
      "grad_norm": 0.017422359436750412,
      "learning_rate": 3.012048192771085e-05,
      "loss": 0.005,
      "step": 427
    },
    {
      "epoch": 428.0,
      "grad_norm": 0.009725161828100681,
      "learning_rate": 2.971887550200803e-05,
      "loss": 0.0047,
      "step": 428
    },
    {
      "epoch": 429.0,
      "grad_norm": 0.023919757455587387,
      "learning_rate": 2.931726907630522e-05,
      "loss": 0.0048,
      "step": 429
    },
    {
      "epoch": 430.0,
      "grad_norm": 0.04076967015862465,
      "learning_rate": 2.891566265060241e-05,
      "loss": 0.0048,
      "step": 430
    },
    {
      "epoch": 431.0,
      "grad_norm": 0.002202529925853014,
      "learning_rate": 2.85140562248996e-05,
      "loss": 0.0045,
      "step": 431
    },
    {
      "epoch": 432.0,
      "grad_norm": 0.026460327208042145,
      "learning_rate": 2.8112449799196788e-05,
      "loss": 0.0048,
      "step": 432
    },
    {
      "epoch": 433.0,
      "grad_norm": 0.026399288326501846,
      "learning_rate": 2.7710843373493977e-05,
      "loss": 0.0048,
      "step": 433
    },
    {
      "epoch": 434.0,
      "grad_norm": 0.007069241255521774,
      "learning_rate": 2.7309236947791167e-05,
      "loss": 0.0047,
      "step": 434
    },
    {
      "epoch": 435.0,
      "grad_norm": 0.023836608976125717,
      "learning_rate": 2.6907630522088356e-05,
      "loss": 0.0048,
      "step": 435
    },
    {
      "epoch": 436.0,
      "grad_norm": 0.007101175840944052,
      "learning_rate": 2.6506024096385545e-05,
      "loss": 0.0047,
      "step": 436
    },
    {
      "epoch": 437.0,
      "grad_norm": 0.009633008390665054,
      "learning_rate": 2.6104417670682734e-05,
      "loss": 0.0047,
      "step": 437
    },
    {
      "epoch": 438.0,
      "grad_norm": 0.00965973362326622,
      "learning_rate": 2.570281124497992e-05,
      "loss": 0.0047,
      "step": 438
    },
    {
      "epoch": 439.0,
      "grad_norm": 0.009659296832978725,
      "learning_rate": 2.530120481927711e-05,
      "loss": 0.0047,
      "step": 439
    },
    {
      "epoch": 440.0,
      "grad_norm": 0.00703496765345335,
      "learning_rate": 2.48995983935743e-05,
      "loss": 0.0047,
      "step": 440
    },
    {
      "epoch": 441.0,
      "grad_norm": 0.007030785549432039,
      "learning_rate": 2.4497991967871488e-05,
      "loss": 0.0047,
      "step": 441
    },
    {
      "epoch": 442.0,
      "grad_norm": 0.007043666671961546,
      "learning_rate": 2.4096385542168677e-05,
      "loss": 0.0047,
      "step": 442
    },
    {
      "epoch": 443.0,
      "grad_norm": 0.007040888071060181,
      "learning_rate": 2.3694779116465866e-05,
      "loss": 0.0047,
      "step": 443
    },
    {
      "epoch": 444.0,
      "grad_norm": 0.009699525311589241,
      "learning_rate": 2.3293172690763055e-05,
      "loss": 0.0047,
      "step": 444
    },
    {
      "epoch": 445.0,
      "grad_norm": 0.0007015722803771496,
      "learning_rate": 2.289156626506024e-05,
      "loss": 0.0049,
      "step": 445
    },
    {
      "epoch": 446.0,
      "grad_norm": 0.00966339185833931,
      "learning_rate": 2.248995983935743e-05,
      "loss": 0.0047,
      "step": 446
    },
    {
      "epoch": 447.0,
      "grad_norm": 0.009668245911598206,
      "learning_rate": 2.208835341365462e-05,
      "loss": 0.0047,
      "step": 447
    },
    {
      "epoch": 448.0,
      "grad_norm": 0.007046937942504883,
      "learning_rate": 2.168674698795181e-05,
      "loss": 0.0047,
      "step": 448
    },
    {
      "epoch": 449.0,
      "grad_norm": 0.0070527587085962296,
      "learning_rate": 2.1285140562248998e-05,
      "loss": 0.0047,
      "step": 449
    },
    {
      "epoch": 450.0,
      "grad_norm": 0.007038488052785397,
      "learning_rate": 2.0883534136546184e-05,
      "loss": 0.0047,
      "step": 450
    },
    {
      "epoch": 451.0,
      "grad_norm": 0.0021306683775037527,
      "learning_rate": 2.0481927710843373e-05,
      "loss": 0.0045,
      "step": 451
    },
    {
      "epoch": 452.0,
      "grad_norm": 0.007048788480460644,
      "learning_rate": 2.0080321285140562e-05,
      "loss": 0.0047,
      "step": 452
    },
    {
      "epoch": 453.0,
      "grad_norm": 0.0006838923436589539,
      "learning_rate": 1.967871485943775e-05,
      "loss": 0.0049,
      "step": 453
    },
    {
      "epoch": 454.0,
      "grad_norm": 0.00969866942614317,
      "learning_rate": 1.927710843373494e-05,
      "loss": 0.0047,
      "step": 454
    },
    {
      "epoch": 455.0,
      "grad_norm": 0.009712344966828823,
      "learning_rate": 1.8875502008032127e-05,
      "loss": 0.0047,
      "step": 455
    },
    {
      "epoch": 456.0,
      "grad_norm": 0.009688866324722767,
      "learning_rate": 1.8473895582329316e-05,
      "loss": 0.0047,
      "step": 456
    },
    {
      "epoch": 457.0,
      "grad_norm": 0.007071357686072588,
      "learning_rate": 1.8072289156626505e-05,
      "loss": 0.0047,
      "step": 457
    },
    {
      "epoch": 458.0,
      "grad_norm": 0.014747700653970242,
      "learning_rate": 1.7670682730923694e-05,
      "loss": 0.0045,
      "step": 458
    },
    {
      "epoch": 459.0,
      "grad_norm": 0.007095323875546455,
      "learning_rate": 1.7269076305220887e-05,
      "loss": 0.0047,
      "step": 459
    },
    {
      "epoch": 460.0,
      "grad_norm": 0.0021590902470052242,
      "learning_rate": 1.6867469879518073e-05,
      "loss": 0.0045,
      "step": 460
    },
    {
      "epoch": 461.0,
      "grad_norm": 0.0006770145264454186,
      "learning_rate": 1.6465863453815262e-05,
      "loss": 0.0049,
      "step": 461
    },
    {
      "epoch": 462.0,
      "grad_norm": 0.0006719469674862921,
      "learning_rate": 1.606425702811245e-05,
      "loss": 0.0049,
      "step": 462
    },
    {
      "epoch": 463.0,
      "grad_norm": 0.009731734171509743,
      "learning_rate": 1.566265060240964e-05,
      "loss": 0.0047,
      "step": 463
    },
    {
      "epoch": 464.0,
      "grad_norm": 0.009717794135212898,
      "learning_rate": 1.526104417670683e-05,
      "loss": 0.0047,
      "step": 464
    },
    {
      "epoch": 465.0,
      "grad_norm": 0.009746178984642029,
      "learning_rate": 1.4859437751004016e-05,
      "loss": 0.0047,
      "step": 465
    },
    {
      "epoch": 466.0,
      "grad_norm": 0.007102122530341148,
      "learning_rate": 1.4457831325301205e-05,
      "loss": 0.0047,
      "step": 466
    },
    {
      "epoch": 467.0,
      "grad_norm": 0.00708032725378871,
      "learning_rate": 1.4056224899598394e-05,
      "loss": 0.0047,
      "step": 467
    },
    {
      "epoch": 468.0,
      "grad_norm": 0.007086113095283508,
      "learning_rate": 1.3654618473895583e-05,
      "loss": 0.0047,
      "step": 468
    },
    {
      "epoch": 469.0,
      "grad_norm": 0.014743141829967499,
      "learning_rate": 1.3253012048192772e-05,
      "loss": 0.0045,
      "step": 469
    },
    {
      "epoch": 470.0,
      "grad_norm": 0.0021814049687236547,
      "learning_rate": 1.285140562248996e-05,
      "loss": 0.0045,
      "step": 470
    },
    {
      "epoch": 471.0,
      "grad_norm": 0.007091291714459658,
      "learning_rate": 1.244979919678715e-05,
      "loss": 0.0047,
      "step": 471
    },
    {
      "epoch": 472.0,
      "grad_norm": 0.0071353972889482975,
      "learning_rate": 1.2048192771084338e-05,
      "loss": 0.0047,
      "step": 472
    },
    {
      "epoch": 473.0,
      "grad_norm": 0.009766222909092903,
      "learning_rate": 1.1646586345381528e-05,
      "loss": 0.0047,
      "step": 473
    },
    {
      "epoch": 474.0,
      "grad_norm": 0.017397722229361534,
      "learning_rate": 1.1244979919678715e-05,
      "loss": 0.005,
      "step": 474
    },
    {
      "epoch": 475.0,
      "grad_norm": 0.026444077491760254,
      "learning_rate": 1.0843373493975904e-05,
      "loss": 0.0048,
      "step": 475
    },
    {
      "epoch": 476.0,
      "grad_norm": 0.009789888747036457,
      "learning_rate": 1.0441767068273092e-05,
      "loss": 0.0047,
      "step": 476
    },
    {
      "epoch": 477.0,
      "grad_norm": 0.007101455237716436,
      "learning_rate": 1.0040160642570281e-05,
      "loss": 0.0047,
      "step": 477
    },
    {
      "epoch": 478.0,
      "grad_norm": 0.007093328982591629,
      "learning_rate": 9.63855421686747e-06,
      "loss": 0.0047,
      "step": 478
    },
    {
      "epoch": 479.0,
      "grad_norm": 0.014785111881792545,
      "learning_rate": 9.236947791164658e-06,
      "loss": 0.0045,
      "step": 479
    },
    {
      "epoch": 480.0,
      "grad_norm": 0.014800012111663818,
      "learning_rate": 8.835341365461847e-06,
      "loss": 0.0045,
      "step": 480
    },
    {
      "epoch": 481.0,
      "grad_norm": 0.00710406294092536,
      "learning_rate": 8.433734939759036e-06,
      "loss": 0.0047,
      "step": 481
    },
    {
      "epoch": 482.0,
      "grad_norm": 0.0163735318928957,
      "learning_rate": 8.032128514056226e-06,
      "loss": 0.005,
      "step": 482
    },
    {
      "epoch": 483.0,
      "grad_norm": 0.00218359031714499,
      "learning_rate": 7.630522088353415e-06,
      "loss": 0.0045,
      "step": 483
    },
    {
      "epoch": 484.0,
      "grad_norm": 0.009749580174684525,
      "learning_rate": 7.228915662650602e-06,
      "loss": 0.0047,
      "step": 484
    },
    {
      "epoch": 485.0,
      "grad_norm": 0.009765205904841423,
      "learning_rate": 6.827309236947792e-06,
      "loss": 0.0047,
      "step": 485
    },
    {
      "epoch": 486.0,
      "grad_norm": 0.000685015635099262,
      "learning_rate": 6.42570281124498e-06,
      "loss": 0.0049,
      "step": 486
    },
    {
      "epoch": 487.0,
      "grad_norm": 0.00976163987070322,
      "learning_rate": 6.024096385542169e-06,
      "loss": 0.0047,
      "step": 487
    },
    {
      "epoch": 488.0,
      "grad_norm": 0.0006887281779199839,
      "learning_rate": 5.622489959839358e-06,
      "loss": 0.0049,
      "step": 488
    },
    {
      "epoch": 489.0,
      "grad_norm": 0.009809976443648338,
      "learning_rate": 5.220883534136546e-06,
      "loss": 0.0047,
      "step": 489
    },
    {
      "epoch": 490.0,
      "grad_norm": 0.009795522317290306,
      "learning_rate": 4.819277108433735e-06,
      "loss": 0.0047,
      "step": 490
    },
    {
      "epoch": 491.0,
      "grad_norm": 0.009749695658683777,
      "learning_rate": 4.417670682730924e-06,
      "loss": 0.0047,
      "step": 491
    },
    {
      "epoch": 492.0,
      "grad_norm": 0.0021824862342327833,
      "learning_rate": 4.016064257028113e-06,
      "loss": 0.0045,
      "step": 492
    },
    {
      "epoch": 493.0,
      "grad_norm": 0.007140113040804863,
      "learning_rate": 3.614457831325301e-06,
      "loss": 0.0047,
      "step": 493
    },
    {
      "epoch": 494.0,
      "grad_norm": 0.007124903611838818,
      "learning_rate": 3.21285140562249e-06,
      "loss": 0.0047,
      "step": 494
    },
    {
      "epoch": 495.0,
      "grad_norm": 0.0071158865466713905,
      "learning_rate": 2.811244979919679e-06,
      "loss": 0.0047,
      "step": 495
    },
    {
      "epoch": 496.0,
      "grad_norm": 0.007119073532521725,
      "learning_rate": 2.4096385542168676e-06,
      "loss": 0.0047,
      "step": 496
    },
    {
      "epoch": 497.0,
      "grad_norm": 0.007112898863852024,
      "learning_rate": 2.0080321285140564e-06,
      "loss": 0.0047,
      "step": 497
    },
    {
      "epoch": 498.0,
      "grad_norm": 0.014866248704493046,
      "learning_rate": 1.606425702811245e-06,
      "loss": 0.0045,
      "step": 498
    },
    {
      "epoch": 499.0,
      "grad_norm": 0.007127212826162577,
      "learning_rate": 1.2048192771084338e-06,
      "loss": 0.0047,
      "step": 499
    },
    {
      "epoch": 500.0,
      "grad_norm": 0.007142535876482725,
      "learning_rate": 8.032128514056225e-07,
      "loss": 0.0047,
      "step": 500
    }
  ],
  "logging_steps": 1,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 500,
  "save_steps": 500,
  "total_flos": 3475984143360000.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}