{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 2181,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 47.86506838160147,
      "learning_rate": 1.5151515151515153e-05,
      "loss": 7.8203,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 50.594463388127075,
      "learning_rate": 3.0303030303030306e-05,
      "loss": 7.9062,
      "step": 2
    },
    {
      "epoch": 0.0,
      "grad_norm": 44.62377962212157,
      "learning_rate": 4.545454545454546e-05,
      "loss": 7.4336,
      "step": 3
    },
    {
      "epoch": 0.0,
      "grad_norm": 16.7241203577855,
      "learning_rate": 6.060606060606061e-05,
      "loss": 5.9336,
      "step": 4
    },
    {
      "epoch": 0.0,
      "grad_norm": 11.94319218216758,
      "learning_rate": 7.575757575757576e-05,
      "loss": 5.625,
      "step": 5
    },
    {
      "epoch": 0.0,
      "grad_norm": 9.417781346187903,
      "learning_rate": 9.090909090909092e-05,
      "loss": 5.3477,
      "step": 6
    },
    {
      "epoch": 0.0,
      "grad_norm": 6.764798347328713,
      "learning_rate": 0.00010606060606060606,
      "loss": 5.0625,
      "step": 7
    },
    {
      "epoch": 0.0,
      "grad_norm": 5.947954081633672,
      "learning_rate": 0.00012121212121212122,
      "loss": 4.8945,
      "step": 8
    },
    {
      "epoch": 0.0,
      "grad_norm": 5.180700854603152,
      "learning_rate": 0.00013636363636363637,
      "loss": 4.668,
      "step": 9
    },
    {
      "epoch": 0.0,
      "grad_norm": 4.434918411594634,
      "learning_rate": 0.00015151515151515152,
      "loss": 4.5039,
      "step": 10
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.8064194830813243,
      "learning_rate": 0.00016666666666666666,
      "loss": 4.3906,
      "step": 11
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.0906140766370225,
      "learning_rate": 0.00018181818181818183,
      "loss": 4.3594,
      "step": 12
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.6267713377496085,
      "learning_rate": 0.00019696969696969695,
      "loss": 4.2324,
      "step": 13
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.1937042395765385,
      "learning_rate": 0.00021212121212121213,
      "loss": 4.2773,
      "step": 14
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.1346981140649284,
      "learning_rate": 0.00022727272727272727,
      "loss": 4.1367,
      "step": 15
    },
    {
      "epoch": 0.01,
      "grad_norm": 4.51173522128777,
      "learning_rate": 0.00024242424242424245,
      "loss": 4.082,
      "step": 16
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6229764531307291,
      "learning_rate": 0.00025757575757575756,
      "loss": 4.0293,
      "step": 17
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.3112964584613946,
      "learning_rate": 0.00027272727272727274,
      "loss": 4.0332,
      "step": 18
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.2078779902698566,
      "learning_rate": 0.0002878787878787879,
      "loss": 3.9863,
      "step": 19
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.221090024002337,
      "learning_rate": 0.00030303030303030303,
      "loss": 3.832,
      "step": 20
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.156089372850874,
      "learning_rate": 0.0003181818181818182,
      "loss": 3.8691,
      "step": 21
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.086889713435333,
      "learning_rate": 0.0003333333333333333,
      "loss": 3.9336,
      "step": 22
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.091733182913575,
      "learning_rate": 0.0003484848484848485,
      "loss": 3.7891,
      "step": 23
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9285410195112951,
      "learning_rate": 0.00036363636363636367,
      "loss": 3.7383,
      "step": 24
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9532626094960892,
      "learning_rate": 0.0003787878787878788,
      "loss": 3.7969,
      "step": 25
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.8654363716770672,
      "learning_rate": 0.0003939393939393939,
      "loss": 3.6758,
      "step": 26
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.682826499882736,
      "learning_rate": 0.00040909090909090913,
      "loss": 3.6016,
      "step": 27
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.8458374543911253,
      "learning_rate": 0.00042424242424242425,
      "loss": 3.6484,
      "step": 28
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.6571642660070606,
      "learning_rate": 0.0004393939393939394,
      "loss": 3.6426,
      "step": 29
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.7655837392074143,
      "learning_rate": 0.00045454545454545455,
      "loss": 3.5977,
      "step": 30
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.5770099476206532,
      "learning_rate": 0.0004696969696969697,
      "loss": 3.5391,
      "step": 31
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.5833753044058878,
      "learning_rate": 0.0004848484848484849,
      "loss": 3.584,
      "step": 32
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.5801547849287838,
      "learning_rate": 0.0005,
      "loss": 3.5723,
      "step": 33
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.44871484911017456,
      "learning_rate": 0.0005151515151515151,
      "loss": 3.4551,
      "step": 34
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.44321161005639526,
      "learning_rate": 0.0005303030303030302,
      "loss": 3.5293,
      "step": 35
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.3864970958909212,
      "learning_rate": 0.0005454545454545455,
      "loss": 3.4551,
      "step": 36
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.3549877153443857,
      "learning_rate": 0.0005606060606060606,
      "loss": 3.4414,
      "step": 37
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.378392032822761,
      "learning_rate": 0.0005757575757575758,
      "loss": 3.4824,
      "step": 38
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.3428280159703215,
      "learning_rate": 0.0005909090909090909,
      "loss": 3.3848,
      "step": 39
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.32767673492919924,
      "learning_rate": 0.0006060606060606061,
      "loss": 3.3809,
      "step": 40
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.334821457006923,
      "learning_rate": 0.0006212121212121212,
      "loss": 3.3613,
      "step": 41
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.3202568846515324,
      "learning_rate": 0.0006363636363636364,
      "loss": 3.3984,
      "step": 42
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.2925483664742464,
      "learning_rate": 0.0006515151515151515,
      "loss": 3.3477,
      "step": 43
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.2902142455697401,
      "learning_rate": 0.0006666666666666666,
      "loss": 3.4043,
      "step": 44
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.2598947233441998,
      "learning_rate": 0.0006818181818181818,
      "loss": 3.3457,
      "step": 45
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.2592234564843323,
      "learning_rate": 0.000696969696969697,
      "loss": 3.2793,
      "step": 46
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.23603633650933512,
      "learning_rate": 0.0007121212121212122,
      "loss": 3.3184,
      "step": 47
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.2748689690186025,
      "learning_rate": 0.0007272727272727273,
      "loss": 3.3652,
      "step": 48
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.2629225299528596,
      "learning_rate": 0.0007424242424242425,
      "loss": 3.3027,
      "step": 49
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.2469303847577921,
      "learning_rate": 0.0007575757575757576,
      "loss": 3.2754,
      "step": 50
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.24127030704521918,
      "learning_rate": 0.0007727272727272727,
      "loss": 3.2402,
      "step": 51
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.31586965242752907,
      "learning_rate": 0.0007878787878787878,
      "loss": 3.4004,
      "step": 52
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.25298592867180203,
      "learning_rate": 0.000803030303030303,
      "loss": 3.3125,
      "step": 53
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.26513169032879813,
      "learning_rate": 0.0008181818181818183,
      "loss": 3.3691,
      "step": 54
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.2488295233633195,
      "learning_rate": 0.0008333333333333334,
      "loss": 3.3203,
      "step": 55
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.2631272077495346,
      "learning_rate": 0.0008484848484848485,
      "loss": 3.2871,
      "step": 56
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.24055493018603727,
      "learning_rate": 0.0008636363636363636,
      "loss": 3.3398,
      "step": 57
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.22553643367649442,
      "learning_rate": 0.0008787878787878789,
      "loss": 3.3477,
      "step": 58
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.2200086243783038,
      "learning_rate": 0.000893939393939394,
      "loss": 3.1855,
      "step": 59
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.24842241614450047,
      "learning_rate": 0.0009090909090909091,
      "loss": 3.2324,
      "step": 60
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.2704306608222093,
      "learning_rate": 0.0009242424242424242,
      "loss": 3.3789,
      "step": 61
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.22967337347891578,
      "learning_rate": 0.0009393939393939394,
      "loss": 3.2891,
      "step": 62
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.23586268617017841,
      "learning_rate": 0.0009545454545454546,
      "loss": 3.3359,
      "step": 63
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.26503678560733923,
      "learning_rate": 0.0009696969696969698,
      "loss": 3.2559,
      "step": 64
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.2393056532396721,
      "learning_rate": 0.000984848484848485,
      "loss": 3.2109,
      "step": 65
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.23467709970411488,
      "learning_rate": 0.001,
      "loss": 3.2188,
      "step": 66
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.30035122100687733,
      "learning_rate": 0.0009999994484067654,
      "loss": 3.252,
      "step": 67
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.31983795887246347,
      "learning_rate": 0.0009999977936282788,
      "loss": 3.2637,
      "step": 68
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.3058876706872471,
      "learning_rate": 0.0009999950356681913,
      "loss": 3.2871,
      "step": 69
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.3131573118302609,
      "learning_rate": 0.0009999911745325876,
      "loss": 3.2344,
      "step": 70
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.32983033039370624,
      "learning_rate": 0.0009999862102299873,
      "loss": 3.2148,
      "step": 71
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.4179934133053002,
      "learning_rate": 0.0009999801427713433,
      "loss": 3.2188,
      "step": 72
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.36846692894699595,
      "learning_rate": 0.0009999729721700424,
      "loss": 3.1758,
      "step": 73
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.3466883485133617,
      "learning_rate": 0.000999964698441906,
      "loss": 3.1875,
      "step": 74
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.32155734930680047,
      "learning_rate": 0.0009999553216051892,
      "loss": 3.1172,
      "step": 75
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.3610726622166297,
      "learning_rate": 0.00099994484168058,
      "loss": 3.2617,
      "step": 76
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.33502441630746665,
      "learning_rate": 0.0009999332586912019,
      "loss": 3.3027,
      "step": 77
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.36469811845083544,
      "learning_rate": 0.0009999205726626108,
      "loss": 3.1738,
      "step": 78
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.29051877940050586,
      "learning_rate": 0.000999906783622797,
      "loss": 3.1797,
      "step": 79
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.2690245295740612,
      "learning_rate": 0.0009998918916021842,
      "loss": 3.1699,
      "step": 80
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.31151466708558123,
      "learning_rate": 0.0009998758966336297,
      "loss": 3.1953,
      "step": 81
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.40011631179537493,
      "learning_rate": 0.0009998587987524242,
      "loss": 3.1797,
      "step": 82
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3225160460705253,
      "learning_rate": 0.0009998405979962926,
      "loss": 3.0879,
      "step": 83
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.35480419383806533,
      "learning_rate": 0.000999821294405392,
      "loss": 3.252,
      "step": 84
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3405919001734859,
      "learning_rate": 0.0009998008880223134,
      "loss": 3.1426,
      "step": 85
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3328111817596931,
      "learning_rate": 0.000999779378892081,
      "loss": 3.207,
      "step": 86
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.2793625499167308,
      "learning_rate": 0.0009997567670621522,
      "loss": 3.1445,
      "step": 87
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3091482756082596,
      "learning_rate": 0.0009997330525824165,
      "loss": 3.1426,
      "step": 88
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3088971974772615,
      "learning_rate": 0.0009997082355051976,
      "loss": 3.1621,
      "step": 89
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.35714958752899206,
      "learning_rate": 0.000999682315885251,
      "loss": 3.0801,
      "step": 90
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.34547289451321395,
      "learning_rate": 0.0009996552937797645,
      "loss": 3.1719,
      "step": 91
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3617884365339593,
      "learning_rate": 0.0009996271692483596,
      "loss": 3.1367,
      "step": 92
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.4061372891929322,
      "learning_rate": 0.0009995979423530893,
      "loss": 3.1152,
      "step": 93
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3439714580466316,
      "learning_rate": 0.000999567613158439,
      "loss": 3.0117,
      "step": 94
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3922903238058651,
      "learning_rate": 0.0009995361817313263,
      "loss": 3.0879,
      "step": 95
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.4006442705098572,
      "learning_rate": 0.0009995036481411004,
      "loss": 3.0957,
      "step": 96
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3772155587535038,
      "learning_rate": 0.0009994700124595429,
      "loss": 3.082,
      "step": 97
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.437788387540688,
      "learning_rate": 0.0009994352747608663,
      "loss": 3.0762,
      "step": 98
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.36691457723333887,
      "learning_rate": 0.0009993994351217151,
      "loss": 3.0938,
      "step": 99
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.3827853978673069,
      "learning_rate": 0.000999362493621165,
      "loss": 3.0137,
      "step": 100
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.37390937558803516,
      "learning_rate": 0.0009993244503407226,
      "loss": 3.0508,
      "step": 101
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.43366765376331906,
      "learning_rate": 0.0009992853053643258,
      "loss": 3.0371,
      "step": 102
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.37812126102463417,
      "learning_rate": 0.0009992450587783426,
      "loss": 3.0254,
      "step": 103
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.46964683337495083,
      "learning_rate": 0.000999203710671572,
      "loss": 3.0156,
      "step": 104
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.44669378426211276,
      "learning_rate": 0.0009991612611352438,
      "loss": 2.9902,
      "step": 105
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.43946712676895294,
      "learning_rate": 0.0009991177102630173,
      "loss": 3.1172,
      "step": 106
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.48714643303158084,
      "learning_rate": 0.0009990730581509817,
      "loss": 2.9375,
      "step": 107
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.3638563371980752,
      "learning_rate": 0.0009990273048976566,
      "loss": 3.0332,
      "step": 108
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.5241951770899201,
      "learning_rate": 0.0009989804506039905,
      "loss": 2.9824,
      "step": 109
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.5744540531615141,
      "learning_rate": 0.0009989324953733614,
      "loss": 3.1172,
      "step": 110
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.4255686958261268,
      "learning_rate": 0.0009988834393115766,
      "loss": 3.0273,
      "step": 111
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.5088350023648092,
      "learning_rate": 0.000998833282526872,
      "loss": 3.0234,
      "step": 112
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.465414489065429,
      "learning_rate": 0.0009987820251299122,
      "loss": 3.0605,
      "step": 113
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.40937129209229006,
      "learning_rate": 0.00099872966723379,
      "loss": 2.998,
      "step": 114
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.4728653306856754,
      "learning_rate": 0.0009986762089540266,
      "loss": 3.0156,
      "step": 115
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.47683272968524615,
      "learning_rate": 0.0009986216504085709,
      "loss": 3.0312,
      "step": 116
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.5235318228576897,
      "learning_rate": 0.0009985659917177991,
      "loss": 3.0098,
      "step": 117
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.4591326236995222,
      "learning_rate": 0.0009985092330045155,
      "loss": 3.0098,
      "step": 118
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.488423697078249,
      "learning_rate": 0.0009984513743939508,
      "loss": 2.8809,
      "step": 119
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.43245491355895677,
      "learning_rate": 0.0009983924160137626,
      "loss": 2.9902,
      "step": 120
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.3849951802017609,
      "learning_rate": 0.000998332357994035,
      "loss": 3.0254,
      "step": 121
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4806601114926406,
      "learning_rate": 0.0009982712004672786,
      "loss": 3.0312,
      "step": 122
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.509203715204686,
      "learning_rate": 0.0009982089435684295,
      "loss": 3.0059,
      "step": 123
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.6197170804217604,
      "learning_rate": 0.0009981455874348499,
      "loss": 2.9062,
      "step": 124
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4856484445278653,
      "learning_rate": 0.0009980811322063269,
      "loss": 2.9961,
      "step": 125
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.565560994257061,
      "learning_rate": 0.0009980155780250728,
      "loss": 3.0273,
      "step": 126
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.43687677124627766,
      "learning_rate": 0.0009979489250357243,
      "loss": 2.9863,
      "step": 127
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4577758351617377,
      "learning_rate": 0.0009978811733853431,
      "loss": 3.0098,
      "step": 128
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.5104124723463535,
      "learning_rate": 0.0009978123232234147,
      "loss": 2.8809,
      "step": 129
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4462118754400629,
      "learning_rate": 0.000997742374701848,
      "loss": 3.1074,
      "step": 130
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4899308068113754,
      "learning_rate": 0.0009976713279749754,
      "loss": 2.9316,
      "step": 131
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4397247949935912,
      "learning_rate": 0.0009975991831995528,
      "loss": 2.9863,
      "step": 132
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.5113745508444417,
      "learning_rate": 0.0009975259405347581,
      "loss": 2.9824,
      "step": 133
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4912823690654896,
      "learning_rate": 0.0009974516001421926,
      "loss": 2.9238,
      "step": 134
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4019183484689013,
      "learning_rate": 0.000997376162185878,
      "loss": 2.9023,
      "step": 135
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4384848127871296,
      "learning_rate": 0.0009972996268322594,
      "loss": 2.9609,
      "step": 136
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.5853430416169073,
      "learning_rate": 0.0009972219942502017,
      "loss": 2.9297,
      "step": 137
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4929065589760861,
      "learning_rate": 0.0009971432646109918,
      "loss": 2.8945,
      "step": 138
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4178915363657808,
      "learning_rate": 0.0009970634380883365,
      "loss": 2.8809,
      "step": 139
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.5672325923457944,
      "learning_rate": 0.0009969825148583627,
      "loss": 2.9434,
      "step": 140
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.485765070528646,
      "learning_rate": 0.0009969004950996173,
      "loss": 3.0938,
      "step": 141
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.46650449519267967,
      "learning_rate": 0.0009968173789930668,
      "loss": 2.9355,
      "step": 142
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4562707484345187,
      "learning_rate": 0.0009967331667220958,
      "loss": 2.9102,
      "step": 143
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.46134009259441067,
      "learning_rate": 0.0009966478584725086,
      "loss": 2.9707,
      "step": 144
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.49960752691239535,
      "learning_rate": 0.0009965614544325263,
      "loss": 2.9316,
      "step": 145
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4351680489285944,
      "learning_rate": 0.000996473954792789,
      "loss": 2.834,
      "step": 146
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4570384428103238,
      "learning_rate": 0.0009963853597463532,
      "loss": 2.8633,
      "step": 147
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4250286974719292,
      "learning_rate": 0.000996295669488693,
      "loss": 2.8164,
      "step": 148
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4619348185636194,
      "learning_rate": 0.0009962048842176979,
      "loss": 3.0195,
      "step": 149
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.42913053792230865,
      "learning_rate": 0.0009961130041336748,
      "loss": 2.8633,
      "step": 150
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4461931909130763,
      "learning_rate": 0.0009960200294393449,
      "loss": 2.8691,
      "step": 151
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4254043470691929,
      "learning_rate": 0.0009959259603398453,
      "loss": 2.916,
      "step": 152
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4324459596562772,
      "learning_rate": 0.0009958307970427275,
      "loss": 2.916,
      "step": 153
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.42333087030280897,
      "learning_rate": 0.0009957345397579572,
      "loss": 2.9453,
      "step": 154
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4152251436168111,
      "learning_rate": 0.0009956371886979138,
      "loss": 2.8477,
      "step": 155
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.40902798568662624,
      "learning_rate": 0.00099553874407739,
      "loss": 2.7793,
      "step": 156
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.5034155825062112,
      "learning_rate": 0.0009954392061135916,
      "loss": 2.9297,
      "step": 157
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4690841506392611,
      "learning_rate": 0.0009953385750261364,
      "loss": 2.9902,
      "step": 158
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.5500127501789521,
      "learning_rate": 0.0009952368510370538,
      "loss": 2.8574,
      "step": 159
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.43870794160816823,
      "learning_rate": 0.0009951340343707852,
      "loss": 2.8301,
      "step": 160
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.41943078250381427,
      "learning_rate": 0.0009950301252541823,
      "loss": 2.8535,
      "step": 161
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.5897987665145353,
      "learning_rate": 0.0009949251239165075,
      "loss": 2.8379,
      "step": 162
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.5019479593148473,
      "learning_rate": 0.000994819030589433,
      "loss": 2.9238,
      "step": 163
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.44201148899843645,
      "learning_rate": 0.00099471184550704,
      "loss": 2.7891,
      "step": 164
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.5772043403690974,
      "learning_rate": 0.0009946035689058189,
      "loss": 2.9102,
      "step": 165
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.425964505851097,
      "learning_rate": 0.0009944942010246681,
      "loss": 2.8008,
      "step": 166
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.44838145065604496,
      "learning_rate": 0.0009943837421048942,
      "loss": 2.7754,
      "step": 167
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.4978828124601326,
      "learning_rate": 0.0009942721923902106,
      "loss": 2.8223,
      "step": 168
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.41734008363251524,
      "learning_rate": 0.0009941595521267377,
      "loss": 2.793,
      "step": 169
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.5249039744196557,
      "learning_rate": 0.0009940458215630017,
      "loss": 2.834,
      "step": 170
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.4968909371946741,
      "learning_rate": 0.0009939310009499348,
      "loss": 2.8027,
      "step": 171
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.43611688254477843,
      "learning_rate": 0.000993815090540874,
      "loss": 2.8418,
      "step": 172
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.5444293734575072,
      "learning_rate": 0.000993698090591561,
      "loss": 2.8496,
      "step": 173
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.4716307323928379,
      "learning_rate": 0.0009935800013601416,
      "loss": 2.8066,
      "step": 174
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.4943025229770689,
      "learning_rate": 0.000993460823107164,
      "loss": 2.7363,
      "step": 175
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.47904549432981036,
      "learning_rate": 0.0009933405560955803,
      "loss": 2.8789,
      "step": 176
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.4319629729111764,
      "learning_rate": 0.0009932192005907446,
      "loss": 2.8105,
      "step": 177
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.4595519639951301,
      "learning_rate": 0.0009930967568604118,
      "loss": 2.8672,
      "step": 178
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.4224923528889133,
      "learning_rate": 0.000992973225174739,
      "loss": 2.8184,
      "step": 179
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.4592983193685254,
      "learning_rate": 0.0009928486058062827,
      "loss": 2.7715,
      "step": 180
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.42950000390717896,
      "learning_rate": 0.0009927228990299999,
      "loss": 2.8242,
      "step": 181
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.5345087229025187,
      "learning_rate": 0.0009925961051232468,
      "loss": 2.7617,
      "step": 182
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.44847874227472034,
      "learning_rate": 0.000992468224365778,
      "loss": 2.8691,
      "step": 183
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.45775201146851463,
      "learning_rate": 0.000992339257039746,
      "loss": 2.7773,
      "step": 184
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.5146851960795359,
      "learning_rate": 0.0009922092034297006,
      "loss": 2.832,
      "step": 185
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4271610196766863,
      "learning_rate": 0.0009920780638225891,
      "loss": 2.7188,
      "step": 186
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.39526360961961876,
      "learning_rate": 0.0009919458385077538,
      "loss": 2.7305,
      "step": 187
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.45636934763129333,
      "learning_rate": 0.0009918125277769336,
      "loss": 2.7676,
      "step": 188
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.40604647746061434,
      "learning_rate": 0.0009916781319242614,
      "loss": 2.8027,
      "step": 189
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.46602922716598144,
      "learning_rate": 0.0009915426512462646,
      "loss": 2.7344,
      "step": 190
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4624317389515949,
      "learning_rate": 0.0009914060860418644,
      "loss": 2.6758,
      "step": 191
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4409248359700915,
      "learning_rate": 0.000991268436612374,
      "loss": 2.7852,
      "step": 192
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.3948161040640291,
      "learning_rate": 0.0009911297032614997,
      "loss": 2.7617,
      "step": 193
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4677895296373518,
      "learning_rate": 0.000990989886295339,
      "loss": 2.7656,
      "step": 194
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.5208506258748722,
      "learning_rate": 0.0009908489860223804,
      "loss": 2.877,
      "step": 195
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4369314776376022,
      "learning_rate": 0.000990707002753502,
      "loss": 2.7871,
      "step": 196
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.5794664359413786,
      "learning_rate": 0.0009905639368019724,
      "loss": 2.7969,
      "step": 197
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.46365465521723365,
      "learning_rate": 0.0009904197884834482,
      "loss": 2.7305,
      "step": 198
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.5265372243621155,
      "learning_rate": 0.0009902745581159742,
      "loss": 2.7266,
      "step": 199
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.5003697883085344,
      "learning_rate": 0.0009901282460199829,
      "loss": 2.8691,
      "step": 200
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.5217280029188144,
      "learning_rate": 0.0009899808525182935,
      "loss": 2.8652,
      "step": 201
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4772844517698489,
      "learning_rate": 0.0009898323779361107,
      "loss": 2.7734,
      "step": 202
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4991486677670108,
      "learning_rate": 0.000989682822601025,
      "loss": 2.7812,
      "step": 203
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4039466836864733,
      "learning_rate": 0.0009895321868430113,
      "loss": 2.7793,
      "step": 204
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4518739294939531,
      "learning_rate": 0.0009893804709944281,
      "loss": 2.8652,
      "step": 205
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.5152346371067269,
      "learning_rate": 0.0009892276753900174,
      "loss": 2.7715,
      "step": 206
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.447279392484993,
      "learning_rate": 0.0009890738003669028,
      "loss": 2.7969,
      "step": 207
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4654857512562644,
      "learning_rate": 0.0009889188462645904,
      "loss": 2.7559,
      "step": 208
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4434012847341127,
      "learning_rate": 0.0009887628134249667,
      "loss": 2.6523,
      "step": 209
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4445773571859153,
      "learning_rate": 0.0009886057021922983,
      "loss": 2.7383,
      "step": 210
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4201442119503552,
      "learning_rate": 0.0009884475129132311,
      "loss": 2.7637,
      "step": 211
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4307045098533661,
      "learning_rate": 0.0009882882459367897,
      "loss": 2.7617,
      "step": 212
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.46339782214220343,
      "learning_rate": 0.0009881279016143766,
      "loss": 2.7598,
      "step": 213
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.40531371096782465,
      "learning_rate": 0.0009879664802997707,
      "loss": 2.7188,
      "step": 214
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4096196980079817,
      "learning_rate": 0.000987803982349128,
      "loss": 2.709,
      "step": 215
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4614664591539858,
      "learning_rate": 0.0009876404081209796,
      "loss": 2.748,
      "step": 216
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.48369409238909705,
      "learning_rate": 0.000987475757976231,
      "loss": 2.7285,
      "step": 217
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.42631320904382114,
      "learning_rate": 0.000987310032278162,
      "loss": 2.7012,
      "step": 218
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.44876872798256584,
      "learning_rate": 0.0009871432313924254,
      "loss": 2.8184,
      "step": 219
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.43154716410348914,
      "learning_rate": 0.000986975355687046,
      "loss": 2.7598,
      "step": 220
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4394806600262461,
      "learning_rate": 0.0009868064055324204,
      "loss": 2.8223,
      "step": 221
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.44722499868525356,
      "learning_rate": 0.0009866363813013153,
      "loss": 2.7637,
      "step": 222
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.44688571653914927,
      "learning_rate": 0.0009864652833688676,
      "loss": 2.7188,
      "step": 223
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4499549318354169,
      "learning_rate": 0.0009862931121125836,
      "loss": 2.7129,
      "step": 224
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4008445085745039,
      "learning_rate": 0.000986119867912337,
      "loss": 2.709,
      "step": 225
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.46572587673693955,
      "learning_rate": 0.000985945551150369,
      "loss": 2.7207,
      "step": 226
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4213129397673624,
      "learning_rate": 0.0009857701622112876,
      "loss": 2.7051,
      "step": 227
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4797468166078364,
      "learning_rate": 0.000985593701482066,
      "loss": 2.7285,
      "step": 228
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.43007584076231714,
      "learning_rate": 0.0009854161693520424,
      "loss": 2.7754,
      "step": 229
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5078872982631744,
      "learning_rate": 0.0009852375662129194,
      "loss": 2.7754,
      "step": 230
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.430317644786795,
      "learning_rate": 0.0009850578924587613,
      "loss": 2.7148,
      "step": 231
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.42213116494102537,
      "learning_rate": 0.000984877148485996,
      "loss": 2.752,
      "step": 232
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.45765119118003195,
      "learning_rate": 0.000984695334693412,
      "loss": 2.6973,
      "step": 233
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5477949235256255,
      "learning_rate": 0.000984512451482158,
      "loss": 2.7148,
      "step": 234
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.4143610767069968,
      "learning_rate": 0.0009843284992557431,
      "loss": 2.6387,
      "step": 235
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.43344443895497536,
      "learning_rate": 0.000984143478420034,
      "loss": 2.7461,
      "step": 236
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.49441390057651774,
      "learning_rate": 0.0009839573893832563,
      "loss": 2.6543,
      "step": 237
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.4625381420873913,
      "learning_rate": 0.000983770232555991,
      "loss": 2.6914,
      "step": 238
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.41165887195734596,
      "learning_rate": 0.0009835820083511765,
      "loss": 2.6582,
      "step": 239
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.42381735410468685,
      "learning_rate": 0.0009833927171841055,
      "loss": 2.7441,
      "step": 240
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.41376870066259136,
      "learning_rate": 0.0009832023594724246,
      "loss": 2.7129,
      "step": 241
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.49431462926572356,
      "learning_rate": 0.0009830109356361344,
      "loss": 2.707,
      "step": 242
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.4466298202420403,
      "learning_rate": 0.0009828184460975867,
      "loss": 2.7793,
      "step": 243
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.4674198381685654,
      "learning_rate": 0.0009826248912814855,
      "loss": 2.7461,
      "step": 244
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.426533642204709,
      "learning_rate": 0.0009824302716148847,
      "loss": 2.6738,
      "step": 245
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.4481397452767005,
      "learning_rate": 0.0009822345875271884,
      "loss": 2.6797,
      "step": 246
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.4481130276370919,
      "learning_rate": 0.0009820378394501481,
      "loss": 2.6797,
      "step": 247
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.4113413515165194,
      "learning_rate": 0.0009818400278178636,
      "loss": 2.6934,
      "step": 248
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.4111459701209743,
      "learning_rate": 0.0009816411530667814,
      "loss": 2.7715,
      "step": 249
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.45696644509514434,
      "learning_rate": 0.000981441215635693,
      "loss": 2.7207,
      "step": 250
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4283039504923376,
      "learning_rate": 0.0009812402159657353,
      "loss": 2.8145,
      "step": 251
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.3992372425790597,
      "learning_rate": 0.000981038154500388,
      "loss": 2.7207,
      "step": 252
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.41563624502597973,
      "learning_rate": 0.0009808350316854746,
      "loss": 2.7168,
      "step": 253
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4828670584209116,
      "learning_rate": 0.0009806308479691594,
      "loss": 2.6621,
      "step": 254
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.41747574451489566,
      "learning_rate": 0.0009804256038019482,
      "loss": 2.7695,
      "step": 255
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4675510431316842,
      "learning_rate": 0.0009802192996366857,
      "loss": 2.6914,
      "step": 256
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.417774064542343,
      "learning_rate": 0.0009800119359285563,
      "loss": 2.6621,
      "step": 257
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.49377131301981203,
      "learning_rate": 0.0009798035131350813,
      "loss": 2.7871,
      "step": 258
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4346270580090776,
      "learning_rate": 0.0009795940317161194,
      "loss": 2.7188,
      "step": 259
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.3927118727949113,
      "learning_rate": 0.0009793834921338646,
      "loss": 2.7734,
      "step": 260
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.39755111634654033,
      "learning_rate": 0.0009791718948528457,
      "loss": 2.6855,
      "step": 261
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4454824889220223,
      "learning_rate": 0.0009789592403399252,
      "loss": 2.6621,
      "step": 262
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4111766064146217,
      "learning_rate": 0.0009787455290642985,
      "loss": 2.8184,
      "step": 263
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.38609627945279124,
      "learning_rate": 0.000978530761497492,
      "loss": 2.7324,
      "step": 264
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4538686893856415,
      "learning_rate": 0.0009783149381133633,
      "loss": 2.7148,
      "step": 265
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4812967381761417,
      "learning_rate": 0.0009780980593880992,
      "loss": 2.7676,
      "step": 266
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.42923905802889617,
      "learning_rate": 0.0009778801258002153,
      "loss": 2.709,
      "step": 267
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.45119144941159417,
      "learning_rate": 0.000977661137830554,
      "loss": 2.6309,
      "step": 268
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4097181477792646,
      "learning_rate": 0.0009774410959622845,
      "loss": 2.7207,
      "step": 269
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.45786994158321553,
      "learning_rate": 0.000977220000680901,
      "loss": 2.7695,
      "step": 270
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.44744419928949114,
      "learning_rate": 0.000976997852474223,
      "loss": 2.7051,
      "step": 271
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4112771290915934,
      "learning_rate": 0.0009767746518323914,
      "loss": 2.6387,
      "step": 272
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.44357328667514834,
      "learning_rate": 0.0009765503992478704,
      "loss": 2.6562,
      "step": 273
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.42284509713613033,
      "learning_rate": 0.0009763250952154449,
      "loss": 2.6406,
      "step": 274
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.459532322421133,
      "learning_rate": 0.0009760987402322195,
      "loss": 2.5254,
      "step": 275
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.41708518696703833,
      "learning_rate": 0.0009758713347976178,
      "loss": 2.6777,
      "step": 276
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.39430173051079204,
      "learning_rate": 0.000975642879413381,
      "loss": 2.6348,
      "step": 277
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4115269151673862,
      "learning_rate": 0.0009754133745835665,
      "loss": 2.6875,
      "step": 278
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4433412937461119,
      "learning_rate": 0.0009751828208145482,
      "loss": 2.709,
      "step": 279
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4068266839353352,
      "learning_rate": 0.0009749512186150131,
      "loss": 2.6816,
      "step": 280
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.3916163061968331,
      "learning_rate": 0.0009747185684959625,
      "loss": 2.7344,
      "step": 281
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.39517780012347886,
      "learning_rate": 0.000974484870970709,
      "loss": 2.75,
      "step": 282
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4060936358081258,
      "learning_rate": 0.0009742501265548767,
      "loss": 2.6621,
      "step": 283
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4287075069606316,
      "learning_rate": 0.0009740143357663993,
      "loss": 2.6094,
      "step": 284
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4431728891691112,
      "learning_rate": 0.000973777499125519,
      "loss": 2.7832,
      "step": 285
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.49091368487067183,
      "learning_rate": 0.0009735396171547859,
      "loss": 2.6797,
      "step": 286
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.41032985099334407,
      "learning_rate": 0.0009733006903790564,
      "loss": 2.6738,
      "step": 287
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.45114552875198755,
      "learning_rate": 0.0009730607193254922,
      "loss": 2.6836,
      "step": 288
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.43645305940947604,
      "learning_rate": 0.0009728197045235585,
      "loss": 2.6914,
      "step": 289
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.42661320377132694,
      "learning_rate": 0.0009725776465050242,
      "loss": 2.6836,
      "step": 290
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.440272607791669,
      "learning_rate": 0.0009723345458039594,
      "loss": 2.6602,
      "step": 291
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.3868635723670402,
      "learning_rate": 0.000972090402956735,
      "loss": 2.666,
      "step": 292
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.3899925517820767,
      "learning_rate": 0.0009718452185020212,
      "loss": 2.543,
      "step": 293
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4213229334991681,
      "learning_rate": 0.0009715989929807862,
      "loss": 2.5938,
      "step": 294
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4005167394699552,
      "learning_rate": 0.0009713517269362955,
      "loss": 2.5586,
      "step": 295
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.38883703429454297,
      "learning_rate": 0.0009711034209141101,
      "loss": 2.6113,
      "step": 296
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5041876565539211,
      "learning_rate": 0.0009708540754620856,
      "loss": 2.6895,
      "step": 297
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.43421589470999894,
      "learning_rate": 0.0009706036911303713,
      "loss": 2.6992,
      "step": 298
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4286146869177898,
      "learning_rate": 0.0009703522684714083,
      "loss": 2.6035,
      "step": 299
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.44638278106150464,
      "learning_rate": 0.0009700998080399286,
      "loss": 2.6426,
      "step": 300
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4254542097737601,
      "learning_rate": 0.0009698463103929542,
      "loss": 2.668,
      "step": 301
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4322599236303823,
      "learning_rate": 0.0009695917760897954,
      "loss": 2.6543,
      "step": 302
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4614055221641415,
      "learning_rate": 0.0009693362056920501,
      "loss": 2.7422,
      "step": 303
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4555151131503242,
      "learning_rate": 0.0009690795997636015,
      "loss": 2.7441,
      "step": 304
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.40538208184585517,
      "learning_rate": 0.0009688219588706179,
      "loss": 2.6523,
      "step": 305
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.46343698686531853,
      "learning_rate": 0.0009685632835815518,
      "loss": 2.623,
      "step": 306
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4782340239901532,
      "learning_rate": 0.0009683035744671367,
      "loss": 2.6094,
      "step": 307
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.436492527130298,
      "learning_rate": 0.0009680428321003883,
      "loss": 2.668,
      "step": 308
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4664747054011712,
      "learning_rate": 0.000967781057056601,
      "loss": 2.5566,
      "step": 309
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.44353165258322413,
      "learning_rate": 0.0009675182499133485,
      "loss": 2.627,
      "step": 310
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4232080516655838,
      "learning_rate": 0.0009672544112504813,
      "loss": 2.6426,
      "step": 311
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4288980876850774,
      "learning_rate": 0.0009669895416501257,
      "loss": 2.7773,
      "step": 312
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4489172434835154,
      "learning_rate": 0.0009667236416966833,
      "loss": 2.6367,
      "step": 313
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.44048053056528597,
      "learning_rate": 0.0009664567119768281,
      "loss": 2.5332,
      "step": 314
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4045250459772477,
      "learning_rate": 0.0009661887530795067,
      "loss": 2.5801,
      "step": 315
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4788536626991677,
      "learning_rate": 0.0009659197655959365,
      "loss": 2.7012,
      "step": 316
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.3797595285959082,
      "learning_rate": 0.000965649750119604,
      "loss": 2.625,
      "step": 317
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4156859656995834,
      "learning_rate": 0.0009653787072462643,
      "loss": 2.6875,
      "step": 318
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4064331764679993,
      "learning_rate": 0.0009651066375739388,
      "loss": 2.6758,
      "step": 319
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.41068414665418435,
      "learning_rate": 0.000964833541702915,
      "loss": 2.6855,
      "step": 320
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.44790797270049776,
      "learning_rate": 0.0009645594202357438,
      "loss": 2.6914,
      "step": 321
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.44771742893152927,
      "learning_rate": 0.0009642842737772397,
      "loss": 2.5332,
      "step": 322
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4778766551041836,
      "learning_rate": 0.0009640081029344782,
      "loss": 2.6523,
      "step": 323
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4380406754497066,
      "learning_rate": 0.0009637309083167956,
      "loss": 2.5645,
      "step": 324
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.40397310574373657,
      "learning_rate": 0.0009634526905357859,
      "loss": 2.6484,
      "step": 325
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4152965694529716,
      "learning_rate": 0.000963173450205302,
      "loss": 2.6855,
      "step": 326
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.5501751358392962,
      "learning_rate": 0.0009628931879414517,
      "loss": 2.7266,
      "step": 327
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.40345415907566123,
      "learning_rate": 0.0009626119043625983,
      "loss": 2.7461,
      "step": 328
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4960555406605979,
      "learning_rate": 0.0009623296000893582,
      "loss": 2.6172,
      "step": 329
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4294467414424045,
      "learning_rate": 0.0009620462757446,
      "loss": 2.7031,
      "step": 330
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.46031433706985664,
      "learning_rate": 0.0009617619319534428,
      "loss": 2.7402,
      "step": 331
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.44307037637450325,
      "learning_rate": 0.000961476569343255,
      "loss": 2.7188,
      "step": 332
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.43960170909537327,
      "learning_rate": 0.0009611901885436529,
      "loss": 2.6309,
      "step": 333
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4744631758316367,
      "learning_rate": 0.0009609027901864996,
      "loss": 2.6484,
      "step": 334
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.5140396857910516,
      "learning_rate": 0.0009606143749059029,
      "loss": 2.7461,
      "step": 335
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4151236686302008,
      "learning_rate": 0.0009603249433382144,
      "loss": 2.6328,
      "step": 336
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.44684295205145746,
      "learning_rate": 0.0009600344961220282,
      "loss": 2.6016,
      "step": 337
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4417509215765153,
      "learning_rate": 0.0009597430338981791,
      "loss": 2.627,
      "step": 338
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4135391183933795,
      "learning_rate": 0.0009594505573097414,
      "loss": 2.6992,
      "step": 339
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.41851925637647763,
      "learning_rate": 0.0009591570670020277,
      "loss": 2.6309,
      "step": 340
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.5025700577292715,
      "learning_rate": 0.0009588625636225871,
      "loss": 2.6426,
      "step": 341
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4064042591983844,
      "learning_rate": 0.0009585670478212036,
      "loss": 2.6836,
      "step": 342
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4767747841284965,
      "learning_rate": 0.0009582705202498956,
      "loss": 2.7051,
      "step": 343
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4785262665003249,
      "learning_rate": 0.0009579729815629133,
      "loss": 2.6836,
      "step": 344
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4193758243013917,
      "learning_rate": 0.0009576744324167379,
      "loss": 2.5508,
      "step": 345
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.39206420241113393,
      "learning_rate": 0.0009573748734700804,
      "loss": 2.6484,
      "step": 346
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4695763202072409,
      "learning_rate": 0.0009570743053838796,
      "loss": 2.6035,
      "step": 347
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4388037286852209,
      "learning_rate": 0.0009567727288213005,
      "loss": 2.7539,
      "step": 348
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.433510318627663,
      "learning_rate": 0.0009564701444477337,
      "loss": 2.6465,
      "step": 349
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.40631769355921354,
      "learning_rate": 0.000956166552930793,
      "loss": 2.6562,
      "step": 350
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.45244838444503105,
      "learning_rate": 0.0009558619549403147,
      "loss": 2.6895,
      "step": 351
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.39998980400225326,
      "learning_rate": 0.0009555563511483555,
      "loss": 2.6816,
      "step": 352
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.41550898574424705,
      "learning_rate": 0.0009552497422291912,
      "loss": 2.8047,
      "step": 353
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4329333125151939,
      "learning_rate": 0.0009549421288593157,
      "loss": 2.6133,
      "step": 354
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.39621132866699704,
      "learning_rate": 0.0009546335117174385,
      "loss": 2.6836,
      "step": 355
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4129284382737315,
      "learning_rate": 0.0009543238914844843,
      "loss": 2.6191,
      "step": 356
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.44095517787123767,
      "learning_rate": 0.0009540132688435907,
      "loss": 2.6602,
      "step": 357
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.3703054727107483,
      "learning_rate": 0.0009537016444801074,
      "loss": 2.5723,
      "step": 358
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.3756656384047659,
      "learning_rate": 0.0009533890190815935,
      "loss": 2.6172,
      "step": 359
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.3816442553860363,
      "learning_rate": 0.0009530753933378173,
      "loss": 2.6699,
      "step": 360
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.39584264861030155,
      "learning_rate": 0.0009527607679407545,
      "loss": 2.5664,
      "step": 361
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.4071084065227687,
      "learning_rate": 0.0009524451435845857,
      "loss": 2.6133,
      "step": 362
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.43421118442001727,
      "learning_rate": 0.0009521285209656963,
      "loss": 2.668,
      "step": 363
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.37951702755621614,
      "learning_rate": 0.0009518109007826734,
      "loss": 2.5977,
      "step": 364
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.47383592824513243,
      "learning_rate": 0.0009514922837363059,
      "loss": 2.6328,
      "step": 365
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.43050605816218024,
      "learning_rate": 0.0009511726705295817,
      "loss": 2.5781,
      "step": 366
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.42591662207113423,
      "learning_rate": 0.000950852061867687,
      "loss": 2.623,
      "step": 367
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.404358061304604,
      "learning_rate": 0.0009505304584580038,
      "loss": 2.7461,
      "step": 368
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.43177141742130903,
      "learning_rate": 0.0009502078610101092,
      "loss": 2.6621,
      "step": 369
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.3949184556226651,
      "learning_rate": 0.0009498842702357736,
      "loss": 2.6543,
      "step": 370
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.37580955917584075,
      "learning_rate": 0.0009495596868489587,
      "loss": 2.6582,
      "step": 371
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.4093935207603845,
      "learning_rate": 0.0009492341115658167,
      "loss": 2.6191,
      "step": 372
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.4104858953067128,
      "learning_rate": 0.0009489075451046879,
      "loss": 2.7617,
      "step": 373
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.3982187358560299,
      "learning_rate": 0.0009485799881861,
      "loss": 2.7207,
      "step": 374
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.40110931838074504,
      "learning_rate": 0.0009482514415327654,
      "loss": 2.748,
      "step": 375
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.4297296360133799,
      "learning_rate": 0.000947921905869581,
      "loss": 2.5898,
      "step": 376
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.4110641199119469,
      "learning_rate": 0.0009475913819236248,
      "loss": 2.7227,
      "step": 377
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.4710851937805663,
      "learning_rate": 0.0009472598704241561,
      "loss": 2.6445,
      "step": 378
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.4092041134371988,
      "learning_rate": 0.0009469273721026131,
      "loss": 2.5273,
      "step": 379
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.42477085688590044,
      "learning_rate": 0.0009465938876926111,
      "loss": 2.6191,
      "step": 380
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.41191789759449793,
      "learning_rate": 0.0009462594179299406,
      "loss": 2.6113,
      "step": 381
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4112739124067909,
      "learning_rate": 0.0009459239635525672,
      "loss": 2.584,
      "step": 382
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.47494057296138303,
      "learning_rate": 0.0009455875253006281,
      "loss": 2.6465,
      "step": 383
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4223195164254889,
      "learning_rate": 0.0009452501039164315,
      "loss": 2.6992,
      "step": 384
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4236663238132524,
      "learning_rate": 0.0009449117001444549,
      "loss": 2.623,
      "step": 385
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.3870513186967413,
      "learning_rate": 0.0009445723147313433,
      "loss": 2.6133,
      "step": 386
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4089300581577955,
      "learning_rate": 0.0009442319484259074,
      "loss": 2.6152,
      "step": 387
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.3961470962576737,
      "learning_rate": 0.0009438906019791222,
      "loss": 2.6875,
      "step": 388
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.40795950148331095,
      "learning_rate": 0.0009435482761441251,
      "loss": 2.7344,
      "step": 389
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4626929002567523,
      "learning_rate": 0.000943204971676215,
      "loss": 2.584,
      "step": 390
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4142592429485853,
      "learning_rate": 0.0009428606893328493,
      "loss": 2.6328,
      "step": 391
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4353686799302308,
      "learning_rate": 0.0009425154298736432,
      "loss": 2.5879,
      "step": 392
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.43553268782775145,
      "learning_rate": 0.0009421691940603678,
      "loss": 2.6543,
      "step": 393
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.40424428790165173,
      "learning_rate": 0.0009418219826569488,
      "loss": 2.707,
      "step": 394
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.44716493335201,
      "learning_rate": 0.0009414737964294635,
      "loss": 2.5996,
      "step": 395
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.42955457449268336,
      "learning_rate": 0.000941124636146141,
      "loss": 2.6465,
      "step": 396
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.41656011424918293,
      "learning_rate": 0.0009407745025773589,
      "loss": 2.623,
      "step": 397
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.43982738287142176,
      "learning_rate": 0.0009404233964956423,
      "loss": 2.5879,
      "step": 398
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4251632830074425,
      "learning_rate": 0.0009400713186756625,
      "loss": 2.6035,
      "step": 399
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.429368504123561,
      "learning_rate": 0.0009397182698942342,
      "loss": 2.6836,
      "step": 400
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.39963224611596815,
      "learning_rate": 0.0009393642509303149,
      "loss": 2.6113,
      "step": 401
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.40374220882743844,
      "learning_rate": 0.0009390092625650023,
      "loss": 2.6582,
      "step": 402
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.38509460139249163,
      "learning_rate": 0.0009386533055815332,
      "loss": 2.5586,
      "step": 403
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.38346929752375936,
      "learning_rate": 0.0009382963807652813,
      "loss": 2.6934,
      "step": 404
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.3850844731581421,
      "learning_rate": 0.000937938488903756,
      "loss": 2.6152,
      "step": 405
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.3838204938664295,
      "learning_rate": 0.0009375796307866003,
      "loss": 2.6758,
      "step": 406
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.3986295321751038,
      "learning_rate": 0.0009372198072055888,
      "loss": 2.6113,
      "step": 407
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.4170284066367068,
      "learning_rate": 0.0009368590189546268,
      "loss": 2.5879,
      "step": 408
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.4115269773670043,
      "learning_rate": 0.0009364972668297474,
      "loss": 2.6133,
      "step": 409
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.44021284171632136,
      "learning_rate": 0.0009361345516291111,
      "loss": 2.623,
      "step": 410
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.454924314418825,
      "learning_rate": 0.0009357708741530024,
      "loss": 2.5879,
      "step": 411
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.40376716675723634,
      "learning_rate": 0.00093540623520383,
      "loss": 2.5684,
      "step": 412
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.3980269671693784,
      "learning_rate": 0.000935040635586123,
      "loss": 2.6543,
      "step": 413
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.4568681894719726,
      "learning_rate": 0.0009346740761065305,
      "loss": 2.6875,
      "step": 414
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.43209172812247404,
      "learning_rate": 0.0009343065575738197,
      "loss": 2.582,
      "step": 415
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.4501198477291117,
      "learning_rate": 0.0009339380807988733,
      "loss": 2.6738,
      "step": 416
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.4477105630604591,
      "learning_rate": 0.0009335686465946887,
      "loss": 2.5977,
      "step": 417
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.39405524135466274,
      "learning_rate": 0.0009331982557763754,
      "loss": 2.5273,
      "step": 418
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.4014363449009265,
      "learning_rate": 0.0009328269091611537,
      "loss": 2.5957,
      "step": 419
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.38274951517475075,
      "learning_rate": 0.0009324546075683524,
      "loss": 2.6211,
      "step": 420
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.40504334999806074,
      "learning_rate": 0.0009320813518194083,
      "loss": 2.6621,
      "step": 421
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.41908611523257955,
      "learning_rate": 0.0009317071427378624,
      "loss": 2.5918,
      "step": 422
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.3904118432043913,
      "learning_rate": 0.0009313319811493594,
      "loss": 2.6621,
      "step": 423
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.39958497598815507,
      "learning_rate": 0.000930955867881646,
      "loss": 2.6074,
      "step": 424
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.49344090520170075,
      "learning_rate": 0.0009305788037645681,
      "loss": 2.5938,
      "step": 425
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.43403746171750995,
      "learning_rate": 0.0009302007896300697,
      "loss": 2.5898,
      "step": 426
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.9999039390086278,
      "learning_rate": 0.0009298218263121911,
      "loss": 2.5703,
      "step": 427
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.4179386362447804,
      "learning_rate": 0.0009294419146470668,
      "loss": 2.625,
      "step": 428
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.38597343573793974,
      "learning_rate": 0.0009290610554729234,
      "loss": 2.6875,
      "step": 429
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.4045393659362929,
      "learning_rate": 0.0009286792496300784,
      "loss": 2.6113,
      "step": 430
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.4060785214364889,
      "learning_rate": 0.0009282964979609379,
      "loss": 2.6406,
      "step": 431
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.40443022041827126,
      "learning_rate": 0.0009279128013099947,
      "loss": 2.6016,
      "step": 432
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.40402122602320395,
      "learning_rate": 0.0009275281605238268,
      "loss": 2.6914,
      "step": 433
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.9064899328130556,
      "learning_rate": 0.0009271425764510953,
      "loss": 2.5742,
      "step": 434
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.4139302206622603,
      "learning_rate": 0.0009267560499425423,
      "loss": 2.5293,
      "step": 435
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.45587300009129617,
      "learning_rate": 0.0009263685818509895,
      "loss": 2.709,
      "step": 436
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.4080093244721587,
      "learning_rate": 0.000925980173031336,
      "loss": 2.6348,
      "step": 437
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.3696711052288375,
      "learning_rate": 0.0009255908243405567,
      "loss": 2.6074,
      "step": 438
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.3870937818732974,
      "learning_rate": 0.0009252005366376996,
      "loss": 2.5898,
      "step": 439
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.38354035744479426,
      "learning_rate": 0.0009248093107838852,
      "loss": 2.543,
      "step": 440
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.36833745004149565,
      "learning_rate": 0.0009244171476423036,
      "loss": 2.5918,
      "step": 441
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.39515160058009097,
      "learning_rate": 0.0009240240480782129,
      "loss": 2.5879,
      "step": 442
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.40025070841542854,
      "learning_rate": 0.0009236300129589376,
      "loss": 2.7383,
      "step": 443
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.3856065981868966,
      "learning_rate": 0.0009232350431538657,
      "loss": 2.6426,
      "step": 444
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.39129236404000617,
      "learning_rate": 0.0009228391395344482,
      "loss": 2.5156,
      "step": 445
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.41353822808243396,
      "learning_rate": 0.000922442302974196,
      "loss": 2.6055,
      "step": 446
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.42193795677872076,
      "learning_rate": 0.0009220445343486785,
      "loss": 2.6641,
      "step": 447
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.4409629775966313,
      "learning_rate": 0.0009216458345355217,
      "loss": 2.666,
      "step": 448
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.42214385897132534,
      "learning_rate": 0.0009212462044144061,
      "loss": 2.6484,
      "step": 449
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.434610603439064,
      "learning_rate": 0.0009208456448670648,
      "loss": 2.6152,
      "step": 450
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.4329803838805059,
      "learning_rate": 0.0009204441567772816,
      "loss": 2.5859,
      "step": 451
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5383685669688083,
      "learning_rate": 0.0009200417410308888,
      "loss": 2.6543,
      "step": 452
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.3778278457883534,
      "learning_rate": 0.0009196383985157656,
      "loss": 2.5449,
      "step": 453
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.4085343977162607,
      "learning_rate": 0.000919234130121836,
      "loss": 2.6289,
      "step": 454
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5006561102577484,
      "learning_rate": 0.0009188289367410672,
      "loss": 2.6074,
      "step": 455
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.40381916114767386,
      "learning_rate": 0.0009184228192674666,
      "loss": 2.6309,
      "step": 456
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.43568170295798747,
      "learning_rate": 0.0009180157785970808,
      "loss": 2.6016,
      "step": 457
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.48539511265267726,
      "learning_rate": 0.0009176078156279932,
      "loss": 2.6758,
      "step": 458
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.4168530791614123,
      "learning_rate": 0.0009171989312603226,
      "loss": 2.6211,
      "step": 459
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.4367528778196279,
      "learning_rate": 0.0009167891263962202,
      "loss": 2.5293,
      "step": 460
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.4396744218245187,
      "learning_rate": 0.0009163784019398685,
      "loss": 2.4688,
      "step": 461
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.37954752428137933,
      "learning_rate": 0.0009159667587974785,
      "loss": 2.6543,
      "step": 462
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.3970158351574394,
      "learning_rate": 0.0009155541978772887,
      "loss": 2.6387,
      "step": 463
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.46967947655995257,
      "learning_rate": 0.0009151407200895625,
      "loss": 2.625,
      "step": 464
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.4106900306741828,
      "learning_rate": 0.000914726326346586,
      "loss": 2.5703,
      "step": 465
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.4010869920628895,
      "learning_rate": 0.0009143110175626661,
      "loss": 2.5684,
      "step": 466
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.4128326050737838,
      "learning_rate": 0.0009138947946541291,
      "loss": 2.5703,
      "step": 467
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.42927829665731243,
      "learning_rate": 0.0009134776585393181,
      "loss": 2.6133,
      "step": 468
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.40195538152472804,
      "learning_rate": 0.0009130596101385906,
      "loss": 2.5645,
      "step": 469
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.41765938446940376,
      "learning_rate": 0.0009126406503743174,
      "loss": 2.5371,
      "step": 470
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.4438858362679348,
      "learning_rate": 0.0009122207801708802,
      "loss": 2.5781,
      "step": 471
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.3996366588332302,
      "learning_rate": 0.0009118000004546689,
      "loss": 2.6289,
      "step": 472
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.38866996618651595,
      "learning_rate": 0.0009113783121540807,
      "loss": 2.5762,
      "step": 473
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.40511191844973493,
      "learning_rate": 0.0009109557161995172,
      "loss": 2.5703,
      "step": 474
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.4300546238705736,
      "learning_rate": 0.0009105322135233828,
      "loss": 2.5938,
      "step": 475
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.42724840565464667,
      "learning_rate": 0.0009101078050600821,
      "loss": 2.6641,
      "step": 476
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.4235296746377064,
      "learning_rate": 0.0009096824917460186,
      "loss": 2.5645,
      "step": 477
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.37615620057369303,
      "learning_rate": 0.0009092562745195921,
      "loss": 2.5645,
      "step": 478
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.37005775772349253,
      "learning_rate": 0.0009088291543211967,
      "loss": 2.6445,
      "step": 479
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.39117983507479687,
      "learning_rate": 0.0009084011320932188,
      "loss": 2.5938,
      "step": 480
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.3836914390985446,
      "learning_rate": 0.0009079722087800352,
      "loss": 2.6113,
      "step": 481
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.3950040692151942,
      "learning_rate": 0.0009075423853280106,
      "loss": 2.625,
      "step": 482
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.38303715643858033,
      "learning_rate": 0.0009071116626854958,
      "loss": 2.6523,
      "step": 483
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.38666316712019577,
      "learning_rate": 0.0009066800418028256,
      "loss": 2.5918,
      "step": 484
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.43771191762679595,
      "learning_rate": 0.0009062475236323168,
      "loss": 2.5918,
      "step": 485
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.4094287985606332,
      "learning_rate": 0.0009058141091282656,
      "loss": 2.5352,
      "step": 486
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.4374898911084178,
      "learning_rate": 0.0009053797992469461,
      "loss": 2.5996,
      "step": 487
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.3879396141587866,
      "learning_rate": 0.0009049445949466078,
      "loss": 2.6309,
      "step": 488
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.40457932128857094,
      "learning_rate": 0.0009045084971874737,
      "loss": 2.6406,
      "step": 489
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.40579195597058637,
      "learning_rate": 0.0009040715069317382,
      "loss": 2.6484,
      "step": 490
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.417607509861666,
      "learning_rate": 0.0009036336251435648,
      "loss": 2.5547,
      "step": 491
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.41772573109737576,
      "learning_rate": 0.0009031948527890839,
      "loss": 2.5996,
      "step": 492
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4027452233733011,
      "learning_rate": 0.000902755190836391,
      "loss": 2.6309,
      "step": 493
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4082791823176778,
      "learning_rate": 0.0009023146402555442,
      "loss": 2.5312,
      "step": 494
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4006320792299315,
      "learning_rate": 0.0009018732020185624,
      "loss": 2.6074,
      "step": 495
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.43023935792929663,
      "learning_rate": 0.0009014308770994235,
      "loss": 2.6426,
      "step": 496
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.39653772828439837,
      "learning_rate": 0.0009009876664740605,
      "loss": 2.5645,
      "step": 497
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.40423640250112547,
      "learning_rate": 0.0009005435711203618,
      "loss": 2.6426,
      "step": 498
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4022248877974805,
      "learning_rate": 0.000900098592018167,
      "loss": 2.5586,
      "step": 499
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.41467066487164456,
      "learning_rate": 0.0008996527301492663,
      "loss": 2.6484,
      "step": 500
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.420044220591984,
      "learning_rate": 0.0008992059864973972,
      "loss": 2.5977,
      "step": 501
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.39390914234951374,
      "learning_rate": 0.0008987583620482427,
      "loss": 2.6211,
      "step": 502
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.39320427571359157,
      "learning_rate": 0.0008983098577894292,
      "loss": 2.5312,
      "step": 503
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.40050704629286554,
      "learning_rate": 0.0008978604747105246,
      "loss": 2.623,
      "step": 504
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.39553439196708545,
      "learning_rate": 0.0008974102138030354,
      "loss": 2.4707,
      "step": 505
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4038155644782662,
      "learning_rate": 0.000896959076060405,
      "loss": 2.6504,
      "step": 506
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.3911301989667696,
      "learning_rate": 0.0008965070624780116,
      "loss": 2.5547,
      "step": 507
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.3897152234865207,
      "learning_rate": 0.0008960541740531658,
      "loss": 2.5156,
      "step": 508
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.38281189395560977,
      "learning_rate": 0.0008956004117851083,
      "loss": 2.582,
      "step": 509
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4385339763998877,
      "learning_rate": 0.0008951457766750079,
      "loss": 2.5547,
      "step": 510
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.3948627973713726,
      "learning_rate": 0.0008946902697259593,
      "loss": 2.625,
      "step": 511
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.40252734372945903,
      "learning_rate": 0.0008942338919429805,
      "loss": 2.5938,
      "step": 512
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4338926051778563,
      "learning_rate": 0.0008937766443330113,
      "loss": 2.5859,
      "step": 513
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.38275885183243835,
      "learning_rate": 0.0008933185279049103,
      "loss": 2.5137,
      "step": 514
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.41498018772815753,
      "learning_rate": 0.0008928595436694532,
      "loss": 2.5352,
      "step": 515
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.41848996419800016,
      "learning_rate": 0.0008923996926393305,
      "loss": 2.6582,
      "step": 516
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.3992518304495277,
      "learning_rate": 0.0008919389758291449,
      "loss": 2.6445,
      "step": 517
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.41261684864031717,
      "learning_rate": 0.0008914773942554098,
      "loss": 2.5605,
      "step": 518
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.3912661282935942,
      "learning_rate": 0.000891014948936546,
      "loss": 2.6367,
      "step": 519
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4397473165131659,
      "learning_rate": 0.0008905516408928804,
      "loss": 2.5469,
      "step": 520
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4248143384442833,
      "learning_rate": 0.0008900874711466434,
      "loss": 2.6113,
      "step": 521
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.3690066110089538,
      "learning_rate": 0.0008896224407219666,
      "loss": 2.5391,
      "step": 522
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4076639663337741,
      "learning_rate": 0.0008891565506448804,
      "loss": 2.6035,
      "step": 523
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.37035627057545734,
      "learning_rate": 0.0008886898019433122,
      "loss": 2.5156,
      "step": 524
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.36463150724566507,
      "learning_rate": 0.0008882221956470836,
      "loss": 2.5215,
      "step": 525
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4174200414752731,
      "learning_rate": 0.0008877537327879086,
      "loss": 2.5117,
      "step": 526
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.38701838402299077,
      "learning_rate": 0.0008872844143993908,
      "loss": 2.5488,
      "step": 527
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4089687557400908,
      "learning_rate": 0.0008868142415170218,
      "loss": 2.5098,
      "step": 528
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4533327206921066,
      "learning_rate": 0.0008863432151781781,
      "loss": 2.6016,
      "step": 529
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.38947522760298475,
      "learning_rate": 0.0008858713364221195,
      "loss": 2.4258,
      "step": 530
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.38978723845667845,
      "learning_rate": 0.0008853986062899868,
      "loss": 2.5762,
      "step": 531
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.3843891328146601,
      "learning_rate": 0.0008849250258247986,
      "loss": 2.5391,
      "step": 532
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.44386052686350497,
      "learning_rate": 0.0008844505960714503,
      "loss": 2.5957,
      "step": 533
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.42158027177299473,
      "learning_rate": 0.0008839753180767108,
      "loss": 2.6172,
      "step": 534
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.43062026058996195,
      "learning_rate": 0.0008834991928892204,
      "loss": 2.582,
      "step": 535
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.39971373368295715,
      "learning_rate": 0.000883022221559489,
      "loss": 2.5312,
      "step": 536
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.36197289677905614,
      "learning_rate": 0.0008825444051398934,
      "loss": 2.668,
      "step": 537
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.3761037621044644,
      "learning_rate": 0.0008820657446846745,
      "loss": 2.6348,
      "step": 538
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.4612399993774412,
      "learning_rate": 0.000881586241249936,
      "loss": 2.6289,
      "step": 539
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.3786936554812936,
      "learning_rate": 0.0008811058958936411,
      "loss": 2.5547,
      "step": 540
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.3844396727435307,
      "learning_rate": 0.000880624709675611,
      "loss": 2.5996,
      "step": 541
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.3846273940930797,
      "learning_rate": 0.000880142683657522,
      "loss": 2.5645,
      "step": 542
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.3934699538884206,
      "learning_rate": 0.0008796598189029029,
      "loss": 2.6309,
      "step": 543
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.3742290435852469,
      "learning_rate": 0.0008791761164771338,
      "loss": 2.582,
      "step": 544
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.4655492854344016,
      "learning_rate": 0.0008786915774474424,
      "loss": 2.5762,
      "step": 545
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.3872527495978308,
      "learning_rate": 0.0008782062028829027,
      "loss": 2.6152,
      "step": 546
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.3880325874629398,
      "learning_rate": 0.0008777199938544318,
      "loss": 2.627,
      "step": 547
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.3937507689198629,
      "learning_rate": 0.0008772329514347883,
      "loss": 2.5195,
      "step": 548
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.38145803387841704,
      "learning_rate": 0.0008767450766985694,
      "loss": 2.6055,
      "step": 549
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.38428568475012836,
      "learning_rate": 0.0008762563707222086,
      "loss": 2.5723,
      "step": 550
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.40101610014395617,
      "learning_rate": 0.0008757668345839738,
      "loss": 2.6191,
      "step": 551
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.40216498677266843,
      "learning_rate": 0.0008752764693639638,
      "loss": 2.6465,
      "step": 552
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.3749997121436037,
      "learning_rate": 0.0008747852761441078,
      "loss": 2.4922,
      "step": 553
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.38621358896364216,
      "learning_rate": 0.0008742932560081607,
      "loss": 2.5664,
      "step": 554
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.36867914323468914,
      "learning_rate": 0.0008738004100417025,
      "loss": 2.5547,
      "step": 555
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.37470345289905543,
      "learning_rate": 0.0008733067393321355,
      "loss": 2.5898,
      "step": 556
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.4669160539250825,
      "learning_rate": 0.000872812244968681,
      "loss": 2.6602,
      "step": 557
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.3816902348821588,
      "learning_rate": 0.0008723169280423783,
      "loss": 2.5918,
      "step": 558
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.37160190722264214,
      "learning_rate": 0.0008718207896460811,
      "loss": 2.5996,
      "step": 559
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.38888346559941933,
      "learning_rate": 0.0008713238308744557,
      "loss": 2.5723,
      "step": 560
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.533024336547213,
      "learning_rate": 0.0008708260528239789,
      "loss": 2.5156,
      "step": 561
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.39797024338566395,
      "learning_rate": 0.000870327456592934,
      "loss": 2.6309,
      "step": 562
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.39574126727622766,
      "learning_rate": 0.0008698280432814107,
      "loss": 2.5645,
      "step": 563
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.4069380545789327,
      "learning_rate": 0.000869327813991301,
      "loss": 2.5449,
      "step": 564
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.40841446443837287,
      "learning_rate": 0.0008688267698262971,
      "loss": 2.5195,
      "step": 565
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.38864688614745796,
      "learning_rate": 0.0008683249118918894,
      "loss": 2.5488,
      "step": 566
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.43529146082780773,
      "learning_rate": 0.0008678222412953637,
      "loss": 2.5547,
      "step": 567
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.3826823851623577,
      "learning_rate": 0.0008673187591457987,
      "loss": 2.5254,
      "step": 568
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.3938829932684149,
      "learning_rate": 0.0008668144665540639,
      "loss": 2.5859,
      "step": 569
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.43167521200759856,
      "learning_rate": 0.0008663093646328167,
      "loss": 2.5273,
      "step": 570
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.41042107429837743,
      "learning_rate": 0.0008658034544965003,
      "loss": 2.5078,
      "step": 571
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.3816724236409058,
      "learning_rate": 0.0008652967372613412,
      "loss": 2.6641,
      "step": 572
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.37516647118610935,
      "learning_rate": 0.0008647892140453466,
      "loss": 2.6094,
      "step": 573
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.3813088929443776,
      "learning_rate": 0.0008642808859683021,
      "loss": 2.5195,
      "step": 574
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.4094356123659152,
      "learning_rate": 0.0008637717541517689,
      "loss": 2.6133,
      "step": 575
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.4147293037234312,
      "learning_rate": 0.0008632618197190816,
      "loss": 2.625,
      "step": 576
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.4151940690923111,
      "learning_rate": 0.0008627510837953458,
      "loss": 2.6289,
      "step": 577
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.41950690931263934,
      "learning_rate": 0.0008622395475074355,
      "loss": 2.5078,
      "step": 578
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.39188863839909893,
      "learning_rate": 0.0008617272119839903,
      "loss": 2.5547,
      "step": 579
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.37230177693597133,
      "learning_rate": 0.0008612140783554136,
      "loss": 2.5859,
      "step": 580
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.3833684495553384,
      "learning_rate": 0.0008607001477538696,
      "loss": 2.4805,
      "step": 581
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.4038715764249969,
      "learning_rate": 0.0008601854213132807,
      "loss": 2.5664,
      "step": 582
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.411184075351096,
      "learning_rate": 0.0008596699001693256,
      "loss": 2.541,
      "step": 583
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.42566670963446934,
      "learning_rate": 0.000859153585459436,
      "loss": 2.6074,
      "step": 584
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.40219919387152075,
      "learning_rate": 0.0008586364783227949,
      "loss": 2.5332,
      "step": 585
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.4092075083380097,
      "learning_rate": 0.0008581185799003332,
      "loss": 2.5605,
      "step": 586
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.42008839787470037,
      "learning_rate": 0.0008575998913347283,
      "loss": 2.5332,
      "step": 587
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.40578818762085533,
      "learning_rate": 0.0008570804137704004,
      "loss": 2.5,
      "step": 588
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.40324593621967675,
      "learning_rate": 0.0008565601483535108,
      "loss": 2.5215,
      "step": 589
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.4719930648954397,
      "learning_rate": 0.0008560390962319591,
      "loss": 2.5234,
      "step": 590
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.45987101750834264,
      "learning_rate": 0.0008555172585553804,
      "loss": 2.5508,
      "step": 591
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.4210935421743659,
      "learning_rate": 0.0008549946364751435,
      "loss": 2.5723,
      "step": 592
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5308077899345258,
      "learning_rate": 0.0008544712311443475,
      "loss": 2.5742,
      "step": 593
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.37895119737811317,
      "learning_rate": 0.0008539470437178196,
      "loss": 2.5117,
      "step": 594
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.41819477407411776,
      "learning_rate": 0.000853422075352113,
      "loss": 2.6875,
      "step": 595
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.4160306674502699,
      "learning_rate": 0.0008528963272055035,
      "loss": 2.6113,
      "step": 596
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.384511729601091,
      "learning_rate": 0.0008523698004379877,
      "loss": 2.6387,
      "step": 597
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.40407651128275585,
      "learning_rate": 0.00085184249621128,
      "loss": 2.5781,
      "step": 598
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.40612649078905383,
      "learning_rate": 0.0008513144156888101,
      "loss": 2.5859,
      "step": 599
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.3885578722974638,
      "learning_rate": 0.0008507855600357207,
      "loss": 2.4805,
      "step": 600
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.3852546136484553,
      "learning_rate": 0.0008502559304188644,
      "loss": 2.5391,
      "step": 601
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.46757324823905294,
      "learning_rate": 0.0008497255280068019,
      "loss": 2.6309,
      "step": 602
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.42494877894073096,
      "learning_rate": 0.0008491943539697986,
      "loss": 2.5742,
      "step": 603
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.39227663617577496,
      "learning_rate": 0.0008486624094798226,
      "loss": 2.6094,
      "step": 604
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.40599486476769114,
      "learning_rate": 0.0008481296957105417,
      "loss": 2.6855,
      "step": 605
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.39832212581027243,
      "learning_rate": 0.0008475962138373213,
      "loss": 2.6074,
      "step": 606
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.40938565527504733,
      "learning_rate": 0.0008470619650372211,
      "loss": 2.5938,
      "step": 607
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.3860021547752974,
      "learning_rate": 0.0008465269504889934,
      "loss": 2.5215,
      "step": 608
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.4271237342892912,
      "learning_rate": 0.0008459911713730799,
      "loss": 2.6992,
      "step": 609
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.374168743879086,
      "learning_rate": 0.0008454546288716089,
      "loss": 2.541,
      "step": 610
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.37728270370155564,
      "learning_rate": 0.0008449173241683935,
      "loss": 2.5254,
      "step": 611
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.42162921511882245,
      "learning_rate": 0.0008443792584489281,
      "loss": 2.5273,
      "step": 612
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.40479292900930874,
      "learning_rate": 0.0008438404329003863,
      "loss": 2.5234,
      "step": 613
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.4021903630743752,
      "learning_rate": 0.0008433008487116183,
      "loss": 2.5156,
      "step": 614
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.41243349857170164,
      "learning_rate": 0.0008427605070731481,
      "loss": 2.6582,
      "step": 615
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.39783318318369204,
      "learning_rate": 0.0008422194091771708,
      "loss": 2.4688,
      "step": 616
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.41059492861404434,
      "learning_rate": 0.0008416775562175503,
      "loss": 2.5352,
      "step": 617
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.406683256716611,
      "learning_rate": 0.000841134949389816,
      "loss": 2.5137,
      "step": 618
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.4073380387305802,
      "learning_rate": 0.0008405915898911611,
      "loss": 2.623,
      "step": 619
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.3812654509818732,
      "learning_rate": 0.0008400474789204396,
      "loss": 2.5625,
      "step": 620
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.3575111612548493,
      "learning_rate": 0.0008395026176781626,
      "loss": 2.5605,
      "step": 621
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.36170553922441506,
      "learning_rate": 0.0008389570073664976,
      "loss": 2.5703,
      "step": 622
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.37744700819684684,
      "learning_rate": 0.0008384106491892642,
      "loss": 2.4961,
      "step": 623
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.445430041185866,
      "learning_rate": 0.0008378635443519327,
      "loss": 2.5176,
      "step": 624
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.3589391437747151,
      "learning_rate": 0.0008373156940616199,
      "loss": 2.5254,
      "step": 625
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.3772099710765918,
      "learning_rate": 0.0008367670995270882,
      "loss": 2.5332,
      "step": 626
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.4375148874037196,
      "learning_rate": 0.0008362177619587416,
      "loss": 2.6406,
      "step": 627
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.40010983318596777,
      "learning_rate": 0.0008356676825686238,
      "loss": 2.5215,
      "step": 628
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.38753174477041447,
      "learning_rate": 0.0008351168625704147,
      "loss": 2.6562,
      "step": 629
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.47495125797971377,
      "learning_rate": 0.0008345653031794292,
      "loss": 2.5918,
      "step": 630
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.38106251519892015,
      "learning_rate": 0.0008340130056126125,
      "loss": 2.6406,
      "step": 631
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.4011532518698315,
      "learning_rate": 0.0008334599710885394,
      "loss": 2.6387,
      "step": 632
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.4009146521971397,
      "learning_rate": 0.0008329062008274098,
      "loss": 2.5977,
      "step": 633
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.4041572979327238,
      "learning_rate": 0.000832351696051048,
      "loss": 2.5391,
      "step": 634
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.38989410981580286,
      "learning_rate": 0.000831796457982898,
      "loss": 2.4961,
      "step": 635
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.42692183171733994,
      "learning_rate": 0.0008312404878480222,
      "loss": 2.5957,
      "step": 636
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.38381783222729543,
      "learning_rate": 0.0008306837868730979,
      "loss": 2.4902,
      "step": 637
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.38831169575836727,
      "learning_rate": 0.0008301263562864152,
      "loss": 2.6035,
      "step": 638
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.3845318306181912,
      "learning_rate": 0.0008295681973178737,
      "loss": 2.4941,
      "step": 639
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.4200849007315366,
      "learning_rate": 0.0008290093111989804,
      "loss": 2.582,
      "step": 640
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.38001421223219034,
      "learning_rate": 0.0008284496991628465,
      "loss": 2.4648,
      "step": 641
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.37403323677207806,
      "learning_rate": 0.0008278893624441847,
      "loss": 2.5801,
      "step": 642
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.4066934477143553,
      "learning_rate": 0.000827328302279307,
      "loss": 2.5742,
      "step": 643
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.41086214542894267,
      "learning_rate": 0.0008267665199061211,
      "loss": 2.5176,
      "step": 644
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.39229736660933207,
      "learning_rate": 0.0008262040165641288,
      "loss": 2.6367,
      "step": 645
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.3797836128988465,
      "learning_rate": 0.0008256407934944219,
      "loss": 2.5215,
      "step": 646
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.3804673513352742,
      "learning_rate": 0.0008250768519396807,
      "loss": 2.5234,
      "step": 647
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.3821292286861633,
      "learning_rate": 0.0008245121931441706,
      "loss": 2.5527,
      "step": 648
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.37573588369217326,
      "learning_rate": 0.0008239468183537393,
      "loss": 2.5254,
      "step": 649
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.39349421996964284,
      "learning_rate": 0.0008233807288158146,
      "loss": 2.5742,
      "step": 650
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.3816156126300987,
      "learning_rate": 0.0008228139257794012,
      "loss": 2.5234,
      "step": 651
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.4316678385845775,
      "learning_rate": 0.0008222464104950778,
      "loss": 2.6426,
      "step": 652
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.3842913698226245,
      "learning_rate": 0.000821678184214995,
      "loss": 2.541,
      "step": 653
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.37364409675853094,
      "learning_rate": 0.0008211092481928716,
      "loss": 2.5645,
      "step": 654
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.4099744981942454,
      "learning_rate": 0.0008205396036839927,
      "loss": 2.6074,
      "step": 655
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.39419124688331386,
      "learning_rate": 0.0008199692519452069,
      "loss": 2.5625,
      "step": 656
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.3825493813863763,
      "learning_rate": 0.0008193981942349224,
      "loss": 2.5898,
      "step": 657
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.4550711420576386,
      "learning_rate": 0.0008188264318131056,
      "loss": 2.5312,
      "step": 658
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.38936437716846206,
      "learning_rate": 0.0008182539659412776,
      "loss": 2.5332,
      "step": 659
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.38239292169815303,
      "learning_rate": 0.0008176807978825118,
      "loss": 2.541,
      "step": 660
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.3585433117391254,
      "learning_rate": 0.0008171069289014306,
      "loss": 2.5918,
      "step": 661
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.36928213655508557,
      "learning_rate": 0.0008165323602642028,
      "loss": 2.502,
      "step": 662
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.37805927227264013,
      "learning_rate": 0.0008159570932385414,
      "loss": 2.5469,
      "step": 663
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.36653601034980765,
      "learning_rate": 0.0008153811290936999,
      "loss": 2.5781,
      "step": 664
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.3697101529393566,
      "learning_rate": 0.0008148044691004698,
      "loss": 2.5449,
      "step": 665
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.3651862498042761,
      "learning_rate": 0.0008142271145311783,
      "loss": 2.5117,
      "step": 666
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.36167780385036474,
      "learning_rate": 0.000813649066659685,
      "loss": 2.6074,
      "step": 667
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.3761817005581069,
      "learning_rate": 0.0008130703267613787,
      "loss": 2.6133,
      "step": 668
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.3818777455066434,
      "learning_rate": 0.0008124908961131759,
      "loss": 2.5293,
      "step": 669
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.40207653567175106,
      "learning_rate": 0.0008119107759935163,
      "loss": 2.5527,
      "step": 670
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.36877188173439207,
      "learning_rate": 0.0008113299676823615,
      "loss": 2.541,
      "step": 671
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.3632022104494877,
      "learning_rate": 0.0008107484724611911,
      "loss": 2.5547,
      "step": 672
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.3693115637103313,
      "learning_rate": 0.0008101662916130006,
      "loss": 2.5234,
      "step": 673
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.3744364558550473,
      "learning_rate": 0.0008095834264222979,
      "loss": 2.457,
      "step": 674
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.3991438105774777,
      "learning_rate": 0.0008089998781751009,
      "loss": 2.5762,
      "step": 675
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.4006894421635839,
      "learning_rate": 0.0008084156481589349,
      "loss": 2.5938,
      "step": 676
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.38491306401293696,
      "learning_rate": 0.0008078307376628291,
      "loss": 2.5957,
      "step": 677
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.3867166423830485,
      "learning_rate": 0.0008072451479773143,
      "loss": 2.4805,
      "step": 678
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.3851661531949921,
      "learning_rate": 0.0008066588803944195,
      "loss": 2.5684,
      "step": 679
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.39166812508653603,
      "learning_rate": 0.0008060719362076697,
      "loss": 2.5293,
      "step": 680
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.3966809847675868,
      "learning_rate": 0.0008054843167120826,
      "loss": 2.5215,
      "step": 681
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.40483852815154336,
      "learning_rate": 0.0008048960232041663,
      "loss": 2.5762,
      "step": 682
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.3963331767529512,
      "learning_rate": 0.0008043070569819153,
      "loss": 2.5215,
      "step": 683
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.3803671492257606,
      "learning_rate": 0.0008037174193448089,
      "loss": 2.4473,
      "step": 684
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.38328764917268165,
      "learning_rate": 0.0008031271115938077,
      "loss": 2.5527,
      "step": 685
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.4164623080266839,
      "learning_rate": 0.0008025361350313505,
      "loss": 2.5605,
      "step": 686
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.4596082299111795,
      "learning_rate": 0.0008019444909613523,
      "loss": 2.5703,
      "step": 687
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.3665676926728072,
      "learning_rate": 0.0008013521806892003,
      "loss": 2.5664,
      "step": 688
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.4376379080690397,
      "learning_rate": 0.000800759205521752,
      "loss": 2.5039,
      "step": 689
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.391639093270708,
      "learning_rate": 0.0008001655667673318,
      "loss": 2.4746,
      "step": 690
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.38154322148917114,
      "learning_rate": 0.0007995712657357279,
      "loss": 2.5938,
      "step": 691
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.40261774464967987,
      "learning_rate": 0.0007989763037381904,
      "loss": 2.5176,
      "step": 692
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.40049277741923994,
      "learning_rate": 0.0007983806820874271,
      "loss": 2.4922,
      "step": 693
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.393450580208158,
      "learning_rate": 0.0007977844020976016,
      "loss": 2.6797,
      "step": 694
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.3843374377253232,
      "learning_rate": 0.00079718746508433,
      "loss": 2.5352,
      "step": 695
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.36945920302607915,
      "learning_rate": 0.0007965898723646776,
      "loss": 2.5215,
      "step": 696
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.36660404174951533,
      "learning_rate": 0.0007959916252571573,
      "loss": 2.4258,
      "step": 697
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.4152138789239884,
      "learning_rate": 0.000795392725081725,
      "loss": 2.5195,
      "step": 698
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.3750045224363473,
      "learning_rate": 0.000794793173159778,
      "loss": 2.4707,
      "step": 699
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.40330286924925124,
      "learning_rate": 0.0007941929708141513,
      "loss": 2.5059,
      "step": 700
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.36406297025284357,
      "learning_rate": 0.0007935921193691153,
      "loss": 2.4883,
      "step": 701
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.37025438206919986,
      "learning_rate": 0.0007929906201503722,
      "loss": 2.4922,
      "step": 702
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.3877577740369658,
      "learning_rate": 0.0007923884744850536,
      "loss": 2.5625,
      "step": 703
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.38148240381156756,
      "learning_rate": 0.0007917856837017176,
      "loss": 2.5332,
      "step": 704
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.37322588990091665,
      "learning_rate": 0.0007911822491303452,
      "loss": 2.5332,
      "step": 705
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.38918620107511426,
      "learning_rate": 0.0007905781721023382,
      "loss": 2.5176,
      "step": 706
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.40864719366212615,
      "learning_rate": 0.000789973453950516,
      "loss": 2.5625,
      "step": 707
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.38908942049687306,
      "learning_rate": 0.000789368096009112,
      "loss": 2.4883,
      "step": 708
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.38619666277127124,
      "learning_rate": 0.0007887620996137721,
      "loss": 2.4551,
      "step": 709
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.3847839192199123,
      "learning_rate": 0.0007881554661015497,
      "loss": 2.5664,
      "step": 710
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.3956722223551053,
      "learning_rate": 0.0007875481968109051,
      "loss": 2.5254,
      "step": 711
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.40172249706914726,
      "learning_rate": 0.0007869402930817007,
      "loss": 2.4922,
      "step": 712
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.38140913615641103,
      "learning_rate": 0.0007863317562551987,
      "loss": 2.5156,
      "step": 713
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.39230856289367166,
      "learning_rate": 0.0007857225876740584,
      "loss": 2.6309,
      "step": 714
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.35885879252247777,
      "learning_rate": 0.0007851127886823327,
      "loss": 2.4883,
      "step": 715
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.36721512060164757,
      "learning_rate": 0.0007845023606254658,
      "loss": 2.4883,
      "step": 716
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.3871232641712994,
      "learning_rate": 0.0007838913048502894,
      "loss": 2.5,
      "step": 717
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.41567697104580253,
      "learning_rate": 0.0007832796227050208,
      "loss": 2.5859,
      "step": 718
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.38877485870689193,
      "learning_rate": 0.0007826673155392587,
      "loss": 2.5566,
      "step": 719
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.37496527752273123,
      "learning_rate": 0.000782054384703981,
      "loss": 2.5215,
      "step": 720
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.38511440045689166,
      "learning_rate": 0.0007814408315515418,
      "loss": 2.5273,
      "step": 721
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.40737137525241335,
      "learning_rate": 0.0007808266574356683,
      "loss": 2.5391,
      "step": 722
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.38700892756707167,
      "learning_rate": 0.0007802118637114573,
      "loss": 2.5918,
      "step": 723
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.39700008450951985,
      "learning_rate": 0.0007795964517353734,
      "loss": 2.5156,
      "step": 724
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.3982057179925213,
      "learning_rate": 0.0007789804228652449,
      "loss": 2.5938,
      "step": 725
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.37123994614021605,
      "learning_rate": 0.0007783637784602609,
      "loss": 2.5312,
      "step": 726
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.3737264177023338,
      "learning_rate": 0.0007777465198809692,
      "loss": 2.5059,
      "step": 727
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.3971221570908004,
      "learning_rate": 0.0007771286484892722,
      "loss": 2.5215,
      "step": 728
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.3955484867888956,
      "learning_rate": 0.000776510165648425,
      "loss": 2.5176,
      "step": 729
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.38247261640115693,
      "learning_rate": 0.0007758910727230311,
      "loss": 2.584,
      "step": 730
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3952319624046844,
      "learning_rate": 0.0007752713710790404,
      "loss": 2.418,
      "step": 731
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.398641588379862,
      "learning_rate": 0.0007746510620837459,
      "loss": 2.5117,
      "step": 732
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3667423594590183,
      "learning_rate": 0.0007740301471057807,
      "loss": 2.4902,
      "step": 733
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3596697572279915,
      "learning_rate": 0.0007734086275151146,
      "loss": 2.4668,
      "step": 734
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3764024644085095,
      "learning_rate": 0.0007727865046830517,
      "loss": 2.5195,
      "step": 735
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.37543138705782203,
      "learning_rate": 0.0007721637799822269,
      "loss": 2.5,
      "step": 736
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3733442105590562,
      "learning_rate": 0.0007715404547866032,
      "loss": 2.5527,
      "step": 737
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.38863207529289046,
      "learning_rate": 0.0007709165304714685,
      "loss": 2.5723,
      "step": 738
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.40626794912674263,
      "learning_rate": 0.0007702920084134324,
      "loss": 2.5879,
      "step": 739
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.37988028216714453,
      "learning_rate": 0.0007696668899904236,
      "loss": 2.4922,
      "step": 740
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3639490167206508,
      "learning_rate": 0.0007690411765816864,
      "loss": 2.5508,
      "step": 741
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.38581215821406994,
      "learning_rate": 0.0007684148695677778,
      "loss": 2.498,
      "step": 742
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.36802435523487,
      "learning_rate": 0.000767787970330565,
      "loss": 2.5195,
      "step": 743
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3601577166009757,
      "learning_rate": 0.000767160480253221,
      "loss": 2.5801,
      "step": 744
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.37471865883286043,
      "learning_rate": 0.0007665324007202235,
      "loss": 2.4902,
      "step": 745
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.36950609882402197,
      "learning_rate": 0.0007659037331173498,
      "loss": 2.4629,
      "step": 746
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.39191728328805586,
      "learning_rate": 0.0007652744788316752,
      "loss": 2.5723,
      "step": 747
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3897591719007878,
      "learning_rate": 0.0007646446392515692,
      "loss": 2.4863,
      "step": 748
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.38733214103054375,
      "learning_rate": 0.000764014215766693,
      "loss": 2.6543,
      "step": 749
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.44194924054743895,
      "learning_rate": 0.0007633832097679958,
      "loss": 2.5957,
      "step": 750
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3773043180063744,
      "learning_rate": 0.0007627516226477122,
      "loss": 2.4785,
      "step": 751
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.38745157900418076,
      "learning_rate": 0.0007621194557993589,
      "loss": 2.5254,
      "step": 752
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.3778383532148278,
      "learning_rate": 0.0007614867106177319,
      "loss": 2.5234,
      "step": 753
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.3798722853158081,
      "learning_rate": 0.0007608533884989029,
      "loss": 2.582,
      "step": 754
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.38318555642897995,
      "learning_rate": 0.0007602194908402166,
      "loss": 2.543,
      "step": 755
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.40422798127788384,
      "learning_rate": 0.0007595850190402877,
      "loss": 2.5098,
      "step": 756
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.38272633733924105,
      "learning_rate": 0.0007589499744989976,
      "loss": 2.498,
      "step": 757
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.37413565286660405,
      "learning_rate": 0.0007583143586174916,
      "loss": 2.5332,
      "step": 758
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.40641212917665115,
      "learning_rate": 0.000757678172798175,
      "loss": 2.5469,
      "step": 759
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.3696100176302387,
      "learning_rate": 0.0007570414184447112,
      "loss": 2.4805,
      "step": 760
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.3676817700827917,
      "learning_rate": 0.0007564040969620179,
      "loss": 2.541,
      "step": 761
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.40572607131130967,
      "learning_rate": 0.0007557662097562636,
      "loss": 2.5371,
      "step": 762
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.37612739620612146,
      "learning_rate": 0.0007551277582348658,
      "loss": 2.4922,
      "step": 763
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.39195328328237733,
      "learning_rate": 0.0007544887438064862,
      "loss": 2.5859,
      "step": 764
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.3804220991665399,
      "learning_rate": 0.0007538491678810294,
      "loss": 2.5801,
      "step": 765
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.4298400649035864,
      "learning_rate": 0.0007532090318696381,
      "loss": 2.5605,
      "step": 766
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.3680298254428115,
      "learning_rate": 0.0007525683371846913,
      "loss": 2.5098,
      "step": 767
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.4267291537063427,
      "learning_rate": 0.0007519270852398001,
      "loss": 2.5586,
      "step": 768
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.40776897634269116,
      "learning_rate": 0.000751285277449806,
      "loss": 2.625,
      "step": 769
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.47044404245650306,
      "learning_rate": 0.0007506429152307756,
      "loss": 2.6484,
      "step": 770
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.40118519969159455,
      "learning_rate": 0.00075,
      "loss": 2.5,
      "step": 771
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.3672685239318412,
      "learning_rate": 0.00074935653317599,
      "loss": 2.4902,
      "step": 772
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.39676760400264854,
      "learning_rate": 0.000748712516178473,
      "loss": 2.5059,
      "step": 773
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.40290364141907475,
      "learning_rate": 0.0007480679504283911,
      "loss": 2.5156,
      "step": 774
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.3864459252602057,
      "learning_rate": 0.0007474228373478964,
      "loss": 2.5469,
      "step": 775
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.43199035577256784,
      "learning_rate": 0.0007467771783603492,
      "loss": 2.5859,
      "step": 776
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.4478066524738304,
      "learning_rate": 0.0007461309748903138,
      "loss": 2.5645,
      "step": 777
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.38924994137436286,
      "learning_rate": 0.0007454842283635562,
      "loss": 2.668,
      "step": 778
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.43606676579019055,
      "learning_rate": 0.0007448369402070404,
      "loss": 2.4297,
      "step": 779
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.45955568682252257,
      "learning_rate": 0.0007441891118489254,
      "loss": 2.5938,
      "step": 780
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.3983341658929419,
      "learning_rate": 0.0007435407447185622,
      "loss": 2.6289,
      "step": 781
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.3946297070960736,
      "learning_rate": 0.0007428918402464908,
      "loss": 2.5352,
      "step": 782
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.38349187648494054,
      "learning_rate": 0.0007422423998644359,
      "loss": 2.5449,
      "step": 783
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.41933526071883404,
      "learning_rate": 0.0007415924250053055,
      "loss": 2.502,
      "step": 784
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.37361444094650564,
      "learning_rate": 0.0007409419171031865,
      "loss": 2.5547,
      "step": 785
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.3801589469164429,
      "learning_rate": 0.0007402908775933419,
      "loss": 2.4668,
      "step": 786
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.3674742241117112,
      "learning_rate": 0.0007396393079122077,
      "loss": 2.4941,
      "step": 787
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.39531902072380737,
      "learning_rate": 0.0007389872094973896,
      "loss": 2.4961,
      "step": 788
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.3973728043116334,
      "learning_rate": 0.00073833458378766,
      "loss": 2.6914,
      "step": 789
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.4093747000031155,
      "learning_rate": 0.0007376814322229544,
      "loss": 2.6621,
      "step": 790
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.4031764306645884,
      "learning_rate": 0.0007370277562443688,
      "loss": 2.5977,
      "step": 791
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.40290068084798253,
      "learning_rate": 0.0007363735572941564,
      "loss": 2.4688,
      "step": 792
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.3842872210669781,
      "learning_rate": 0.0007357188368157236,
      "loss": 2.5664,
      "step": 793
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.37729940867801565,
      "learning_rate": 0.0007350635962536284,
      "loss": 2.3965,
      "step": 794
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.369462241174564,
      "learning_rate": 0.0007344078370535756,
      "loss": 2.5352,
      "step": 795
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.4111835636782617,
      "learning_rate": 0.0007337515606624148,
      "loss": 2.5918,
      "step": 796
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3961816308530645,
      "learning_rate": 0.0007330947685281362,
      "loss": 2.5605,
      "step": 797
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.37022835627101197,
      "learning_rate": 0.0007324374620998682,
      "loss": 2.5605,
      "step": 798
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.36107607351475063,
      "learning_rate": 0.000731779642827874,
      "loss": 2.4219,
      "step": 799
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.4005167819430563,
      "learning_rate": 0.0007311213121635483,
      "loss": 2.4766,
      "step": 800
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3757664045132119,
      "learning_rate": 0.0007304624715594139,
      "loss": 2.4922,
      "step": 801
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.36924014069899397,
      "learning_rate": 0.0007298031224691193,
      "loss": 2.5742,
      "step": 802
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.40291953740323144,
      "learning_rate": 0.0007291432663474339,
      "loss": 2.5723,
      "step": 803
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.380096096573123,
      "learning_rate": 0.0007284829046502467,
      "loss": 2.4961,
      "step": 804
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.37938972825124007,
      "learning_rate": 0.0007278220388345619,
      "loss": 2.4766,
      "step": 805
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3855006661108689,
      "learning_rate": 0.0007271606703584958,
      "loss": 2.5312,
      "step": 806
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.36254657205207236,
      "learning_rate": 0.000726498800681274,
      "loss": 2.502,
      "step": 807
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3648317612173727,
      "learning_rate": 0.0007258364312632279,
      "loss": 2.4883,
      "step": 808
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.38035685095729976,
      "learning_rate": 0.0007251735635657915,
      "loss": 2.4902,
      "step": 809
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3592064545823401,
      "learning_rate": 0.000724510199051498,
      "loss": 2.459,
      "step": 810
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3799549120705248,
      "learning_rate": 0.0007238463391839769,
      "loss": 2.5801,
      "step": 811
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3966907417214172,
      "learning_rate": 0.0007231819854279508,
      "loss": 2.5547,
      "step": 812
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3752850238491139,
      "learning_rate": 0.0007225171392492316,
      "loss": 2.459,
      "step": 813
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3854662469195201,
      "learning_rate": 0.0007218518021147182,
      "loss": 2.541,
      "step": 814
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.39455274308551896,
      "learning_rate": 0.0007211859754923923,
      "loss": 2.5586,
      "step": 815
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.4022940474705106,
      "learning_rate": 0.0007205196608513158,
      "loss": 2.5684,
      "step": 816
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.379501676411674,
      "learning_rate": 0.0007198528596616272,
      "loss": 2.4336,
      "step": 817
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.39869358619505474,
      "learning_rate": 0.0007191855733945387,
      "loss": 2.4805,
      "step": 818
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.39755135976229916,
      "learning_rate": 0.0007185178035223327,
      "loss": 2.4805,
      "step": 819
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.4141430370908538,
      "learning_rate": 0.0007178495515183583,
      "loss": 2.543,
      "step": 820
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.41190899196468495,
      "learning_rate": 0.000717180818857029,
      "loss": 2.5156,
      "step": 821
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.38273629209562104,
      "learning_rate": 0.0007165116070138182,
      "loss": 2.6309,
      "step": 822
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.40991240714961513,
      "learning_rate": 0.0007158419174652569,
      "loss": 2.5137,
      "step": 823
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.4205020578135034,
      "learning_rate": 0.00071517175168893,
      "loss": 2.498,
      "step": 824
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.3900451964591823,
      "learning_rate": 0.0007145011111634732,
      "loss": 2.5176,
      "step": 825
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.4426412148763417,
      "learning_rate": 0.0007138299973685694,
      "loss": 2.6133,
      "step": 826
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.3982923361729594,
      "learning_rate": 0.0007131584117849459,
      "loss": 2.5527,
      "step": 827
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.35498521446141107,
      "learning_rate": 0.0007124863558943713,
      "loss": 2.4473,
      "step": 828
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.3618349943778806,
      "learning_rate": 0.0007118138311796514,
      "loss": 2.4961,
      "step": 829
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.39235025718117905,
      "learning_rate": 0.0007111408391246262,
      "loss": 2.5859,
      "step": 830
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.412008821773245,
      "learning_rate": 0.0007104673812141675,
      "loss": 2.502,
      "step": 831
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.39131355177336363,
      "learning_rate": 0.0007097934589341745,
      "loss": 2.4941,
      "step": 832
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.37895932683024003,
      "learning_rate": 0.0007091190737715711,
      "loss": 2.4629,
      "step": 833
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.3937763929701672,
      "learning_rate": 0.0007084442272143026,
      "loss": 2.5293,
      "step": 834
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.3832275957805093,
      "learning_rate": 0.000707768920751332,
      "loss": 2.5742,
      "step": 835
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.36570441160098455,
      "learning_rate": 0.0007070931558726373,
      "loss": 2.5723,
      "step": 836
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.3875688800040636,
      "learning_rate": 0.0007064169340692076,
      "loss": 2.5645,
      "step": 837
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.3816614293015164,
      "learning_rate": 0.0007057402568330407,
      "loss": 2.582,
      "step": 838
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.38687354381570616,
      "learning_rate": 0.0007050631256571389,
      "loss": 2.5723,
      "step": 839
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.3693282900629493,
      "learning_rate": 0.000704385542035506,
      "loss": 2.5039,
      "step": 840
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.37175007596100884,
      "learning_rate": 0.000703707507463144,
      "loss": 2.5938,
      "step": 841
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.3916527958437485,
      "learning_rate": 0.0007030290234360505,
      "loss": 2.4434,
      "step": 842
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.36992035321696304,
      "learning_rate": 0.0007023500914512139,
      "loss": 2.4863,
      "step": 843
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.4027603829129113,
      "learning_rate": 0.0007016707130066116,
      "loss": 2.5,
      "step": 844
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.36363218145901466,
      "learning_rate": 0.0007009908896012055,
      "loss": 2.5703,
      "step": 845
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.38853226026437954,
      "learning_rate": 0.0007003106227349399,
      "loss": 2.4863,
      "step": 846
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.36759112519080567,
      "learning_rate": 0.000699629913908737,
      "loss": 2.4668,
      "step": 847
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.4134401889717951,
      "learning_rate": 0.0006989487646244943,
      "loss": 2.4727,
      "step": 848
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.391702164670627,
      "learning_rate": 0.0006982671763850814,
      "loss": 2.4746,
      "step": 849
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.38123449492419664,
      "learning_rate": 0.0006975851506943359,
      "loss": 2.5078,
      "step": 850
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.3889232296657741,
      "learning_rate": 0.0006969026890570611,
      "loss": 2.5352,
      "step": 851
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.3715024165274851,
      "learning_rate": 0.0006962197929790216,
      "loss": 2.502,
      "step": 852
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.37010604134222663,
      "learning_rate": 0.0006955364639669409,
      "loss": 2.5176,
      "step": 853
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.39519030978887426,
      "learning_rate": 0.0006948527035284978,
      "loss": 2.4961,
      "step": 854
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.3764926204997517,
      "learning_rate": 0.0006941685131723225,
      "loss": 2.5391,
      "step": 855
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.3744214751487865,
      "learning_rate": 0.0006934838944079943,
      "loss": 2.5781,
      "step": 856
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.36261239116752636,
      "learning_rate": 0.0006927988487460378,
      "loss": 2.4199,
      "step": 857
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.38671819015243253,
      "learning_rate": 0.0006921133776979186,
      "loss": 2.5156,
      "step": 858
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.35100311767487324,
      "learning_rate": 0.0006914274827760418,
      "loss": 2.4805,
      "step": 859
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.39135609465616145,
      "learning_rate": 0.0006907411654937475,
      "loss": 2.6016,
      "step": 860
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.39970413283730877,
      "learning_rate": 0.0006900544273653075,
      "loss": 2.5273,
      "step": 861
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.34789786827277275,
      "learning_rate": 0.000689367269905922,
      "loss": 2.5664,
      "step": 862
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.35751042285393936,
      "learning_rate": 0.0006886796946317168,
      "loss": 2.543,
      "step": 863
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5615084487013351,
      "learning_rate": 0.0006879917030597397,
      "loss": 2.5312,
      "step": 864
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3647539688588585,
      "learning_rate": 0.0006873032967079561,
      "loss": 2.5742,
      "step": 865
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.377648717230249,
      "learning_rate": 0.0006866144770952474,
      "loss": 2.5273,
      "step": 866
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.38204631164903796,
      "learning_rate": 0.0006859252457414067,
      "loss": 2.5254,
      "step": 867
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.40052596812591684,
      "learning_rate": 0.0006852356041671351,
      "loss": 2.5254,
      "step": 868
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.40933824372154953,
      "learning_rate": 0.0006845455538940394,
      "loss": 2.5332,
      "step": 869
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3592013916082447,
      "learning_rate": 0.0006838550964446276,
      "loss": 2.4707,
      "step": 870
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3833325679242676,
      "learning_rate": 0.0006831642333423067,
      "loss": 2.5371,
      "step": 871
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3715143191384816,
      "learning_rate": 0.000682472966111378,
      "loss": 2.5039,
      "step": 872
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.38130046564556086,
      "learning_rate": 0.0006817812962770348,
      "loss": 2.5859,
      "step": 873
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.37962175704157547,
      "learning_rate": 0.0006810892253653589,
      "loss": 2.4414,
      "step": 874
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.36304239899816376,
      "learning_rate": 0.0006803967549033167,
      "loss": 2.5605,
      "step": 875
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.369363629008378,
      "learning_rate": 0.0006797038864187564,
      "loss": 2.5352,
      "step": 876
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.36677046703828087,
      "learning_rate": 0.0006790106214404043,
      "loss": 2.5625,
      "step": 877
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3568611961181725,
      "learning_rate": 0.0006783169614978614,
      "loss": 2.5215,
      "step": 878
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3714111926749449,
      "learning_rate": 0.0006776229081216001,
      "loss": 2.5254,
      "step": 879
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3744654766400233,
      "learning_rate": 0.0006769284628429611,
      "loss": 2.5742,
      "step": 880
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3708877731011553,
      "learning_rate": 0.0006762336271941498,
      "loss": 2.5723,
      "step": 881
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3779027349973449,
      "learning_rate": 0.0006755384027082326,
      "loss": 2.4004,
      "step": 882
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.39084915867158937,
      "learning_rate": 0.0006748427909191342,
      "loss": 2.5117,
      "step": 883
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.3871346346784643,
      "learning_rate": 0.0006741467933616335,
      "loss": 2.5664,
      "step": 884
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.3883443177275864,
      "learning_rate": 0.0006734504115713604,
      "loss": 2.4258,
      "step": 885
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.3798422474963951,
      "learning_rate": 0.0006727536470847932,
      "loss": 2.4629,
      "step": 886
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.37074270102006707,
      "learning_rate": 0.000672056501439254,
      "loss": 2.4941,
      "step": 887
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.38083548218719376,
      "learning_rate": 0.0006713589761729063,
      "loss": 2.502,
      "step": 888
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.3828930883250762,
      "learning_rate": 0.0006706610728247508,
      "loss": 2.4199,
      "step": 889
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.36392898681046537,
      "learning_rate": 0.0006699627929346227,
      "loss": 2.4766,
      "step": 890
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.36759988039675084,
      "learning_rate": 0.0006692641380431879,
      "loss": 2.6094,
      "step": 891
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.37490599933829616,
      "learning_rate": 0.0006685651096919393,
      "loss": 2.4355,
      "step": 892
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.34953845875893325,
      "learning_rate": 0.0006678657094231944,
      "loss": 2.5156,
      "step": 893
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.42715209717135605,
      "learning_rate": 0.0006671659387800909,
      "loss": 2.5039,
      "step": 894
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.37269144077007194,
      "learning_rate": 0.000666465799306584,
      "loss": 2.4785,
      "step": 895
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.4097524859539718,
      "learning_rate": 0.0006657652925474423,
      "loss": 2.5176,
      "step": 896
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.36325455090479447,
      "learning_rate": 0.000665064420048245,
      "loss": 2.543,
      "step": 897
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.3888638945868926,
      "learning_rate": 0.0006643631833553785,
      "loss": 2.541,
      "step": 898
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.3657022707739725,
      "learning_rate": 0.000663661584016032,
      "loss": 2.5273,
      "step": 899
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.40186682693170706,
      "learning_rate": 0.0006629596235781957,
      "loss": 2.4824,
      "step": 900
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.43353798566479607,
      "learning_rate": 0.0006622573035906556,
      "loss": 2.5449,
      "step": 901
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.37413182501953196,
      "learning_rate": 0.0006615546256029921,
      "loss": 2.3848,
      "step": 902
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.3530420457992454,
      "learning_rate": 0.0006608515911655743,
      "loss": 2.4883,
      "step": 903
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.3934749298824924,
      "learning_rate": 0.0006601482018295591,
      "loss": 2.5371,
      "step": 904
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.42452406561982164,
      "learning_rate": 0.0006594444591468851,
      "loss": 2.5586,
      "step": 905
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.3854335382672011,
      "learning_rate": 0.0006587403646702713,
      "loss": 2.5195,
      "step": 906
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.3736700737616124,
      "learning_rate": 0.0006580359199532126,
      "loss": 2.4883,
      "step": 907
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.40616891189532334,
      "learning_rate": 0.000657331126549977,
      "loss": 2.4277,
      "step": 908
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.3833761636144324,
      "learning_rate": 0.0006566259860156014,
      "loss": 2.4336,
      "step": 909
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.38384699987027976,
      "learning_rate": 0.0006559204999058888,
      "loss": 2.4961,
      "step": 910
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.6759016337691449,
      "learning_rate": 0.0006552146697774049,
      "loss": 2.4863,
      "step": 911
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.38879677379992583,
      "learning_rate": 0.0006545084971874737,
      "loss": 2.4688,
      "step": 912
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.3954748315069902,
      "learning_rate": 0.0006538019836941758,
      "loss": 2.4863,
      "step": 913
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.3903870224242552,
      "learning_rate": 0.0006530951308563431,
      "loss": 2.5449,
      "step": 914
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.36294507090228834,
      "learning_rate": 0.0006523879402335567,
      "loss": 2.5137,
      "step": 915
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.4136287890077231,
      "learning_rate": 0.0006516804133861429,
      "loss": 2.627,
      "step": 916
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.45837351783744873,
      "learning_rate": 0.0006509725518751698,
      "loss": 2.4141,
      "step": 917
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.38557990476725756,
      "learning_rate": 0.0006502643572624438,
      "loss": 2.4883,
      "step": 918
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.39666333238933565,
      "learning_rate": 0.0006495558311105064,
      "loss": 2.4453,
      "step": 919
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.4471124828440117,
      "learning_rate": 0.0006488469749826305,
      "loss": 2.543,
      "step": 920
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.4341613425000251,
      "learning_rate": 0.000648137790442817,
      "loss": 2.4609,
      "step": 921
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.40239499855667504,
      "learning_rate": 0.0006474282790557916,
      "loss": 2.5293,
      "step": 922
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.3871362816269737,
      "learning_rate": 0.000646718442387001,
      "loss": 2.5449,
      "step": 923
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.39599755828706545,
      "learning_rate": 0.0006460082820026094,
      "loss": 2.5039,
      "step": 924
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.39837047419858895,
      "learning_rate": 0.0006452977994694959,
      "loss": 2.5801,
      "step": 925
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.37254270533614553,
      "learning_rate": 0.0006445869963552496,
      "loss": 2.5273,
      "step": 926
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.3748998268356706,
      "learning_rate": 0.0006438758742281672,
      "loss": 2.5859,
      "step": 927
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.404303813734554,
      "learning_rate": 0.0006431644346572495,
      "loss": 2.5195,
      "step": 928
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.3687473851573776,
      "learning_rate": 0.0006424526792121974,
      "loss": 2.4629,
      "step": 929
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.39022544326706016,
      "learning_rate": 0.0006417406094634089,
      "loss": 2.4824,
      "step": 930
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.37901287300771386,
      "learning_rate": 0.0006410282269819756,
      "loss": 2.5898,
      "step": 931
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.45416127723642374,
      "learning_rate": 0.0006403155333396787,
      "loss": 2.5195,
      "step": 932
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.3796862572411226,
      "learning_rate": 0.0006396025301089863,
      "loss": 2.584,
      "step": 933
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.37402996633038027,
      "learning_rate": 0.0006388892188630493,
      "loss": 2.5859,
      "step": 934
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.37885693170464396,
      "learning_rate": 0.0006381756011756982,
      "loss": 2.5977,
      "step": 935
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.37616884853877636,
      "learning_rate": 0.0006374616786214403,
      "loss": 2.4805,
      "step": 936
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.3603646400305853,
      "learning_rate": 0.0006367474527754544,
      "loss": 2.5176,
      "step": 937
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.38037725520591464,
      "learning_rate": 0.0006360329252135894,
      "loss": 2.502,
      "step": 938
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.37808306868147634,
      "learning_rate": 0.0006353180975123595,
      "loss": 2.502,
      "step": 939
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.3942336879257232,
      "learning_rate": 0.0006346029712489413,
      "loss": 2.5254,
      "step": 940
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.3909610675410671,
      "learning_rate": 0.0006338875480011698,
      "loss": 2.5664,
      "step": 941
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.38468147287358,
      "learning_rate": 0.0006331718293475357,
      "loss": 2.4297,
      "step": 942
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.37877288321911456,
      "learning_rate": 0.0006324558168671811,
      "loss": 2.6426,
      "step": 943
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.3905877517285995,
      "learning_rate": 0.0006317395121398968,
      "loss": 2.5625,
      "step": 944
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.38771307225051094,
      "learning_rate": 0.0006310229167461179,
      "loss": 2.5195,
      "step": 945
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.38118411778238404,
      "learning_rate": 0.0006303060322669214,
      "loss": 2.5156,
      "step": 946
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.4255379075908999,
      "learning_rate": 0.0006295888602840214,
      "loss": 2.3887,
      "step": 947
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.41530662085284853,
      "learning_rate": 0.0006288714023797671,
      "loss": 2.5059,
      "step": 948
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.35854793009573355,
      "learning_rate": 0.000628153660137138,
      "loss": 2.5176,
      "step": 949
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.40875699564923024,
      "learning_rate": 0.0006274356351397413,
      "loss": 2.4609,
      "step": 950
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.39039041106089567,
      "learning_rate": 0.0006267173289718079,
      "loss": 2.4297,
      "step": 951
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.4061758542972262,
      "learning_rate": 0.000625998743218189,
      "loss": 2.5723,
      "step": 952
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.38549094528798405,
      "learning_rate": 0.000625279879464353,
      "loss": 2.4551,
      "step": 953
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.38450461365762184,
      "learning_rate": 0.000624560739296381,
      "loss": 2.5176,
      "step": 954
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.42813318002441825,
      "learning_rate": 0.0006238413243009648,
      "loss": 2.459,
      "step": 955
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.3854561925541816,
      "learning_rate": 0.000623121636065402,
      "loss": 2.4531,
      "step": 956
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.37572786118185314,
      "learning_rate": 0.0006224016761775933,
      "loss": 2.4707,
      "step": 957
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.3958486435800659,
      "learning_rate": 0.0006216814462260386,
      "loss": 2.5176,
      "step": 958
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.39005402924358834,
      "learning_rate": 0.0006209609477998338,
      "loss": 2.5508,
      "step": 959
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.3622530641112439,
      "learning_rate": 0.0006202401824886674,
      "loss": 2.4902,
      "step": 960
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.3523461753737025,
      "learning_rate": 0.0006195191518828162,
      "loss": 2.5449,
      "step": 961
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.35392071601479097,
      "learning_rate": 0.0006187978575731427,
      "loss": 2.4805,
      "step": 962
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.3871745485505628,
      "learning_rate": 0.0006180763011510911,
      "loss": 2.5059,
      "step": 963
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.35949585457808364,
      "learning_rate": 0.000617354484208684,
      "loss": 2.4668,
      "step": 964
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.37175955190457827,
      "learning_rate": 0.0006166324083385189,
      "loss": 2.5566,
      "step": 965
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.36399988359239416,
      "learning_rate": 0.0006159100751337642,
      "loss": 2.416,
      "step": 966
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.38718611674570663,
      "learning_rate": 0.0006151874861881565,
      "loss": 2.4492,
      "step": 967
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.36185057893116523,
      "learning_rate": 0.0006144646430959964,
      "loss": 2.5859,
      "step": 968
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.38623422371133753,
      "learning_rate": 0.0006137415474521454,
      "loss": 2.543,
      "step": 969
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.40661543183551263,
      "learning_rate": 0.0006130182008520222,
      "loss": 2.4844,
      "step": 970
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.3670110172979152,
      "learning_rate": 0.000612294604891599,
      "loss": 2.5371,
      "step": 971
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.37472223696403734,
      "learning_rate": 0.0006115707611673986,
      "loss": 2.5391,
      "step": 972
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.3815935270455712,
      "learning_rate": 0.0006108466712764902,
      "loss": 2.4766,
      "step": 973
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.37648060347575113,
      "learning_rate": 0.0006101223368164858,
      "loss": 2.5449,
      "step": 974
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.3891105969924535,
      "learning_rate": 0.0006093977593855375,
      "loss": 2.5254,
      "step": 975
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.3818082391620932,
      "learning_rate": 0.0006086729405823335,
      "loss": 2.4961,
      "step": 976
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.36747479133917693,
      "learning_rate": 0.0006079478820060943,
      "loss": 2.5254,
      "step": 977
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.36168337004546874,
      "learning_rate": 0.0006072225852565695,
      "loss": 2.4648,
      "step": 978
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.3492572884318507,
      "learning_rate": 0.0006064970519340341,
      "loss": 2.5273,
      "step": 979
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.37850689212372074,
      "learning_rate": 0.0006057712836392856,
      "loss": 2.4336,
      "step": 980
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.35404043884698444,
      "learning_rate": 0.0006050452819736389,
      "loss": 2.5605,
      "step": 981
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.36771412826923,
      "learning_rate": 0.000604319048538925,
      "loss": 2.4258,
      "step": 982
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.3965692274460368,
      "learning_rate": 0.0006035925849374855,
      "loss": 2.5332,
      "step": 983
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.46890599714753484,
      "learning_rate": 0.0006028658927721697,
      "loss": 2.5566,
      "step": 984
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.36763586253314146,
      "learning_rate": 0.0006021389736463321,
      "loss": 2.5703,
      "step": 985
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.3905584976092622,
      "learning_rate": 0.0006014118291638271,
      "loss": 2.4336,
      "step": 986
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.3846185091769459,
      "learning_rate": 0.0006006844609290065,
      "loss": 2.5918,
      "step": 987
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.4123503333105017,
      "learning_rate": 0.0005999568705467161,
      "loss": 2.6426,
      "step": 988
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.42465124702436563,
      "learning_rate": 0.0005992290596222915,
      "loss": 2.4961,
      "step": 989
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.3671533809380912,
      "learning_rate": 0.0005985010297615551,
      "loss": 2.541,
      "step": 990
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.37172100806976016,
      "learning_rate": 0.0005977727825708123,
      "loss": 2.5156,
      "step": 991
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.3876307436572213,
      "learning_rate": 0.0005970443196568478,
      "loss": 2.4414,
      "step": 992
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3974455556892383,
      "learning_rate": 0.0005963156426269227,
      "loss": 2.4023,
      "step": 993
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.40053206931462,
      "learning_rate": 0.0005955867530887702,
      "loss": 2.4746,
      "step": 994
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.37794618545914943,
      "learning_rate": 0.0005948576526505923,
      "loss": 2.4707,
      "step": 995
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.38095626404837074,
      "learning_rate": 0.0005941283429210568,
      "loss": 2.502,
      "step": 996
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3959132921837636,
      "learning_rate": 0.0005933988255092926,
      "loss": 2.4941,
      "step": 997
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5092405012607447,
      "learning_rate": 0.0005926691020248874,
      "loss": 2.4395,
      "step": 998
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.407620146205852,
      "learning_rate": 0.0005919391740778833,
      "loss": 2.502,
      "step": 999
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3813025358410362,
      "learning_rate": 0.0005912090432787736,
      "loss": 2.4941,
      "step": 1000
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3807756158001032,
      "learning_rate": 0.000590478711238499,
      "loss": 2.377,
      "step": 1001
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3969649551158979,
      "learning_rate": 0.0005897481795684446,
      "loss": 2.5195,
      "step": 1002
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.4076938936424566,
      "learning_rate": 0.0005890174498804355,
      "loss": 2.4414,
      "step": 1003
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3861719273151608,
      "learning_rate": 0.0005882865237867339,
      "loss": 2.5117,
      "step": 1004
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3852640224380834,
      "learning_rate": 0.0005875554029000353,
      "loss": 2.498,
      "step": 1005
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.36590919608348926,
      "learning_rate": 0.0005868240888334653,
      "loss": 2.3984,
      "step": 1006
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3850297990299645,
      "learning_rate": 0.0005860925832005753,
      "loss": 2.4277,
      "step": 1007
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.39793439249491824,
      "learning_rate": 0.0005853608876153395,
      "loss": 2.5645,
      "step": 1008
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.37791209263098224,
      "learning_rate": 0.0005846290036921512,
      "loss": 2.5273,
      "step": 1009
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3757287157181429,
      "learning_rate": 0.0005838969330458195,
      "loss": 2.3867,
      "step": 1010
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.38087268142808056,
      "learning_rate": 0.0005831646772915651,
      "loss": 2.5762,
      "step": 1011
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3757154478393314,
      "learning_rate": 0.0005824322380450173,
      "loss": 2.3984,
      "step": 1012
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3698839870572313,
      "learning_rate": 0.0005816996169222102,
      "loss": 2.5449,
      "step": 1013
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3669492082373975,
      "learning_rate": 0.0005809668155395793,
      "loss": 2.4609,
      "step": 1014
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.39191018050166887,
      "learning_rate": 0.0005802338355139578,
      "loss": 2.4648,
      "step": 1015
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.36805341146951054,
      "learning_rate": 0.0005795006784625728,
      "loss": 2.4844,
      "step": 1016
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.4077878100096171,
      "learning_rate": 0.0005787673460030423,
      "loss": 2.5684,
      "step": 1017
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.37095134310008465,
      "learning_rate": 0.000578033839753371,
      "loss": 2.3945,
      "step": 1018
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.3856512078602976,
      "learning_rate": 0.0005773001613319476,
      "loss": 2.4414,
      "step": 1019
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.3988731702981723,
      "learning_rate": 0.00057656631235754,
      "loss": 2.4609,
      "step": 1020
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.4183173031192875,
      "learning_rate": 0.0005758322944492929,
      "loss": 2.457,
      "step": 1021
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.39689672535411896,
      "learning_rate": 0.0005750981092267237,
      "loss": 2.5801,
      "step": 1022
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.379885714959307,
      "learning_rate": 0.0005743637583097183,
      "loss": 2.4629,
      "step": 1023
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.38355177179511196,
      "learning_rate": 0.0005736292433185291,
      "loss": 2.4805,
      "step": 1024
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.3807150424375656,
      "learning_rate": 0.0005728945658737699,
      "loss": 2.6094,
      "step": 1025
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.3700386795418782,
      "learning_rate": 0.0005721597275964133,
      "loss": 2.4629,
      "step": 1026
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.38424639807480254,
      "learning_rate": 0.0005714247301077865,
      "loss": 2.4531,
      "step": 1027
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.3674080260504661,
      "learning_rate": 0.0005706895750295682,
      "loss": 2.4922,
      "step": 1028
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.4008197183887381,
      "learning_rate": 0.0005699542639837844,
      "loss": 2.4785,
      "step": 1029
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.3729497614390487,
      "learning_rate": 0.0005692187985928055,
      "loss": 2.4824,
      "step": 1030
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.37886800420117583,
      "learning_rate": 0.0005684831804793427,
      "loss": 2.459,
      "step": 1031
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.38698769889995815,
      "learning_rate": 0.0005677474112664438,
      "loss": 2.5625,
      "step": 1032
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.3711565505708607,
      "learning_rate": 0.0005670114925774899,
      "loss": 2.4512,
      "step": 1033
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.37945436771535446,
      "learning_rate": 0.0005662754260361924,
      "loss": 2.5391,
      "step": 1034
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.3943102366275468,
      "learning_rate": 0.0005655392132665884,
      "loss": 2.5176,
      "step": 1035
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.40172709087269914,
      "learning_rate": 0.000564802855893038,
      "loss": 2.4844,
      "step": 1036
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.42452196155094446,
      "learning_rate": 0.0005640663555402198,
      "loss": 2.4941,
      "step": 1037
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3959593751454868,
      "learning_rate": 0.0005633297138331285,
      "loss": 2.4551,
      "step": 1038
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.380588377004395,
      "learning_rate": 0.0005625929323970705,
      "loss": 2.4746,
      "step": 1039
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.36780089176268577,
      "learning_rate": 0.0005618560128576603,
      "loss": 2.4961,
      "step": 1040
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.373594586975474,
      "learning_rate": 0.0005611189568408173,
      "loss": 2.4746,
      "step": 1041
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.37715183134110153,
      "learning_rate": 0.0005603817659727619,
      "loss": 2.4902,
      "step": 1042
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3694890518622045,
      "learning_rate": 0.0005596444418800121,
      "loss": 2.5039,
      "step": 1043
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3770896678683744,
      "learning_rate": 0.0005589069861893798,
      "loss": 2.4688,
      "step": 1044
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3646712864785194,
      "learning_rate": 0.0005581694005279673,
      "loss": 2.5195,
      "step": 1045
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3925124612113276,
      "learning_rate": 0.0005574316865231637,
      "loss": 2.627,
      "step": 1046
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3927791746230083,
      "learning_rate": 0.0005566938458026411,
      "loss": 2.5371,
      "step": 1047
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3916955920677205,
      "learning_rate": 0.0005559558799943514,
      "loss": 2.5586,
      "step": 1048
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.38989855541360313,
      "learning_rate": 0.0005552177907265223,
      "loss": 2.5137,
      "step": 1049
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.38844378990855555,
      "learning_rate": 0.000554479579627654,
      "loss": 2.498,
      "step": 1050
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.36043883182970304,
      "learning_rate": 0.0005537412483265157,
      "loss": 2.5664,
      "step": 1051
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.41357351855316704,
      "learning_rate": 0.0005530027984521413,
      "loss": 2.5527,
      "step": 1052
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3689925705247356,
      "learning_rate": 0.0005522642316338268,
      "loss": 2.5332,
      "step": 1053
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.4006136778474805,
      "learning_rate": 0.0005515255495011259,
      "loss": 2.4785,
      "step": 1054
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3552835938634182,
      "learning_rate": 0.0005507867536838472,
      "loss": 2.4902,
      "step": 1055
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.36703065785833333,
      "learning_rate": 0.0005500478458120492,
      "loss": 2.4746,
      "step": 1056
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3778820810287155,
      "learning_rate": 0.0005493088275160387,
      "loss": 2.5391,
      "step": 1057
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3875998134966287,
      "learning_rate": 0.0005485697004263657,
      "loss": 2.4883,
      "step": 1058
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3662623819139003,
      "learning_rate": 0.0005478304661738199,
      "loss": 2.5605,
      "step": 1059
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3843371416038381,
      "learning_rate": 0.0005470911263894279,
      "loss": 2.3594,
      "step": 1060
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3598614522642916,
      "learning_rate": 0.0005463516827044491,
      "loss": 2.4121,
      "step": 1061
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3783559640723365,
      "learning_rate": 0.000545612136750372,
      "loss": 2.5254,
      "step": 1062
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.36281677878082924,
      "learning_rate": 0.0005448724901589107,
      "loss": 2.4844,
      "step": 1063
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3900998612340574,
      "learning_rate": 0.0005441327445620014,
      "loss": 2.4648,
      "step": 1064
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.4401261615351944,
      "learning_rate": 0.0005433929015917988,
      "loss": 2.4824,
      "step": 1065
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3656500946014901,
      "learning_rate": 0.0005426529628806724,
      "loss": 2.3887,
      "step": 1066
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3685952760170525,
      "learning_rate": 0.0005419129300612029,
      "loss": 2.4707,
      "step": 1067
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.36834145776989063,
      "learning_rate": 0.000541172804766179,
      "loss": 2.5371,
      "step": 1068
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.39918784792484235,
      "learning_rate": 0.0005404325886285927,
      "loss": 2.4102,
      "step": 1069
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.364640126073911,
      "learning_rate": 0.000539692283281637,
      "loss": 2.5391,
      "step": 1070
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3891280195313063,
      "learning_rate": 0.0005389518903587017,
      "loss": 2.5488,
      "step": 1071
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3980167162732819,
      "learning_rate": 0.0005382114114933695,
      "loss": 2.4492,
      "step": 1072
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3750712404965566,
      "learning_rate": 0.0005374708483194132,
      "loss": 2.4316,
      "step": 1073
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3772291455870346,
      "learning_rate": 0.000536730202470791,
      "loss": 2.5117,
      "step": 1074
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3658008254755044,
      "learning_rate": 0.0005359894755816443,
      "loss": 2.5176,
      "step": 1075
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.37306442721876326,
      "learning_rate": 0.0005352486692862926,
      "loss": 2.4062,
      "step": 1076
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.36962299581561814,
      "learning_rate": 0.0005345077852192307,
      "loss": 2.5078,
      "step": 1077
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.37801324224291427,
      "learning_rate": 0.0005337668250151254,
      "loss": 2.459,
      "step": 1078
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3784871771421614,
      "learning_rate": 0.0005330257903088111,
      "loss": 2.3984,
      "step": 1079
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.37157980746532415,
      "learning_rate": 0.000532284682735287,
      "loss": 2.5391,
      "step": 1080
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.3663205405569991,
      "learning_rate": 0.0005315435039297124,
      "loss": 2.4414,
      "step": 1081
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.3965529003994781,
      "learning_rate": 0.0005308022555274046,
      "loss": 2.4746,
      "step": 1082
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.38723406477088423,
      "learning_rate": 0.0005300609391638336,
      "loss": 2.5547,
      "step": 1083
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.37955317044232456,
      "learning_rate": 0.0005293195564746201,
      "loss": 2.5,
      "step": 1084
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.403877456396635,
      "learning_rate": 0.0005285781090955304,
      "loss": 2.4961,
      "step": 1085
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.4065689900795475,
      "learning_rate": 0.0005278365986624743,
      "loss": 2.4277,
      "step": 1086
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.37373822842012766,
      "learning_rate": 0.0005270950268115001,
      "loss": 2.5547,
      "step": 1087
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.3904302893335085,
      "learning_rate": 0.0005263533951787919,
      "loss": 2.5176,
      "step": 1088
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.38624571356153403,
      "learning_rate": 0.000525611705400666,
      "loss": 2.5371,
      "step": 1089
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.3938190423754996,
      "learning_rate": 0.0005248699591135664,
      "loss": 2.5293,
      "step": 1090
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.40112627255076716,
      "learning_rate": 0.0005241281579540618,
      "loss": 2.502,
      "step": 1091
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.4028268776188261,
      "learning_rate": 0.0005233863035588427,
      "loss": 2.5469,
      "step": 1092
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.388842141354559,
      "learning_rate": 0.0005226443975647161,
      "loss": 2.5234,
      "step": 1093
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.40786279047056834,
      "learning_rate": 0.0005219024416086036,
      "loss": 2.5,
      "step": 1094
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.3859483143616971,
      "learning_rate": 0.0005211604373275366,
      "loss": 2.5508,
      "step": 1095
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.3615474305229973,
      "learning_rate": 0.0005204183863586533,
      "loss": 2.5039,
      "step": 1096
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.37210324186596083,
      "learning_rate": 0.0005196762903391951,
      "loss": 2.5352,
      "step": 1097
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.3545522475151223,
      "learning_rate": 0.0005189341509065023,
      "loss": 2.4824,
      "step": 1098
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.36626939398397046,
      "learning_rate": 0.0005181919696980112,
      "loss": 2.5078,
      "step": 1099
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.37191811294172294,
      "learning_rate": 0.0005174497483512506,
      "loss": 2.4883,
      "step": 1100
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.371649504200305,
      "learning_rate": 0.0005167074885038374,
      "loss": 2.4707,
      "step": 1101
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.3742431809115339,
      "learning_rate": 0.0005159651917934735,
      "loss": 2.416,
      "step": 1102
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.3919536222857386,
      "learning_rate": 0.0005152228598579428,
      "loss": 2.4883,
      "step": 1103
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.3651675380132402,
      "learning_rate": 0.000514480494335106,
      "loss": 2.416,
      "step": 1104
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.36870996182056043,
      "learning_rate": 0.0005137380968628983,
      "loss": 2.5488,
      "step": 1105
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.3623239053914332,
      "learning_rate": 0.0005129956690793255,
      "loss": 2.4785,
      "step": 1106
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.36423447385728025,
      "learning_rate": 0.0005122532126224601,
      "loss": 2.4375,
      "step": 1107
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.37324672659066926,
      "learning_rate": 0.0005115107291304378,
      "loss": 2.4473,
      "step": 1108
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.37501525560717314,
      "learning_rate": 0.0005107682202414544,
      "loss": 2.4902,
      "step": 1109
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.39048338677508054,
      "learning_rate": 0.0005100256875937613,
      "loss": 2.5156,
      "step": 1110
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.38721350521487763,
      "learning_rate": 0.0005092831328256625,
      "loss": 2.4844,
      "step": 1111
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.3848824873474211,
      "learning_rate": 0.0005085405575755105,
      "loss": 2.5195,
      "step": 1112
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.4030077554964755,
      "learning_rate": 0.0005077979634817034,
      "loss": 2.502,
      "step": 1113
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.4252664696907678,
      "learning_rate": 0.0005070553521826808,
      "loss": 2.5273,
      "step": 1114
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.4003667536821472,
      "learning_rate": 0.00050631272531692,
      "loss": 2.416,
      "step": 1115
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.4186577962839968,
      "learning_rate": 0.0005055700845229327,
      "loss": 2.5312,
      "step": 1116
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.40462771537305103,
      "learning_rate": 0.000504827431439262,
      "loss": 2.5254,
      "step": 1117
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.3742429106705449,
      "learning_rate": 0.000504084767704477,
      "loss": 2.4531,
      "step": 1118
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.38619362005638286,
      "learning_rate": 0.0005033420949571712,
      "loss": 2.5645,
      "step": 1119
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.38695456107768633,
      "learning_rate": 0.0005025994148359574,
      "loss": 2.4062,
      "step": 1120
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.4330544285094602,
      "learning_rate": 0.0005018567289794651,
      "loss": 2.4121,
      "step": 1121
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.3811946670692585,
      "learning_rate": 0.0005011140390263362,
      "loss": 2.5332,
      "step": 1122
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.38875930974701656,
      "learning_rate": 0.0005003713466152218,
      "loss": 2.4238,
      "step": 1123
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.3878583607263707,
      "learning_rate": 0.0004996286533847783,
      "loss": 2.4551,
      "step": 1124
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.3915465461122664,
      "learning_rate": 0.000498885960973664,
      "loss": 2.4336,
      "step": 1125
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.38092728878149645,
      "learning_rate": 0.000498143271020535,
      "loss": 2.5156,
      "step": 1126
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.35730752514668646,
      "learning_rate": 0.0004974005851640428,
      "loss": 2.5176,
      "step": 1127
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4309013579584056,
      "learning_rate": 0.000496657905042829,
      "loss": 2.6055,
      "step": 1128
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.37168528923416616,
      "learning_rate": 0.0004959152322955232,
      "loss": 2.5273,
      "step": 1129
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4223791027137694,
      "learning_rate": 0.0004951725685607382,
      "loss": 2.3926,
      "step": 1130
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.35134333776227616,
      "learning_rate": 0.0004944299154770673,
      "loss": 2.4297,
      "step": 1131
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.3812116007861276,
      "learning_rate": 0.0004936872746830802,
      "loss": 2.4727,
      "step": 1132
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4063256246644419,
      "learning_rate": 0.0004929446478173195,
      "loss": 2.4688,
      "step": 1133
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.35674081761523396,
      "learning_rate": 0.0004922020365182968,
      "loss": 2.5078,
      "step": 1134
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.36709762540828983,
      "learning_rate": 0.0004914594424244897,
      "loss": 2.502,
      "step": 1135
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.36408527191865875,
      "learning_rate": 0.0004907168671743376,
      "loss": 2.4551,
      "step": 1136
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.38850356850676093,
      "learning_rate": 0.0004899743124062387,
      "loss": 2.5703,
      "step": 1137
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.46278092845261615,
      "learning_rate": 0.0004892317797585456,
      "loss": 2.4863,
      "step": 1138
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.3768979483315128,
      "learning_rate": 0.0004884892708695623,
      "loss": 2.4805,
      "step": 1139
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.37985831967921874,
      "learning_rate": 0.0004877467873775402,
      "loss": 2.4219,
      "step": 1140
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.44236139549323505,
      "learning_rate": 0.00048700433092067473,
      "loss": 2.5273,
      "step": 1141
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.47082427347175565,
      "learning_rate": 0.0004862619031371019,
      "loss": 2.5098,
      "step": 1142
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.3808099677977431,
      "learning_rate": 0.0004855195056648942,
      "loss": 2.5,
      "step": 1143
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4171295816287706,
      "learning_rate": 0.00048477714014205734,
      "loss": 2.3828,
      "step": 1144
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4711016528102692,
      "learning_rate": 0.00048403480820652644,
      "loss": 2.4648,
      "step": 1145
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.4264643700818479,
      "learning_rate": 0.0004832925114961629,
      "loss": 2.5742,
      "step": 1146
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3792771275285108,
      "learning_rate": 0.0004825502516487497,
      "loss": 2.4082,
      "step": 1147
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.40129151190678886,
      "learning_rate": 0.00048180803030198896,
      "loss": 2.5488,
      "step": 1148
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.46231760788905996,
      "learning_rate": 0.0004810658490934979,
      "loss": 2.502,
      "step": 1149
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.4053267826237742,
      "learning_rate": 0.000480323709660805,
      "loss": 2.4102,
      "step": 1150
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.41974797952223575,
      "learning_rate": 0.0004795816136413467,
      "loss": 2.4961,
      "step": 1151
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3897165348231591,
      "learning_rate": 0.00047883956267246353,
      "loss": 2.4375,
      "step": 1152
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.38578495282011754,
      "learning_rate": 0.00047809755839139657,
      "loss": 2.4121,
      "step": 1153
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3946060390686367,
      "learning_rate": 0.0004773556024352841,
      "loss": 2.5254,
      "step": 1154
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.39012358167256894,
      "learning_rate": 0.00047661369644115754,
      "loss": 2.6016,
      "step": 1155
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.40001268707235416,
      "learning_rate": 0.0004758718420459383,
      "loss": 2.4297,
      "step": 1156
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3935568085618047,
      "learning_rate": 0.0004751300408864339,
      "loss": 2.541,
      "step": 1157
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.37991416949291956,
      "learning_rate": 0.00047438829459933414,
      "loss": 2.4316,
      "step": 1158
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3772750260080625,
      "learning_rate": 0.0004736466048212082,
      "loss": 2.4902,
      "step": 1159
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.4052555293555998,
      "learning_rate": 0.0004729049731885002,
      "loss": 2.5078,
      "step": 1160
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.414452744729495,
      "learning_rate": 0.000472163401337526,
      "loss": 2.3516,
      "step": 1161
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3841435928329603,
      "learning_rate": 0.00047142189090446985,
      "loss": 2.4492,
      "step": 1162
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.37237473902637114,
      "learning_rate": 0.0004706804435253802,
      "loss": 2.3457,
      "step": 1163
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3981691834576357,
      "learning_rate": 0.0004699390608361665,
      "loss": 2.4785,
      "step": 1164
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3843073797953735,
      "learning_rate": 0.0004691977444725955,
      "loss": 2.4355,
      "step": 1165
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.36792953735891304,
      "learning_rate": 0.0004684564960702877,
      "loss": 2.5156,
      "step": 1166
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.3762020290020908,
      "learning_rate": 0.0004677153172647131,
      "loss": 2.4883,
      "step": 1167
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.3804765308860921,
      "learning_rate": 0.00046697420969118894,
      "loss": 2.4648,
      "step": 1168
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.37704078590487144,
      "learning_rate": 0.00046623317498487466,
      "loss": 2.4043,
      "step": 1169
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.3807896889992381,
      "learning_rate": 0.0004654922147807694,
      "loss": 2.457,
      "step": 1170
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.3921669304008037,
      "learning_rate": 0.00046475133071370757,
      "loss": 2.4082,
      "step": 1171
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.39518860236033126,
      "learning_rate": 0.00046401052441835574,
      "loss": 2.4902,
      "step": 1172
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.3786442227938226,
      "learning_rate": 0.000463269797529209,
      "loss": 2.4297,
      "step": 1173
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.3732680660397845,
      "learning_rate": 0.00046252915168058697,
      "loss": 2.4395,
      "step": 1174
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.38082153295844656,
      "learning_rate": 0.0004617885885066305,
      "loss": 2.4707,
      "step": 1175
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.37428513592733187,
      "learning_rate": 0.0004610481096412984,
      "loss": 2.4883,
      "step": 1176
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.40465464791753825,
      "learning_rate": 0.000460307716718363,
      "loss": 2.5234,
      "step": 1177
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.4220150719450851,
      "learning_rate": 0.0004595674113714074,
      "loss": 2.4648,
      "step": 1178
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.42165498665868406,
      "learning_rate": 0.0004588271952338212,
      "loss": 2.373,
      "step": 1179
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.38875081633051073,
      "learning_rate": 0.00045808706993879714,
      "loss": 2.3965,
      "step": 1180
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.39817450248202113,
      "learning_rate": 0.00045734703711932767,
      "loss": 2.498,
      "step": 1181
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.3899871543362889,
      "learning_rate": 0.0004566070984082013,
      "loss": 2.5527,
      "step": 1182
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.39841607148093067,
      "learning_rate": 0.00045586725543799865,
      "loss": 2.5117,
      "step": 1183
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.3887974537177613,
      "learning_rate": 0.00045512750984108937,
      "loss": 2.5176,
      "step": 1184
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.378915019923162,
      "learning_rate": 0.000454387863249628,
      "loss": 2.4961,
      "step": 1185
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.3729774737586374,
      "learning_rate": 0.00045364831729555096,
      "loss": 2.4961,
      "step": 1186
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.41601482047643457,
      "learning_rate": 0.0004529088736105721,
      "loss": 2.5527,
      "step": 1187
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.3770216607911211,
      "learning_rate": 0.0004521695338261802,
      "loss": 2.5273,
      "step": 1188
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3860547805344984,
      "learning_rate": 0.0004514302995736344,
      "loss": 2.4297,
      "step": 1189
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.38291138159853294,
      "learning_rate": 0.0004506911724839613,
      "loss": 2.5,
      "step": 1190
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3836051645020376,
      "learning_rate": 0.0004499521541879508,
      "loss": 2.5,
      "step": 1191
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3860518970147892,
      "learning_rate": 0.00044921324631615303,
      "loss": 2.4414,
      "step": 1192
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.37526120135899554,
      "learning_rate": 0.0004484744504988742,
      "loss": 2.4707,
      "step": 1193
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.4108787368682809,
      "learning_rate": 0.00044773576836617336,
      "loss": 2.4531,
      "step": 1194
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.38312761607757817,
      "learning_rate": 0.0004469972015478588,
      "loss": 2.5039,
      "step": 1195
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.38368005988981574,
      "learning_rate": 0.0004462587516734844,
      "loss": 2.4199,
      "step": 1196
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.40523716426007883,
      "learning_rate": 0.00044552042037234596,
      "loss": 2.5078,
      "step": 1197
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.4175928245597615,
      "learning_rate": 0.00044478220927347774,
      "loss": 2.5176,
      "step": 1198
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3704680280545028,
      "learning_rate": 0.00044404412000564875,
      "loss": 2.4336,
      "step": 1199
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3851967809346016,
      "learning_rate": 0.000443306154197359,
      "loss": 2.4844,
      "step": 1200
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.39809588160698567,
      "learning_rate": 0.00044256831347683646,
      "loss": 2.5234,
      "step": 1201
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3818299410503429,
      "learning_rate": 0.0004418305994720328,
      "loss": 2.5098,
      "step": 1202
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.4114198685654004,
      "learning_rate": 0.0004410930138106203,
      "loss": 2.3848,
      "step": 1203
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.4271773862088436,
      "learning_rate": 0.000440355558119988,
      "loss": 2.5312,
      "step": 1204
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3736939659083574,
      "learning_rate": 0.00043961823402723814,
      "loss": 2.4668,
      "step": 1205
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.38630377075987055,
      "learning_rate": 0.0004388810431591829,
      "loss": 2.502,
      "step": 1206
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.4107327486970333,
      "learning_rate": 0.0004381439871423398,
      "loss": 2.5312,
      "step": 1207
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3657623577118704,
      "learning_rate": 0.00043740706760292966,
      "loss": 2.5273,
      "step": 1208
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.37771322530231266,
      "learning_rate": 0.0004366702861668716,
      "loss": 2.4531,
      "step": 1209
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3467066366926856,
      "learning_rate": 0.00043593364445978036,
      "loss": 2.4316,
      "step": 1210
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.3704963694773179,
      "learning_rate": 0.0004351971441069622,
      "loss": 2.4336,
      "step": 1211
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.3919473932053192,
      "learning_rate": 0.0004344607867334116,
      "loss": 2.4902,
      "step": 1212
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.35535288621520267,
      "learning_rate": 0.00043372457396380766,
      "loss": 2.4727,
      "step": 1213
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.3578230290251067,
      "learning_rate": 0.00043298850742251013,
      "loss": 2.4824,
      "step": 1214
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.43243627756297714,
      "learning_rate": 0.0004322525887335563,
      "loss": 2.4355,
      "step": 1215
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.4073102201749991,
      "learning_rate": 0.00043151681952065734,
      "loss": 2.4805,
      "step": 1216
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.3949442273443569,
      "learning_rate": 0.00043078120140719456,
      "loss": 2.498,
      "step": 1217
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.4000551673492126,
      "learning_rate": 0.0004300457360162158,
      "loss": 2.4043,
      "step": 1218
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.4029436255985288,
      "learning_rate": 0.0004293104249704319,
      "loss": 2.5039,
      "step": 1219
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.36680969281044024,
      "learning_rate": 0.00042857526989221355,
      "loss": 2.4082,
      "step": 1220
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.3772701213128429,
      "learning_rate": 0.00042784027240358674,
      "loss": 2.498,
      "step": 1221
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.3917897643509441,
      "learning_rate": 0.0004271054341262301,
      "loss": 2.4395,
      "step": 1222
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.3812790974191783,
      "learning_rate": 0.000426370756681471,
      "loss": 2.4961,
      "step": 1223
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.3654839117365957,
      "learning_rate": 0.0004256362416902817,
      "loss": 2.4883,
      "step": 1224
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.36978975328353525,
      "learning_rate": 0.00042490189077327637,
      "loss": 2.4746,
      "step": 1225
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.38124062467717207,
      "learning_rate": 0.00042416770555070703,
      "loss": 2.4688,
      "step": 1226
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.38637079591990114,
      "learning_rate": 0.00042343368764246,
      "loss": 2.5059,
      "step": 1227
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.4046034402891046,
      "learning_rate": 0.0004226998386680524,
      "loss": 2.5391,
      "step": 1228
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.39498432720757864,
      "learning_rate": 0.000421966160246629,
      "loss": 2.4922,
      "step": 1229
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.37988781075327466,
      "learning_rate": 0.00042123265399695783,
      "loss": 2.3984,
      "step": 1230
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.389870712012582,
      "learning_rate": 0.0004204993215374273,
      "loss": 2.4746,
      "step": 1231
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.4156045772579545,
      "learning_rate": 0.00041976616448604226,
      "loss": 2.5645,
      "step": 1232
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.43029526168639043,
      "learning_rate": 0.00041903318446042076,
      "loss": 2.5059,
      "step": 1233
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.9708579456125126,
      "learning_rate": 0.00041830038307778984,
      "loss": 2.459,
      "step": 1234
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.38332751406397186,
      "learning_rate": 0.0004175677619549828,
      "loss": 2.4824,
      "step": 1235
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.3974532679045108,
      "learning_rate": 0.000416835322708435,
      "loss": 2.4199,
      "step": 1236
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.3784187167298708,
      "learning_rate": 0.00041610306695418056,
      "loss": 2.4238,
      "step": 1237
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.4294143299491514,
      "learning_rate": 0.0004153709963078488,
      "loss": 2.457,
      "step": 1238
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.40991033159782686,
      "learning_rate": 0.0004146391123846606,
      "loss": 2.5859,
      "step": 1239
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.376478004782134,
      "learning_rate": 0.0004139074167994249,
      "loss": 2.4727,
      "step": 1240
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.3794702949199632,
      "learning_rate": 0.00041317591116653486,
      "loss": 2.4824,
      "step": 1241
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.4376437556242257,
      "learning_rate": 0.0004124445970999648,
      "loss": 2.4473,
      "step": 1242
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.37376547900932655,
      "learning_rate": 0.00041171347621326627,
      "loss": 2.582,
      "step": 1243
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.3699891114447574,
      "learning_rate": 0.00041098255011956465,
      "loss": 2.4004,
      "step": 1244
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.3491913225906994,
      "learning_rate": 0.00041025182043155547,
      "loss": 2.5566,
      "step": 1245
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.3879555780031632,
      "learning_rate": 0.000409521288761501,
      "loss": 2.375,
      "step": 1246
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.4037988420384874,
      "learning_rate": 0.00040879095672122646,
      "loss": 2.375,
      "step": 1247
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.3966693592349872,
      "learning_rate": 0.0004080608259221167,
      "loss": 2.4531,
      "step": 1248
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.408382154705785,
      "learning_rate": 0.0004073308979751126,
      "loss": 2.5508,
      "step": 1249
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.3787142049009457,
      "learning_rate": 0.0004066011744907074,
      "loss": 2.4199,
      "step": 1250
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.37059528866300934,
      "learning_rate": 0.00040587165707894326,
      "loss": 2.4668,
      "step": 1251
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.39986701108107303,
      "learning_rate": 0.0004051423473494076,
      "loss": 2.4434,
      "step": 1252
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.3834833160771361,
      "learning_rate": 0.0004044132469112299,
      "loss": 2.3965,
      "step": 1253
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.45955945806709303,
      "learning_rate": 0.00040368435737307733,
      "loss": 2.4844,
      "step": 1254
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.4059679783005138,
      "learning_rate": 0.00040295568034315224,
      "loss": 2.5547,
      "step": 1255
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.3810664377849195,
      "learning_rate": 0.0004022272174291878,
      "loss": 2.5156,
      "step": 1256
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.40852360138906985,
      "learning_rate": 0.0004014989702384449,
      "loss": 2.3535,
      "step": 1257
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.41575822967271403,
      "learning_rate": 0.00040077094037770843,
      "loss": 2.5,
      "step": 1258
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.4011136518685322,
      "learning_rate": 0.0004000431294532838,
      "loss": 2.5156,
      "step": 1259
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.4044502306218274,
      "learning_rate": 0.0003993155390709935,
      "loss": 2.4727,
      "step": 1260
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.3686797625642993,
      "learning_rate": 0.0003985881708361729,
      "loss": 2.4375,
      "step": 1261
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.40163536920159737,
      "learning_rate": 0.00039786102635366784,
      "loss": 2.4473,
      "step": 1262
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.38472731477911465,
      "learning_rate": 0.0003971341072278302,
      "loss": 2.3984,
      "step": 1263
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.3901932893035491,
      "learning_rate": 0.00039640741506251457,
      "loss": 2.418,
      "step": 1264
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.39801046234421406,
      "learning_rate": 0.00039568095146107495,
      "loss": 2.5684,
      "step": 1265
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.3843938454075936,
      "learning_rate": 0.00039495471802636096,
      "loss": 2.5176,
      "step": 1266
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.39405399637821886,
      "learning_rate": 0.0003942287163607145,
      "loss": 2.4258,
      "step": 1267
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.41317579539688404,
      "learning_rate": 0.0003935029480659658,
      "loss": 2.4688,
      "step": 1268
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.40890993166166384,
      "learning_rate": 0.00039277741474343054,
      "loss": 2.4238,
      "step": 1269
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.3764410110718153,
      "learning_rate": 0.0003920521179939057,
      "loss": 2.5059,
      "step": 1270
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.4041015689127568,
      "learning_rate": 0.00039132705941766644,
      "loss": 2.4941,
      "step": 1271
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.4000356217977423,
      "learning_rate": 0.0003906022406144624,
      "loss": 2.3809,
      "step": 1272
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.4167700509182729,
      "learning_rate": 0.0003898776631835143,
      "loss": 2.4414,
      "step": 1273
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.3815377951736372,
      "learning_rate": 0.00038915332872350994,
      "loss": 2.334,
      "step": 1274
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.37496396652372105,
      "learning_rate": 0.00038842923883260135,
      "loss": 2.4219,
      "step": 1275
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.379922663278086,
      "learning_rate": 0.00038770539510840093,
      "loss": 2.5195,
      "step": 1276
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3830146717553855,
      "learning_rate": 0.00038698179914797783,
      "loss": 2.4453,
      "step": 1277
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3892903332705004,
      "learning_rate": 0.0003862584525478545,
      "loss": 2.4883,
      "step": 1278
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.39422968421903776,
      "learning_rate": 0.00038553535690400353,
      "loss": 2.3809,
      "step": 1279
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.38666924555023635,
      "learning_rate": 0.00038481251381184355,
      "loss": 2.3691,
      "step": 1280
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.38195880398799903,
      "learning_rate": 0.00038408992486623584,
      "loss": 2.4004,
      "step": 1281
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.37627869722354884,
      "learning_rate": 0.00038336759166148117,
      "loss": 2.4668,
      "step": 1282
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.4119951153473973,
      "learning_rate": 0.0003826455157913159,
      "loss": 2.4453,
      "step": 1283
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3804895568648521,
      "learning_rate": 0.00038192369884890886,
      "loss": 2.4668,
      "step": 1284
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.39425909818290655,
      "learning_rate": 0.00038120214242685723,
      "loss": 2.5391,
      "step": 1285
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.37038726050260756,
      "learning_rate": 0.00038048084811718373,
      "loss": 2.4062,
      "step": 1286
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3742745809637095,
      "learning_rate": 0.0003797598175113327,
      "loss": 2.4824,
      "step": 1287
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3989290138222216,
      "learning_rate": 0.0003790390522001662,
      "loss": 2.4629,
      "step": 1288
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3818419102274193,
      "learning_rate": 0.0003783185537739615,
      "loss": 2.5469,
      "step": 1289
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3771875668794178,
      "learning_rate": 0.00037759832382240697,
      "loss": 2.4766,
      "step": 1290
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.40494048376189434,
      "learning_rate": 0.00037687836393459826,
      "loss": 2.418,
      "step": 1291
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3822591504422196,
      "learning_rate": 0.0003761586756990354,
      "loss": 2.5469,
      "step": 1292
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.37194181030754225,
      "learning_rate": 0.0003754392607036191,
      "loss": 2.5137,
      "step": 1293
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.38247166279129746,
      "learning_rate": 0.0003747201205356472,
      "loss": 2.5,
      "step": 1294
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3570135460571694,
      "learning_rate": 0.0003740012567818111,
      "loss": 2.4883,
      "step": 1295
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.35672528571102496,
      "learning_rate": 0.0003732826710281922,
      "loss": 2.5215,
      "step": 1296
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3737285425851964,
      "learning_rate": 0.0003725643648602588,
      "loss": 2.4766,
      "step": 1297
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.3741730780579149,
      "learning_rate": 0.0003718463398628621,
      "loss": 2.4277,
      "step": 1298
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.37535752600415206,
      "learning_rate": 0.0003711285976202331,
      "loss": 2.4492,
      "step": 1299
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.37416561257048686,
      "learning_rate": 0.0003704111397159787,
      "loss": 2.4258,
      "step": 1300
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.40088252933934926,
      "learning_rate": 0.0003696939677330788,
      "loss": 2.4199,
      "step": 1301
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.36938383582288725,
      "learning_rate": 0.00036897708325388213,
      "loss": 2.4395,
      "step": 1302
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.38142720740820957,
      "learning_rate": 0.0003682604878601034,
      "loss": 2.3809,
      "step": 1303
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.38642571334106807,
      "learning_rate": 0.000367544183132819,
      "loss": 2.4414,
      "step": 1304
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.4163210792915275,
      "learning_rate": 0.0003668281706524645,
      "loss": 2.375,
      "step": 1305
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.4113635841342351,
      "learning_rate": 0.0003661124519988304,
      "loss": 2.4336,
      "step": 1306
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.39545226148502177,
      "learning_rate": 0.00036539702875105893,
      "loss": 2.4316,
      "step": 1307
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.3928141414538381,
      "learning_rate": 0.0003646819024876406,
      "loss": 2.3672,
      "step": 1308
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.4163811464032783,
      "learning_rate": 0.0003639670747864107,
      "loss": 2.4277,
      "step": 1309
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.3846496290546263,
      "learning_rate": 0.00036325254722454584,
      "loss": 2.4941,
      "step": 1310
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.39984951824633874,
      "learning_rate": 0.00036253832137855997,
      "loss": 2.4395,
      "step": 1311
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.37613275383080125,
      "learning_rate": 0.00036182439882430183,
      "loss": 2.498,
      "step": 1312
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.3804329238380219,
      "learning_rate": 0.00036111078113695096,
      "loss": 2.4707,
      "step": 1313
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.38491413164496757,
      "learning_rate": 0.0003603974698910139,
      "loss": 2.4844,
      "step": 1314
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.3734358873791372,
      "learning_rate": 0.0003596844666603214,
      "loss": 2.4219,
      "step": 1315
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.3952186816136156,
      "learning_rate": 0.0003589717730180245,
      "loss": 2.4238,
      "step": 1316
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.4008908629376709,
      "learning_rate": 0.00035825939053659117,
      "loss": 2.4316,
      "step": 1317
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.39543017152192134,
      "learning_rate": 0.00035754732078780273,
      "loss": 2.502,
      "step": 1318
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.3810169468661811,
      "learning_rate": 0.00035683556534275076,
      "loss": 2.4844,
      "step": 1319
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.37636724534619503,
      "learning_rate": 0.00035612412577183303,
      "loss": 2.4082,
      "step": 1320
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.39023775896296287,
      "learning_rate": 0.00035541300364475063,
      "loss": 2.4805,
      "step": 1321
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.39543245853825343,
      "learning_rate": 0.0003547022005305043,
      "loss": 2.4395,
      "step": 1322
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3803762630214228,
      "learning_rate": 0.0003539917179973907,
      "loss": 2.4707,
      "step": 1323
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.37016340781943596,
      "learning_rate": 0.00035328155761299917,
      "loss": 2.4961,
      "step": 1324
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.37003315931711844,
      "learning_rate": 0.0003525717209442085,
      "loss": 2.5098,
      "step": 1325
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.4005560323207818,
      "learning_rate": 0.00035186220955718306,
      "loss": 2.4512,
      "step": 1326
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3835461453428004,
      "learning_rate": 0.0003511530250173696,
      "loss": 2.498,
      "step": 1327
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.38044809553955433,
      "learning_rate": 0.00035044416888949364,
      "loss": 2.4902,
      "step": 1328
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3907777243319403,
      "learning_rate": 0.0003497356427375562,
      "loss": 2.5,
      "step": 1329
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.37442402768100524,
      "learning_rate": 0.00034902744812483034,
      "loss": 2.5625,
      "step": 1330
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3697743425766204,
      "learning_rate": 0.00034831958661385714,
      "loss": 2.5098,
      "step": 1331
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.38446729603672364,
      "learning_rate": 0.0003476120597664434,
      "loss": 2.416,
      "step": 1332
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.37388203832019773,
      "learning_rate": 0.00034690486914365704,
      "loss": 2.5156,
      "step": 1333
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3784450516699846,
      "learning_rate": 0.00034619801630582435,
      "loss": 2.4395,
      "step": 1334
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.4226843806135174,
      "learning_rate": 0.00034549150281252633,
      "loss": 2.5293,
      "step": 1335
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.36531243619224063,
      "learning_rate": 0.0003447853302225952,
      "loss": 2.4297,
      "step": 1336
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3915689221713253,
      "learning_rate": 0.00034407950009411126,
      "loss": 2.5527,
      "step": 1337
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3804715239970928,
      "learning_rate": 0.00034337401398439873,
      "loss": 2.4707,
      "step": 1338
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.37734698519328086,
      "learning_rate": 0.00034266887345002305,
      "loss": 2.4512,
      "step": 1339
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3823414058540003,
      "learning_rate": 0.0003419640800467874,
      "loss": 2.4473,
      "step": 1340
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3673406340523773,
      "learning_rate": 0.0003412596353297288,
      "loss": 2.4258,
      "step": 1341
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.37572662441683835,
      "learning_rate": 0.00034055554085311493,
      "loss": 2.5723,
      "step": 1342
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.390644309415998,
      "learning_rate": 0.00033985179817044105,
      "loss": 2.5098,
      "step": 1343
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3994547295240882,
      "learning_rate": 0.0003391484088344257,
      "loss": 2.4609,
      "step": 1344
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3641830084205201,
      "learning_rate": 0.00033844537439700807,
      "loss": 2.377,
      "step": 1345
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.36774908880277374,
      "learning_rate": 0.00033774269640934445,
      "loss": 2.5332,
      "step": 1346
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3649039940073676,
      "learning_rate": 0.0003370403764218045,
      "loss": 2.4316,
      "step": 1347
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.398223554725233,
      "learning_rate": 0.000336338415983968,
      "loss": 2.4707,
      "step": 1348
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3788570465122782,
      "learning_rate": 0.00033563681664462155,
      "loss": 2.4004,
      "step": 1349
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3982684550233149,
      "learning_rate": 0.000334935579951755,
      "loss": 2.5703,
      "step": 1350
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.37306061233041127,
      "learning_rate": 0.0003342347074525578,
      "loss": 2.3828,
      "step": 1351
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.39156598798489056,
      "learning_rate": 0.0003335342006934161,
      "loss": 2.4551,
      "step": 1352
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3989991158726481,
      "learning_rate": 0.00033283406121990914,
      "loss": 2.5332,
      "step": 1353
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3812533289720542,
      "learning_rate": 0.0003321342905768057,
      "loss": 2.498,
      "step": 1354
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.36965986657429084,
      "learning_rate": 0.00033143489030806086,
      "loss": 2.4219,
      "step": 1355
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3846426981653418,
      "learning_rate": 0.00033073586195681227,
      "loss": 2.4219,
      "step": 1356
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3751882514659656,
      "learning_rate": 0.00033003720706537736,
      "loss": 2.4141,
      "step": 1357
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.40996661300987813,
      "learning_rate": 0.0003293389271752492,
      "loss": 2.3145,
      "step": 1358
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3864765678137452,
      "learning_rate": 0.00032864102382709374,
      "loss": 2.376,
      "step": 1359
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.4096499196539133,
      "learning_rate": 0.000327943498560746,
      "loss": 2.4551,
      "step": 1360
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.40532681655086683,
      "learning_rate": 0.00032724635291520694,
      "loss": 2.4102,
      "step": 1361
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.40869363824913957,
      "learning_rate": 0.00032654958842863967,
      "loss": 2.3926,
      "step": 1362
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.39697134864804096,
      "learning_rate": 0.0003258532066383667,
      "loss": 2.3652,
      "step": 1363
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4248209694298927,
      "learning_rate": 0.000325157209080866,
      "loss": 2.4395,
      "step": 1364
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4198923410381533,
      "learning_rate": 0.00032446159729176743,
      "loss": 2.5215,
      "step": 1365
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.42403204475438855,
      "learning_rate": 0.0003237663728058502,
      "loss": 2.4941,
      "step": 1366
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.391524255002629,
      "learning_rate": 0.0003230715371570389,
      "loss": 2.3438,
      "step": 1367
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.37564253258544084,
      "learning_rate": 0.00032237709187839996,
      "loss": 2.4453,
      "step": 1368
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4095157898610107,
      "learning_rate": 0.0003216830385021388,
      "loss": 2.5996,
      "step": 1369
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.40206497045500267,
      "learning_rate": 0.0003209893785595959,
      "loss": 2.3984,
      "step": 1370
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4053316535558721,
      "learning_rate": 0.00032029611358124366,
      "loss": 2.5449,
      "step": 1371
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.3960398383502729,
      "learning_rate": 0.00031960324509668336,
      "loss": 2.4648,
      "step": 1372
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4115378848210557,
      "learning_rate": 0.0003189107746346412,
      "loss": 2.4629,
      "step": 1373
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.3882788742889099,
      "learning_rate": 0.0003182187037229653,
      "loss": 2.4648,
      "step": 1374
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.3967600570758219,
      "learning_rate": 0.0003175270338886221,
      "loss": 2.4922,
      "step": 1375
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.3720870082312021,
      "learning_rate": 0.00031683576665769345,
      "loss": 2.4277,
      "step": 1376
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.410079874083045,
      "learning_rate": 0.0003161449035553724,
      "loss": 2.4648,
      "step": 1377
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.38796623317898943,
      "learning_rate": 0.00031545444610596077,
      "loss": 2.4199,
      "step": 1378
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.39404762800350274,
      "learning_rate": 0.000314764395832865,
      "loss": 2.4375,
      "step": 1379
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.42245742218640053,
      "learning_rate": 0.0003140747542585934,
      "loss": 2.4082,
      "step": 1380
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.39204095104946063,
      "learning_rate": 0.00031338552290475266,
      "loss": 2.4219,
      "step": 1381
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.39406155410352167,
      "learning_rate": 0.00031269670329204396,
      "loss": 2.4375,
      "step": 1382
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.40659238427129935,
      "learning_rate": 0.0003120082969402604,
      "loss": 2.4355,
      "step": 1383
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.37891751134681984,
      "learning_rate": 0.00031132030536828314,
      "loss": 2.5215,
      "step": 1384
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.39564416251090284,
      "learning_rate": 0.00031063273009407805,
      "loss": 2.4844,
      "step": 1385
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.38994296001594797,
      "learning_rate": 0.00030994557263469265,
      "loss": 2.4805,
      "step": 1386
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3829461461689725,
      "learning_rate": 0.0003092588345062526,
      "loss": 2.4492,
      "step": 1387
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3840437863086631,
      "learning_rate": 0.0003085725172239582,
      "loss": 2.4492,
      "step": 1388
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.40966656061448803,
      "learning_rate": 0.0003078866223020815,
      "loss": 2.5195,
      "step": 1389
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.40588834673968527,
      "learning_rate": 0.0003072011512539624,
      "loss": 2.4805,
      "step": 1390
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.4015263973454593,
      "learning_rate": 0.00030651610559200574,
      "loss": 2.3848,
      "step": 1391
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.39833440203525716,
      "learning_rate": 0.00030583148682767757,
      "loss": 2.418,
      "step": 1392
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.38907713113616743,
      "learning_rate": 0.00030514729647150243,
      "loss": 2.4531,
      "step": 1393
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3747407829408176,
      "learning_rate": 0.0003044635360330592,
      "loss": 2.4707,
      "step": 1394
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.38081625822734066,
      "learning_rate": 0.00030378020702097845,
      "loss": 2.5,
      "step": 1395
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.376797721260927,
      "learning_rate": 0.000303097310942939,
      "loss": 2.3438,
      "step": 1396
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.40116220155155063,
      "learning_rate": 0.0003024148493056641,
      "loss": 2.4531,
      "step": 1397
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.4074043504317936,
      "learning_rate": 0.00030173282361491865,
      "loss": 2.4805,
      "step": 1398
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3805265071264406,
      "learning_rate": 0.0003010512353755057,
      "loss": 2.5117,
      "step": 1399
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3877909362473007,
      "learning_rate": 0.00030037008609126313,
      "loss": 2.4805,
      "step": 1400
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3993531908807153,
      "learning_rate": 0.0002996893772650602,
      "loss": 2.5098,
      "step": 1401
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.4054498218276188,
      "learning_rate": 0.0002990091103987945,
      "loss": 2.4805,
      "step": 1402
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.40607571782213886,
      "learning_rate": 0.0002983292869933886,
      "loss": 2.5293,
      "step": 1403
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.401764926336853,
      "learning_rate": 0.0002976499085487862,
      "loss": 2.4492,
      "step": 1404
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3894236438752315,
      "learning_rate": 0.00029697097656394963,
      "loss": 2.4648,
      "step": 1405
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.4020694427199162,
      "learning_rate": 0.00029629249253685595,
      "loss": 2.4102,
      "step": 1406
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.39940659523608124,
      "learning_rate": 0.00029561445796449416,
      "loss": 2.4531,
      "step": 1407
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.40321699878976364,
      "learning_rate": 0.0002949368743428612,
      "loss": 2.3613,
      "step": 1408
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.4262596541065508,
      "learning_rate": 0.0002942597431669593,
      "loss": 2.4453,
      "step": 1409
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.39175412316295116,
      "learning_rate": 0.0002935830659307924,
      "loss": 2.4727,
      "step": 1410
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.41137069759649936,
      "learning_rate": 0.0002929068441273629,
      "loss": 2.4141,
      "step": 1411
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.4108469848996005,
      "learning_rate": 0.0002922310792486681,
      "loss": 2.4375,
      "step": 1412
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.42341695035743526,
      "learning_rate": 0.00029155577278569745,
      "loss": 2.4238,
      "step": 1413
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.3906920189288247,
      "learning_rate": 0.00029088092622842895,
      "loss": 2.4297,
      "step": 1414
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.4027905175102389,
      "learning_rate": 0.00029020654106582544,
      "loss": 2.4688,
      "step": 1415
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.40037605619674577,
      "learning_rate": 0.0002895326187858326,
      "loss": 2.4375,
      "step": 1416
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.3832213403373976,
      "learning_rate": 0.00028885916087537377,
      "loss": 2.3906,
      "step": 1417
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.4011452937733642,
      "learning_rate": 0.00028818616882034877,
      "loss": 2.4668,
      "step": 1418
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.41422815596662044,
      "learning_rate": 0.0002875136441056286,
      "loss": 2.4102,
      "step": 1419
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.3880070007255337,
      "learning_rate": 0.000286841588215054,
      "loss": 2.4219,
      "step": 1420
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.39446089711576926,
      "learning_rate": 0.0002861700026314308,
      "loss": 2.459,
      "step": 1421
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.3975518357586395,
      "learning_rate": 0.00028549888883652686,
      "loss": 2.4863,
      "step": 1422
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.3782470283195577,
      "learning_rate": 0.00028482824831107,
      "loss": 2.3652,
      "step": 1423
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.3707486331683162,
      "learning_rate": 0.000284158082534743,
      "loss": 2.4805,
      "step": 1424
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.3953633574794342,
      "learning_rate": 0.00028348839298618177,
      "loss": 2.4238,
      "step": 1425
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.42198440952359945,
      "learning_rate": 0.0002828191811429709,
      "loss": 2.418,
      "step": 1426
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.40103712862458935,
      "learning_rate": 0.00028215044848164164,
      "loss": 2.5176,
      "step": 1427
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.41186547046161665,
      "learning_rate": 0.00028148219647766747,
      "loss": 2.4551,
      "step": 1428
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.39255821639556754,
      "learning_rate": 0.00028081442660546124,
      "loss": 2.5156,
      "step": 1429
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.38600439458256386,
      "learning_rate": 0.0002801471403383728,
      "loss": 2.4023,
      "step": 1430
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.3907649635113824,
      "learning_rate": 0.00027948033914868415,
      "loss": 2.3809,
      "step": 1431
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.3814169786605639,
      "learning_rate": 0.00027881402450760775,
      "loss": 2.4766,
      "step": 1432
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.40091586305624966,
      "learning_rate": 0.00027814819788528165,
      "loss": 2.4062,
      "step": 1433
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.4197429600886839,
      "learning_rate": 0.00027748286075076836,
      "loss": 2.4785,
      "step": 1434
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.3881004107167376,
      "learning_rate": 0.00027681801457204937,
      "loss": 2.4199,
      "step": 1435
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.3908449094875445,
      "learning_rate": 0.00027615366081602306,
      "loss": 2.4629,
      "step": 1436
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.37652263398023633,
      "learning_rate": 0.0002754898009485021,
      "loss": 2.5293,
      "step": 1437
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.39359173026528305,
      "learning_rate": 0.0002748264364342085,
      "loss": 2.3867,
      "step": 1438
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.4130191797649521,
      "learning_rate": 0.00027416356873677204,
      "loss": 2.5195,
      "step": 1439
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.39169062935754007,
      "learning_rate": 0.0002735011993187258,
      "loss": 2.4688,
      "step": 1440
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.37961403450182524,
      "learning_rate": 0.0002728393296415042,
      "loss": 2.3457,
      "step": 1441
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.3837926629145711,
      "learning_rate": 0.00027217796116543817,
      "loss": 2.3984,
      "step": 1442
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.4113822268111202,
      "learning_rate": 0.0002715170953497532,
      "loss": 2.4219,
      "step": 1443
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.39802143803375967,
      "learning_rate": 0.00027085673365256614,
      "loss": 2.3633,
      "step": 1444
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.3976227117834746,
      "learning_rate": 0.00027019687753088075,
      "loss": 2.3906,
      "step": 1445
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.40506857665876767,
      "learning_rate": 0.00026953752844058597,
      "loss": 2.5039,
      "step": 1446
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.3869627939789709,
      "learning_rate": 0.0002688786878364516,
      "loss": 2.4238,
      "step": 1447
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.40597936487315744,
      "learning_rate": 0.00026822035717212597,
      "loss": 2.3477,
      "step": 1448
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.41597901116578323,
      "learning_rate": 0.00026756253790013193,
      "loss": 2.4824,
      "step": 1449
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.452823013919672,
      "learning_rate": 0.0002669052314718641,
      "loss": 2.377,
      "step": 1450
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.42929305786957955,
      "learning_rate": 0.0002662484393375855,
      "loss": 2.459,
      "step": 1451
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.41597704886231074,
      "learning_rate": 0.00026559216294642446,
      "loss": 2.4434,
      "step": 1452
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.4178102431708016,
      "learning_rate": 0.0002649364037463718,
      "loss": 2.4785,
      "step": 1453
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.4061136736616461,
      "learning_rate": 0.0002642811631842764,
      "loss": 2.4785,
      "step": 1454
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3998130963224607,
      "learning_rate": 0.0002636264427058439,
      "loss": 2.4941,
      "step": 1455
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.40483844907256644,
      "learning_rate": 0.00026297224375563123,
      "loss": 2.5684,
      "step": 1456
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.40919449493204646,
      "learning_rate": 0.00026231856777704575,
      "loss": 2.3945,
      "step": 1457
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3728367465438491,
      "learning_rate": 0.00026166541621234026,
      "loss": 2.4434,
      "step": 1458
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.38190706021619947,
      "learning_rate": 0.00026101279050261045,
      "loss": 2.4844,
      "step": 1459
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.37868430728490143,
      "learning_rate": 0.00026036069208779247,
      "loss": 2.4668,
      "step": 1460
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.386631876782959,
      "learning_rate": 0.0002597091224066581,
      "loss": 2.3574,
      "step": 1461
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3808065403762871,
      "learning_rate": 0.00025905808289681365,
      "loss": 2.418,
      "step": 1462
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.409399023879909,
      "learning_rate": 0.0002584075749946946,
      "loss": 2.3359,
      "step": 1463
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3961411379101835,
      "learning_rate": 0.00025775760013556424,
      "loss": 2.4922,
      "step": 1464
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3889902208185602,
      "learning_rate": 0.0002571081597535095,
      "loss": 2.3633,
      "step": 1465
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3933191939672592,
      "learning_rate": 0.00025645925528143776,
      "loss": 2.4023,
      "step": 1466
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3894980551664617,
      "learning_rate": 0.0002558108881510747,
      "loss": 2.5391,
      "step": 1467
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.38919989851556847,
      "learning_rate": 0.00025516305979295963,
      "loss": 2.4395,
      "step": 1468
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.39085477646072464,
      "learning_rate": 0.0002545157716364439,
      "loss": 2.4492,
      "step": 1469
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.39185960925462826,
      "learning_rate": 0.00025386902510968624,
      "loss": 2.416,
      "step": 1470
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.39659619226379117,
      "learning_rate": 0.00025322282163965095,
      "loss": 2.4668,
      "step": 1471
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3945479407179201,
      "learning_rate": 0.00025257716265210384,
      "loss": 2.4199,
      "step": 1472
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.40362230421073964,
      "learning_rate": 0.0002519320495716091,
      "loss": 2.3691,
      "step": 1473
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.4032505817379387,
      "learning_rate": 0.00025128748382152716,
      "loss": 2.4434,
      "step": 1474
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.3894450000115197,
      "learning_rate": 0.00025064346682401016,
      "loss": 2.3867,
      "step": 1475
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.39333642946061137,
      "learning_rate": 0.0002500000000000001,
      "loss": 2.4961,
      "step": 1476
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.380642888740323,
      "learning_rate": 0.0002493570847692246,
      "loss": 2.3633,
      "step": 1477
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.39465347532510586,
      "learning_rate": 0.00024871472255019424,
      "loss": 2.4219,
      "step": 1478
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.3846566680911326,
      "learning_rate": 0.00024807291476019994,
      "loss": 2.4941,
      "step": 1479
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.4054112467227602,
      "learning_rate": 0.00024743166281530877,
      "loss": 2.4395,
      "step": 1480
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.40120037490618027,
      "learning_rate": 0.000246790968130362,
      "loss": 2.498,
      "step": 1481
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.39366313873804754,
      "learning_rate": 0.0002461508321189706,
      "loss": 2.4785,
      "step": 1482
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.36656519666783105,
      "learning_rate": 0.00024551125619351385,
      "loss": 2.4492,
      "step": 1483
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.3692021299773204,
      "learning_rate": 0.00024487224176513453,
      "loss": 2.4805,
      "step": 1484
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.39437678257760655,
      "learning_rate": 0.0002442337902437365,
      "loss": 2.4258,
      "step": 1485
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.39739687843326466,
      "learning_rate": 0.0002435959030379824,
      "loss": 2.4316,
      "step": 1486
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.37374593519983706,
      "learning_rate": 0.00024295858155528888,
      "loss": 2.4551,
      "step": 1487
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.396486844327594,
      "learning_rate": 0.00024232182720182523,
      "loss": 2.4844,
      "step": 1488
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.4154806509382246,
      "learning_rate": 0.00024168564138250855,
      "loss": 2.502,
      "step": 1489
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.3965120903379412,
      "learning_rate": 0.00024105002550100246,
      "loss": 2.502,
      "step": 1490
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.4002575623421327,
      "learning_rate": 0.00024041498095971254,
      "loss": 2.3984,
      "step": 1491
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.3963232737104374,
      "learning_rate": 0.0002397805091597835,
      "loss": 2.4766,
      "step": 1492
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.38844060404464975,
      "learning_rate": 0.0002391466115010973,
      "loss": 2.5039,
      "step": 1493
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.38271076218345434,
      "learning_rate": 0.00023851328938226808,
      "loss": 2.3848,
      "step": 1494
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.41630839786697194,
      "learning_rate": 0.00023788054420064109,
      "loss": 2.4844,
      "step": 1495
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.4186225231278926,
      "learning_rate": 0.00023724837735228773,
      "loss": 2.4512,
      "step": 1496
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.41552413664352295,
      "learning_rate": 0.00023661679023200422,
      "loss": 2.4414,
      "step": 1497
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.3865520538414148,
      "learning_rate": 0.00023598578423330714,
      "loss": 2.3594,
      "step": 1498
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.3992860270864843,
      "learning_rate": 0.00023535536074843083,
      "loss": 2.5137,
      "step": 1499
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.4102996417285064,
      "learning_rate": 0.00023472552116832502,
      "loss": 2.4199,
      "step": 1500
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.38962386949098593,
      "learning_rate": 0.0002340962668826503,
      "loss": 2.3633,
      "step": 1501
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.3945745303355589,
      "learning_rate": 0.00023346759927977663,
      "loss": 2.4316,
      "step": 1502
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.4091405297846134,
      "learning_rate": 0.0002328395197467789,
      "loss": 2.3984,
      "step": 1503
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.4057042905957915,
      "learning_rate": 0.00023221202966943515,
      "loss": 2.4609,
      "step": 1504
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.39834408221534806,
      "learning_rate": 0.0002315851304322223,
      "loss": 2.4961,
      "step": 1505
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.39158954562533943,
      "learning_rate": 0.0002309588234183137,
      "loss": 2.4375,
      "step": 1506
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.39392756235404325,
      "learning_rate": 0.00023033311000957653,
      "loss": 2.3926,
      "step": 1507
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.39453995078060167,
      "learning_rate": 0.00022970799158656758,
      "loss": 2.3691,
      "step": 1508
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.38132927891676194,
      "learning_rate": 0.0002290834695285316,
      "loss": 2.457,
      "step": 1509
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.3813660416646438,
      "learning_rate": 0.00022845954521339678,
      "loss": 2.4375,
      "step": 1510
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.4105227416800663,
      "learning_rate": 0.0002278362200177732,
      "loss": 2.4336,
      "step": 1511
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.40149996454142345,
      "learning_rate": 0.00022721349531694852,
      "loss": 2.3711,
      "step": 1512
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.3940951508231731,
      "learning_rate": 0.0002265913724848855,
      "loss": 2.4434,
      "step": 1513
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.39614882438685894,
      "learning_rate": 0.00022596985289421946,
      "loss": 2.5508,
      "step": 1514
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.39201095452202567,
      "learning_rate": 0.00022534893791625405,
      "loss": 2.3418,
      "step": 1515
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3829711011441404,
      "learning_rate": 0.00022472862892095968,
      "loss": 2.3262,
      "step": 1516
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.4155857773378795,
      "learning_rate": 0.00022410892727696896,
      "loss": 2.4688,
      "step": 1517
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.40802972142932936,
      "learning_rate": 0.0002234898343515751,
      "loss": 2.4805,
      "step": 1518
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3921465814110065,
      "learning_rate": 0.00022287135151072792,
      "loss": 2.3281,
      "step": 1519
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3854933760546064,
      "learning_rate": 0.00022225348011903096,
      "loss": 2.3906,
      "step": 1520
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.36938497642761936,
      "learning_rate": 0.0002216362215397393,
      "loss": 2.3789,
      "step": 1521
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3902707781932321,
      "learning_rate": 0.00022101957713475522,
      "loss": 2.5703,
      "step": 1522
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3918652199106501,
      "learning_rate": 0.00022040354826462666,
      "loss": 2.4629,
      "step": 1523
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.39006337352483195,
      "learning_rate": 0.0002197881362885426,
      "loss": 2.3496,
      "step": 1524
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.4136254320136678,
      "learning_rate": 0.0002191733425643318,
      "loss": 2.4492,
      "step": 1525
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.39365700882900345,
      "learning_rate": 0.00021855916844845826,
      "loss": 2.4434,
      "step": 1526
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.41380102295696675,
      "learning_rate": 0.00021794561529601898,
      "loss": 2.4727,
      "step": 1527
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3944092375027099,
      "learning_rate": 0.00021733268446074138,
      "loss": 2.5,
      "step": 1528
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.4208280056768133,
      "learning_rate": 0.00021672037729497917,
      "loss": 2.4355,
      "step": 1529
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.39294642333512897,
      "learning_rate": 0.0002161086951497106,
      "loss": 2.418,
      "step": 1530
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.4130908591516093,
      "learning_rate": 0.00021549763937453442,
      "loss": 2.3535,
      "step": 1531
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3897212550234977,
      "learning_rate": 0.00021488721131766736,
      "loss": 2.4395,
      "step": 1532
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.403200823783983,
      "learning_rate": 0.00021427741232594183,
      "loss": 2.4766,
      "step": 1533
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.4044097810979833,
      "learning_rate": 0.0002136682437448013,
      "loss": 2.5078,
      "step": 1534
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3897582942878178,
      "learning_rate": 0.0002130597069182994,
      "loss": 2.3906,
      "step": 1535
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.40245754586618054,
      "learning_rate": 0.0002124518031890948,
      "loss": 2.3672,
      "step": 1536
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3962817075622861,
      "learning_rate": 0.0002118445338984502,
      "loss": 2.4512,
      "step": 1537
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.39108522498479625,
      "learning_rate": 0.00021123790038622808,
      "loss": 2.3223,
      "step": 1538
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.4001615373376366,
      "learning_rate": 0.0002106319039908879,
      "loss": 2.4551,
      "step": 1539
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.41069275963179014,
      "learning_rate": 0.00021002654604948412,
      "loss": 2.4785,
      "step": 1540
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.433923399467506,
      "learning_rate": 0.00020942182789766172,
      "loss": 2.4258,
      "step": 1541
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.40214733289038446,
      "learning_rate": 0.00020881775086965492,
      "loss": 2.3633,
      "step": 1542
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.42421932592811706,
      "learning_rate": 0.00020821431629828246,
      "loss": 2.3477,
      "step": 1543
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.40617511555128233,
      "learning_rate": 0.00020761152551494643,
      "loss": 2.5742,
      "step": 1544
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.39992730840170365,
      "learning_rate": 0.00020700937984962798,
      "loss": 2.3965,
      "step": 1545
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.3920924424816319,
      "learning_rate": 0.0002064078806308848,
      "loss": 2.377,
      "step": 1546
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.43518895352669096,
      "learning_rate": 0.00020580702918584882,
      "loss": 2.4648,
      "step": 1547
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5282261294824463,
      "learning_rate": 0.000205206826840222,
      "loss": 2.4512,
      "step": 1548
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.39629153346615165,
      "learning_rate": 0.0002046072749182751,
      "loss": 2.4199,
      "step": 1549
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.39757948374774893,
      "learning_rate": 0.00020400837474284273,
      "loss": 2.3652,
      "step": 1550
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.38325760371909656,
      "learning_rate": 0.0002034101276353224,
      "loss": 2.3906,
      "step": 1551
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.40001901246650423,
      "learning_rate": 0.00020281253491567027,
      "loss": 2.4102,
      "step": 1552
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.45607592323432816,
      "learning_rate": 0.0002022155979023984,
      "loss": 2.5137,
      "step": 1553
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.3997611957467963,
      "learning_rate": 0.000201619317912573,
      "loss": 2.4785,
      "step": 1554
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.4017348030109825,
      "learning_rate": 0.00020102369626180962,
      "loss": 2.5,
      "step": 1555
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.38402854074787257,
      "learning_rate": 0.0002004287342642721,
      "loss": 2.4883,
      "step": 1556
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.41379946873738405,
      "learning_rate": 0.00019983443323266824,
      "loss": 2.4316,
      "step": 1557
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.40502164419864933,
      "learning_rate": 0.00019924079447824805,
      "loss": 2.4668,
      "step": 1558
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.38251216510864344,
      "learning_rate": 0.00019864781931079977,
      "loss": 2.4766,
      "step": 1559
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.413275179551919,
      "learning_rate": 0.00019805550903864773,
      "loss": 2.5391,
      "step": 1560
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.39987117782129133,
      "learning_rate": 0.00019746386496864948,
      "loss": 2.4629,
      "step": 1561
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.38681149905770296,
      "learning_rate": 0.00019687288840619226,
      "loss": 2.4805,
      "step": 1562
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.3849136696600826,
      "learning_rate": 0.0001962825806551911,
      "loss": 2.4043,
      "step": 1563
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.40917750751476833,
      "learning_rate": 0.0001956929430180846,
      "loss": 2.4141,
      "step": 1564
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.39198102285646824,
      "learning_rate": 0.00019510397679583374,
      "loss": 2.3652,
      "step": 1565
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.38115528212243494,
      "learning_rate": 0.0001945156832879174,
      "loss": 2.3984,
      "step": 1566
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.3978284284630193,
      "learning_rate": 0.00019392806379233036,
      "loss": 2.5195,
      "step": 1567
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.389565981808393,
      "learning_rate": 0.00019334111960558065,
      "loss": 2.4375,
      "step": 1568
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.38474799595514414,
      "learning_rate": 0.00019275485202268573,
      "loss": 2.4121,
      "step": 1569
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.3955682763856454,
      "learning_rate": 0.00019216926233717085,
      "loss": 2.4023,
      "step": 1570
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.40765661895332667,
      "learning_rate": 0.00019158435184106498,
      "loss": 2.4668,
      "step": 1571
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.39693209307020233,
      "learning_rate": 0.00019100012182489905,
      "loss": 2.4004,
      "step": 1572
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.4595611241457044,
      "learning_rate": 0.00019041657357770226,
      "loss": 2.3984,
      "step": 1573
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.43052203892242824,
      "learning_rate": 0.00018983370838699943,
      "loss": 2.4668,
      "step": 1574
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.4057846393521001,
      "learning_rate": 0.00018925152753880892,
      "loss": 2.4902,
      "step": 1575
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.410220480470681,
      "learning_rate": 0.00018867003231763847,
      "loss": 2.3809,
      "step": 1576
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.414914125964014,
      "learning_rate": 0.00018808922400648375,
      "loss": 2.4727,
      "step": 1577
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.4084845117350386,
      "learning_rate": 0.00018750910388682428,
      "loss": 2.4766,
      "step": 1578
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.3910417982190957,
      "learning_rate": 0.00018692967323862125,
      "loss": 2.3926,
      "step": 1579
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.3903653482906688,
      "learning_rate": 0.00018635093334031517,
      "loss": 2.4102,
      "step": 1580
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.4078679888600526,
      "learning_rate": 0.00018577288546882165,
      "loss": 2.3926,
      "step": 1581
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.40082923050524893,
      "learning_rate": 0.00018519553089953023,
      "loss": 2.4414,
      "step": 1582
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.39242062757307283,
      "learning_rate": 0.0001846188709063001,
      "loss": 2.4844,
      "step": 1583
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4183420719703342,
      "learning_rate": 0.00018404290676145857,
      "loss": 2.3633,
      "step": 1584
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.39448680736546615,
      "learning_rate": 0.00018346763973579722,
      "loss": 2.4492,
      "step": 1585
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4107031997876655,
      "learning_rate": 0.00018289307109856939,
      "loss": 2.4766,
      "step": 1586
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.40874035197791647,
      "learning_rate": 0.0001823192021174882,
      "loss": 2.3945,
      "step": 1587
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.3921408435635881,
      "learning_rate": 0.0001817460340587223,
      "loss": 2.4355,
      "step": 1588
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4145479274484156,
      "learning_rate": 0.00018117356818689445,
      "loss": 2.4414,
      "step": 1589
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4080667950389335,
      "learning_rate": 0.00018060180576507756,
      "loss": 2.3828,
      "step": 1590
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.39180117135708753,
      "learning_rate": 0.00018003074805479313,
      "loss": 2.3789,
      "step": 1591
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4052099554485318,
      "learning_rate": 0.00017946039631600724,
      "loss": 2.4668,
      "step": 1592
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4315530844770923,
      "learning_rate": 0.00017889075180712837,
      "loss": 2.5176,
      "step": 1593
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4280651859432795,
      "learning_rate": 0.00017832181578500512,
      "loss": 2.2266,
      "step": 1594
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4068452600947267,
      "learning_rate": 0.0001777535895049221,
      "loss": 2.5605,
      "step": 1595
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.39115205839980943,
      "learning_rate": 0.0001771860742205988,
      "loss": 2.4688,
      "step": 1596
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.39815390075307316,
      "learning_rate": 0.00017661927118418525,
      "loss": 2.4297,
      "step": 1597
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4306912990691876,
      "learning_rate": 0.00017605318164626066,
      "loss": 2.3945,
      "step": 1598
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.41243597275346966,
      "learning_rate": 0.00017548780685582949,
      "loss": 2.4883,
      "step": 1599
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4319440775070539,
      "learning_rate": 0.00017492314806031922,
      "loss": 2.4551,
      "step": 1600
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.38373276728212613,
      "learning_rate": 0.00017435920650557806,
      "loss": 2.4238,
      "step": 1601
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4006152295719899,
      "learning_rate": 0.00017379598343587112,
      "loss": 2.4004,
      "step": 1602
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.3782892417383588,
      "learning_rate": 0.00017323348009387878,
      "loss": 2.3848,
      "step": 1603
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.41511896244383545,
      "learning_rate": 0.0001726716977206929,
      "loss": 2.4473,
      "step": 1604
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.43087438242624465,
      "learning_rate": 0.00017211063755581525,
      "loss": 2.5254,
      "step": 1605
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.426288087758836,
      "learning_rate": 0.0001715503008371536,
      "loss": 2.4102,
      "step": 1606
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.4629291119676336,
      "learning_rate": 0.0001709906888010196,
      "loss": 2.5156,
      "step": 1607
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.4358766810516288,
      "learning_rate": 0.00017043180268212638,
      "loss": 2.4336,
      "step": 1608
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.4074395554184586,
      "learning_rate": 0.00016987364371358481,
      "loss": 2.4023,
      "step": 1609
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.4343386293686442,
      "learning_rate": 0.00016931621312690214,
      "loss": 2.6016,
      "step": 1610
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.4072023288816863,
      "learning_rate": 0.00016875951215197777,
      "loss": 2.3594,
      "step": 1611
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.42534302747820146,
      "learning_rate": 0.00016820354201710214,
      "loss": 2.5332,
      "step": 1612
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.40064058831358007,
      "learning_rate": 0.00016764830394895203,
      "loss": 2.3867,
      "step": 1613
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.40615005661488296,
      "learning_rate": 0.00016709379917259027,
      "loss": 2.4238,
      "step": 1614
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.41587675797985163,
      "learning_rate": 0.00016654002891146091,
      "loss": 2.416,
      "step": 1615
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.3915176005054673,
      "learning_rate": 0.00016598699438738764,
      "loss": 2.4297,
      "step": 1616
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.4269015929412336,
      "learning_rate": 0.00016543469682057105,
      "loss": 2.498,
      "step": 1617
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.4035223101883089,
      "learning_rate": 0.00016488313742958526,
      "loss": 2.4082,
      "step": 1618
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.432302702488857,
      "learning_rate": 0.00016433231743137646,
      "loss": 2.4453,
      "step": 1619
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.40994485512159656,
      "learning_rate": 0.0001637822380412584,
      "loss": 2.4238,
      "step": 1620
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.44632907368271024,
      "learning_rate": 0.00016323290047291195,
      "loss": 2.4434,
      "step": 1621
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.40461201007421005,
      "learning_rate": 0.0001626843059383803,
      "loss": 2.457,
      "step": 1622
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.3906237666289408,
      "learning_rate": 0.00016213645564806752,
      "loss": 2.4512,
      "step": 1623
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.4038844817827262,
      "learning_rate": 0.0001615893508107359,
      "loss": 2.4062,
      "step": 1624
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.3824171132167749,
      "learning_rate": 0.00016104299263350252,
      "loss": 2.3867,
      "step": 1625
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.4148006654215999,
      "learning_rate": 0.00016049738232183758,
      "loss": 2.4492,
      "step": 1626
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.3838928385834824,
      "learning_rate": 0.0001599525210795606,
      "loss": 2.3945,
      "step": 1627
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.4578344843453192,
      "learning_rate": 0.00015940841010883889,
      "loss": 2.4395,
      "step": 1628
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.42275156195660946,
      "learning_rate": 0.00015886505061018413,
      "loss": 2.4863,
      "step": 1629
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.42453262274013515,
      "learning_rate": 0.0001583224437824498,
      "loss": 2.4414,
      "step": 1630
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.37113517997388296,
      "learning_rate": 0.0001577805908228293,
      "loss": 2.4355,
      "step": 1631
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.3948101417734408,
      "learning_rate": 0.00015723949292685191,
      "loss": 2.4434,
      "step": 1632
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.40222791130031577,
      "learning_rate": 0.0001566991512883818,
      "loss": 2.3828,
      "step": 1633
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.4088611703818109,
      "learning_rate": 0.00015615956709961378,
      "loss": 2.4121,
      "step": 1634
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.4155029985667745,
      "learning_rate": 0.00015562074155107215,
      "loss": 2.5195,
      "step": 1635
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.3988856063530602,
      "learning_rate": 0.0001550826758316068,
      "loss": 2.4922,
      "step": 1636
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.40237282285116305,
      "learning_rate": 0.00015454537112839122,
      "loss": 2.4961,
      "step": 1637
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.41687547825483995,
      "learning_rate": 0.00015400882862692033,
      "loss": 2.334,
      "step": 1638
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.4151750203047279,
      "learning_rate": 0.00015347304951100665,
      "loss": 2.3906,
      "step": 1639
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.383722607539598,
      "learning_rate": 0.00015293803496277907,
      "loss": 2.3945,
      "step": 1640
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.3773525471249892,
      "learning_rate": 0.00015240378616267886,
      "loss": 2.4531,
      "step": 1641
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.40877094528810876,
      "learning_rate": 0.00015187030428945843,
      "loss": 2.4395,
      "step": 1642
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.38802208344540307,
      "learning_rate": 0.0001513375905201776,
      "loss": 2.4023,
      "step": 1643
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.3888711474645471,
      "learning_rate": 0.00015080564603020142,
      "loss": 2.5137,
      "step": 1644
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.39254313699708987,
      "learning_rate": 0.0001502744719931982,
      "loss": 2.4629,
      "step": 1645
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.405776105808149,
      "learning_rate": 0.00014974406958113558,
      "loss": 2.4922,
      "step": 1646
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.3831463301767907,
      "learning_rate": 0.00014921443996427947,
      "loss": 2.4883,
      "step": 1647
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.40065728363423886,
      "learning_rate": 0.0001486855843111901,
      "loss": 2.3496,
      "step": 1648
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.39195065370379956,
      "learning_rate": 0.0001481575037887201,
      "loss": 2.4609,
      "step": 1649
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.41142387327726987,
      "learning_rate": 0.00014763019956201253,
      "loss": 2.3906,
      "step": 1650
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.39927294036833977,
      "learning_rate": 0.0001471036727944966,
      "loss": 2.4512,
      "step": 1651
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.4099099199368349,
      "learning_rate": 0.0001465779246478872,
      "loss": 2.3848,
      "step": 1652
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.40787577836401073,
      "learning_rate": 0.00014605295628218045,
      "loss": 2.4766,
      "step": 1653
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.3917904891793823,
      "learning_rate": 0.0001455287688556527,
      "loss": 2.4355,
      "step": 1654
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.40618569396412335,
      "learning_rate": 0.00014500536352485673,
      "loss": 2.4238,
      "step": 1655
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5698852932298657,
      "learning_rate": 0.00014448274144461965,
      "loss": 2.5254,
      "step": 1656
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.41814315739801566,
      "learning_rate": 0.00014396090376804112,
      "loss": 2.4316,
      "step": 1657
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.40425049906269234,
      "learning_rate": 0.00014343985164648926,
      "loss": 2.334,
      "step": 1658
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.4027740792290728,
      "learning_rate": 0.00014291958622959973,
      "loss": 2.4355,
      "step": 1659
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.4132078484911027,
      "learning_rate": 0.00014240010866527176,
      "loss": 2.4258,
      "step": 1660
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.4071309415641029,
      "learning_rate": 0.00014188142009966686,
      "loss": 2.4219,
      "step": 1661
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.3956745794678067,
      "learning_rate": 0.0001413635216772053,
      "loss": 2.3418,
      "step": 1662
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.3901965613640078,
      "learning_rate": 0.000140846414540564,
      "loss": 2.3281,
      "step": 1663
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.4084215485285634,
      "learning_rate": 0.00014033009983067452,
      "loss": 2.4707,
      "step": 1664
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.3932217090129575,
      "learning_rate": 0.00013981457868671927,
      "loss": 2.4648,
      "step": 1665
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.40220746295967197,
      "learning_rate": 0.0001392998522461305,
      "loss": 2.4531,
      "step": 1666
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.40585812810218896,
      "learning_rate": 0.00013878592164458635,
      "loss": 2.4219,
      "step": 1667
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.41264995301361557,
      "learning_rate": 0.00013827278801600978,
      "loss": 2.3926,
      "step": 1668
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.4116552517334253,
      "learning_rate": 0.0001377604524925647,
      "loss": 2.4824,
      "step": 1669
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.4139604625506058,
      "learning_rate": 0.00013724891620465424,
      "loss": 2.3965,
      "step": 1670
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.3917803189876735,
      "learning_rate": 0.0001367381802809185,
      "loss": 2.3164,
      "step": 1671
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.3801060028357546,
      "learning_rate": 0.00013622824584823113,
      "loss": 2.3984,
      "step": 1672
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.3973918040854658,
      "learning_rate": 0.00013571911403169795,
      "loss": 2.4609,
      "step": 1673
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.38907327270668307,
      "learning_rate": 0.0001352107859546533,
      "loss": 2.3516,
      "step": 1674
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.49440347541915497,
      "learning_rate": 0.00013470326273865886,
      "loss": 2.4355,
      "step": 1675
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.42922259255982886,
      "learning_rate": 0.00013419654550349985,
      "loss": 2.3848,
      "step": 1676
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.4086806648818565,
      "learning_rate": 0.00013369063536718346,
      "loss": 2.4023,
      "step": 1677
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.39119519326647567,
      "learning_rate": 0.00013318553344593632,
      "loss": 2.4082,
      "step": 1678
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.41348359244911015,
      "learning_rate": 0.00013268124085420136,
      "loss": 2.4297,
      "step": 1679
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.4313181772062602,
      "learning_rate": 0.0001321777587046364,
      "loss": 2.3848,
      "step": 1680
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.3962525549005578,
      "learning_rate": 0.00013167508810811059,
      "loss": 2.4219,
      "step": 1681
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.4048263976153775,
      "learning_rate": 0.0001311732301737029,
      "loss": 2.3672,
      "step": 1682
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.42156170327304027,
      "learning_rate": 0.0001306721860086991,
      "loss": 2.4023,
      "step": 1683
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.4206948839157917,
      "learning_rate": 0.00013017195671858928,
      "loss": 2.4688,
      "step": 1684
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.4112614039936992,
      "learning_rate": 0.0001296725434070661,
      "loss": 2.4082,
      "step": 1685
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.4017983091898466,
      "learning_rate": 0.00012917394717602121,
      "loss": 2.4453,
      "step": 1686
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.413934163963259,
      "learning_rate": 0.00012867616912554426,
      "loss": 2.5195,
      "step": 1687
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.42382026480492513,
      "learning_rate": 0.00012817921035391882,
      "loss": 2.4102,
      "step": 1688
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.4317301446580033,
      "learning_rate": 0.00012768307195762168,
      "loss": 2.4512,
      "step": 1689
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.39203484763082425,
      "learning_rate": 0.00012718775503131908,
      "loss": 2.3379,
      "step": 1690
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.4066739855107007,
      "learning_rate": 0.0001266932606678646,
      "loss": 2.4355,
      "step": 1691
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.41674341507602103,
      "learning_rate": 0.00012619958995829756,
      "loss": 2.4355,
      "step": 1692
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.4021754547237056,
      "learning_rate": 0.0001257067439918394,
      "loss": 2.3594,
      "step": 1693
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.3917874993416449,
      "learning_rate": 0.00012521472385589234,
      "loss": 2.3965,
      "step": 1694
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.4062725873832379,
      "learning_rate": 0.00012472353063603626,
      "loss": 2.3398,
      "step": 1695
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.40094876602677315,
      "learning_rate": 0.0001242331654160263,
      "loss": 2.5,
      "step": 1696
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.4130105001941233,
      "learning_rate": 0.0001237436292777914,
      "loss": 2.4785,
      "step": 1697
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.39947127703944435,
      "learning_rate": 0.00012325492330143061,
      "loss": 2.4434,
      "step": 1698
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.38477030257465494,
      "learning_rate": 0.00012276704856521175,
      "loss": 2.3984,
      "step": 1699
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.3950136832469225,
      "learning_rate": 0.00012228000614556816,
      "loss": 2.5449,
      "step": 1700
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.4148022367984241,
      "learning_rate": 0.00012179379711709738,
      "loss": 2.4941,
      "step": 1701
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.4071837025752915,
      "learning_rate": 0.0001213084225525577,
      "loss": 2.4277,
      "step": 1702
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.3920089452208134,
      "learning_rate": 0.00012082388352286627,
      "loss": 2.4746,
      "step": 1703
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.4521496854624818,
      "learning_rate": 0.00012034018109709716,
      "loss": 2.4023,
      "step": 1704
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.404327527002345,
      "learning_rate": 0.00011985731634247809,
      "loss": 2.4023,
      "step": 1705
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.4097231184447067,
      "learning_rate": 0.00011937529032438904,
      "loss": 2.5078,
      "step": 1706
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.4139370343708315,
      "learning_rate": 0.00011889410410635887,
      "loss": 2.4258,
      "step": 1707
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.40178478362803144,
      "learning_rate": 0.0001184137587500641,
      "loss": 2.4238,
      "step": 1708
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.38585118205534996,
      "learning_rate": 0.00011793425531532564,
      "loss": 2.3496,
      "step": 1709
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.40805925306888935,
      "learning_rate": 0.00011745559486010671,
      "loss": 2.4395,
      "step": 1710
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.40328503691413636,
      "learning_rate": 0.00011697777844051105,
      "loss": 2.4805,
      "step": 1711
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.4200518174169621,
      "learning_rate": 0.00011650080711077964,
      "loss": 2.5273,
      "step": 1712
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.38217707909344734,
      "learning_rate": 0.00011602468192328936,
      "loss": 2.332,
      "step": 1713
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.403006881350276,
      "learning_rate": 0.00011554940392854973,
      "loss": 2.3867,
      "step": 1714
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.4141328135935977,
      "learning_rate": 0.00011507497417520146,
      "loss": 2.4629,
      "step": 1715
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.4117234148541873,
      "learning_rate": 0.00011460139371001339,
      "loss": 2.3623,
      "step": 1716
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.42944997283046055,
      "learning_rate": 0.00011412866357788049,
      "loss": 2.4102,
      "step": 1717
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.4089728793141541,
      "learning_rate": 0.00011365678482182207,
      "loss": 2.4336,
      "step": 1718
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.41724355141998126,
      "learning_rate": 0.0001131857584829783,
      "loss": 2.4434,
      "step": 1719
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.3931223557025858,
      "learning_rate": 0.0001127155856006093,
      "loss": 2.4375,
      "step": 1720
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.4027181662399743,
      "learning_rate": 0.00011224626721209141,
      "loss": 2.4297,
      "step": 1721
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.4216515210069573,
      "learning_rate": 0.0001117778043529164,
      "loss": 2.4746,
      "step": 1722
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.39400803023765374,
      "learning_rate": 0.0001113101980566879,
      "loss": 2.375,
      "step": 1723
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.40297702397935126,
      "learning_rate": 0.00011084344935511958,
      "loss": 2.5508,
      "step": 1724
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.3809988162805518,
      "learning_rate": 0.00011037755927803345,
      "loss": 2.3008,
      "step": 1725
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.39834135432522727,
      "learning_rate": 0.00010991252885335651,
      "loss": 2.5234,
      "step": 1726
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.41875928039039617,
      "learning_rate": 0.00010944835910711958,
      "loss": 2.4609,
      "step": 1727
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.45552089215675096,
      "learning_rate": 0.00010898505106345396,
      "loss": 2.3027,
      "step": 1728
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.4105229653523523,
      "learning_rate": 0.00010852260574459022,
      "loss": 2.4492,
      "step": 1729
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.39974596065586687,
      "learning_rate": 0.00010806102417085512,
      "loss": 2.3145,
      "step": 1730
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.38795887715101984,
      "learning_rate": 0.00010760030736066951,
      "loss": 2.373,
      "step": 1731
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.4137763383933172,
      "learning_rate": 0.00010714045633054687,
      "loss": 2.3789,
      "step": 1732
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.401626550609146,
      "learning_rate": 0.00010668147209508971,
      "loss": 2.459,
      "step": 1733
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.4232086940415331,
      "learning_rate": 0.00010622335566698877,
      "loss": 2.4414,
      "step": 1734
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.40153220507483833,
      "learning_rate": 0.00010576610805701942,
      "loss": 2.4316,
      "step": 1735
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.4178715903083533,
      "learning_rate": 0.00010530973027404073,
      "loss": 2.3828,
      "step": 1736
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.4089350846273742,
      "learning_rate": 0.00010485422332499212,
      "loss": 2.4355,
      "step": 1737
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.405450632061439,
      "learning_rate": 0.00010439958821489165,
      "loss": 2.3828,
      "step": 1738
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.41452247264054143,
      "learning_rate": 0.00010394582594683428,
      "loss": 2.418,
      "step": 1739
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.4023778058136272,
      "learning_rate": 0.0001034929375219884,
      "loss": 2.3848,
      "step": 1740
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.44014129069742397,
      "learning_rate": 0.00010304092393959514,
      "loss": 2.4316,
      "step": 1741
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.41063080685629266,
      "learning_rate": 0.00010258978619696468,
      "loss": 2.4043,
      "step": 1742
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.388311476080845,
      "learning_rate": 0.00010213952528947551,
      "loss": 2.377,
      "step": 1743
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.4230505545141105,
      "learning_rate": 0.00010169014221057089,
      "loss": 2.4805,
      "step": 1744
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.42248692935184967,
      "learning_rate": 0.00010124163795175734,
      "loss": 2.2197,
      "step": 1745
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.3830789546776496,
      "learning_rate": 0.00010079401350260287,
      "loss": 2.4551,
      "step": 1746
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.41814080023137806,
      "learning_rate": 0.00010034726985073362,
      "loss": 2.4785,
      "step": 1747
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.40141611746547884,
      "learning_rate": 9.9901407981833e-05,
      "loss": 2.4746,
      "step": 1748
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.4162659889326509,
      "learning_rate": 9.94564288796384e-05,
      "loss": 2.3408,
      "step": 1749
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.4084958839467262,
      "learning_rate": 9.901233352593953e-05,
      "loss": 2.459,
      "step": 1750
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.4041304118928708,
      "learning_rate": 9.856912290057668e-05,
      "loss": 2.3652,
      "step": 1751
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.4106196954901805,
      "learning_rate": 9.812679798143748e-05,
      "loss": 2.3379,
      "step": 1752
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.4047339758480756,
      "learning_rate": 9.768535974445586e-05,
      "loss": 2.3672,
      "step": 1753
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.4335289005324956,
      "learning_rate": 9.724480916360906e-05,
      "loss": 2.416,
      "step": 1754
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.3980885033744236,
      "learning_rate": 9.68051472109162e-05,
      "loss": 2.3965,
      "step": 1755
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.41056771961873784,
      "learning_rate": 9.636637485643529e-05,
      "loss": 2.3867,
      "step": 1756
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4345687047789334,
      "learning_rate": 9.592849306826174e-05,
      "loss": 2.4766,
      "step": 1757
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4147308440023137,
      "learning_rate": 9.549150281252633e-05,
      "loss": 2.4922,
      "step": 1758
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4209935877038958,
      "learning_rate": 9.505540505339223e-05,
      "loss": 2.3691,
      "step": 1759
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.41962233092256873,
      "learning_rate": 9.4620200753054e-05,
      "loss": 2.3965,
      "step": 1760
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.42146435255363396,
      "learning_rate": 9.418589087173441e-05,
      "loss": 2.4707,
      "step": 1761
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4063258616060406,
      "learning_rate": 9.375247636768325e-05,
      "loss": 2.4688,
      "step": 1762
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4215702162095022,
      "learning_rate": 9.331995819717443e-05,
      "loss": 2.4785,
      "step": 1763
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4210158722772454,
      "learning_rate": 9.288833731450419e-05,
      "loss": 2.4805,
      "step": 1764
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.3850837780308625,
      "learning_rate": 9.245761467198948e-05,
      "loss": 2.4629,
      "step": 1765
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.41261041440932866,
      "learning_rate": 9.20277912199648e-05,
      "loss": 2.3145,
      "step": 1766
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.39195484694268407,
      "learning_rate": 9.159886790678123e-05,
      "loss": 2.4023,
      "step": 1767
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4020489835687364,
      "learning_rate": 9.11708456788033e-05,
      "loss": 2.4199,
      "step": 1768
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.40355935810366783,
      "learning_rate": 9.074372548040793e-05,
      "loss": 2.5352,
      "step": 1769
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4012851152961733,
      "learning_rate": 9.031750825398145e-05,
      "loss": 2.4473,
      "step": 1770
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4031631798502843,
      "learning_rate": 8.98921949399179e-05,
      "loss": 2.4238,
      "step": 1771
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.3850168249499637,
      "learning_rate": 8.94677864766173e-05,
      "loss": 2.5059,
      "step": 1772
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4163646930021875,
      "learning_rate": 8.904428380048269e-05,
      "loss": 2.3809,
      "step": 1773
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.39881730640299645,
      "learning_rate": 8.862168784591929e-05,
      "loss": 2.3809,
      "step": 1774
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.39112989615690164,
      "learning_rate": 8.819999954533115e-05,
      "loss": 2.4316,
      "step": 1775
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4071491287044505,
      "learning_rate": 8.777921982911996e-05,
      "loss": 2.4531,
      "step": 1776
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.41482568557160576,
      "learning_rate": 8.735934962568253e-05,
      "loss": 2.4785,
      "step": 1777
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.3950495735746128,
      "learning_rate": 8.694038986140945e-05,
      "loss": 2.5215,
      "step": 1778
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.42762114289471964,
      "learning_rate": 8.652234146068206e-05,
      "loss": 2.373,
      "step": 1779
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.4009665885969022,
      "learning_rate": 8.610520534587086e-05,
      "loss": 2.3281,
      "step": 1780
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.395170320198477,
      "learning_rate": 8.568898243733397e-05,
      "loss": 2.3828,
      "step": 1781
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.4142297930303977,
      "learning_rate": 8.527367365341409e-05,
      "loss": 2.4902,
      "step": 1782
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.4229144398994197,
      "learning_rate": 8.485927991043757e-05,
      "loss": 2.4453,
      "step": 1783
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.42860115585593644,
      "learning_rate": 8.444580212271125e-05,
      "loss": 2.502,
      "step": 1784
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.4159250272683492,
      "learning_rate": 8.403324120252159e-05,
      "loss": 2.3926,
      "step": 1785
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.40524576434482984,
      "learning_rate": 8.362159806013175e-05,
      "loss": 2.4219,
      "step": 1786
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.42466982168157785,
      "learning_rate": 8.321087360377988e-05,
      "loss": 2.4727,
      "step": 1787
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.4124665514043204,
      "learning_rate": 8.280106873967752e-05,
      "loss": 2.4941,
      "step": 1788
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.42994559682097805,
      "learning_rate": 8.239218437200679e-05,
      "loss": 2.3184,
      "step": 1789
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.414505820270918,
      "learning_rate": 8.198422140291939e-05,
      "loss": 2.375,
      "step": 1790
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.41279182534886355,
      "learning_rate": 8.157718073253351e-05,
      "loss": 2.418,
      "step": 1791
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.40788080897801177,
      "learning_rate": 8.117106325893287e-05,
      "loss": 2.4062,
      "step": 1792
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.3957513735512191,
      "learning_rate": 8.076586987816404e-05,
      "loss": 2.4199,
      "step": 1793
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.3965612285483393,
      "learning_rate": 8.036160148423449e-05,
      "loss": 2.3633,
      "step": 1794
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.4231465604056877,
      "learning_rate": 7.995825896911141e-05,
      "loss": 2.4023,
      "step": 1795
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.4018279567521213,
      "learning_rate": 7.955584322271853e-05,
      "loss": 2.4961,
      "step": 1796
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.40830148961658796,
      "learning_rate": 7.915435513293523e-05,
      "loss": 2.3164,
      "step": 1797
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.38584682993343844,
      "learning_rate": 7.875379558559387e-05,
      "loss": 2.4258,
      "step": 1798
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.39568199844639895,
      "learning_rate": 7.835416546447838e-05,
      "loss": 2.4199,
      "step": 1799
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4071945374661125,
      "learning_rate": 7.795546565132167e-05,
      "loss": 2.3359,
      "step": 1800
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.39744588308797385,
      "learning_rate": 7.755769702580412e-05,
      "loss": 2.4688,
      "step": 1801
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.42275657747417866,
      "learning_rate": 7.716086046555193e-05,
      "loss": 2.4141,
      "step": 1802
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4150537762417974,
      "learning_rate": 7.676495684613432e-05,
      "loss": 2.4531,
      "step": 1803
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4332053464201804,
      "learning_rate": 7.636998704106252e-05,
      "loss": 2.4473,
      "step": 1804
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4205389501895092,
      "learning_rate": 7.597595192178702e-05,
      "loss": 2.4043,
      "step": 1805
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.40562073924382963,
      "learning_rate": 7.558285235769646e-05,
      "loss": 2.3887,
      "step": 1806
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.3971003392172051,
      "learning_rate": 7.519068921611494e-05,
      "loss": 2.4336,
      "step": 1807
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4010467899402174,
      "learning_rate": 7.479946336230047e-05,
      "loss": 2.3848,
      "step": 1808
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.40123780614971527,
      "learning_rate": 7.440917565944349e-05,
      "loss": 2.2852,
      "step": 1809
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4044050568499585,
      "learning_rate": 7.4019826968664e-05,
      "loss": 2.3359,
      "step": 1810
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.3899473757251684,
      "learning_rate": 7.363141814901053e-05,
      "loss": 2.3535,
      "step": 1811
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4057237375526231,
      "learning_rate": 7.32439500574577e-05,
      "loss": 2.3516,
      "step": 1812
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.398782225940268,
      "learning_rate": 7.285742354890473e-05,
      "loss": 2.3945,
      "step": 1813
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.39755791692411435,
      "learning_rate": 7.247183947617325e-05,
      "loss": 2.457,
      "step": 1814
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.39324623316283736,
      "learning_rate": 7.20871986900053e-05,
      "loss": 2.416,
      "step": 1815
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4045110047062737,
      "learning_rate": 7.170350203906218e-05,
      "loss": 2.4609,
      "step": 1816
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.39448788914785504,
      "learning_rate": 7.132075036992158e-05,
      "loss": 2.3652,
      "step": 1817
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4289802739001266,
      "learning_rate": 7.093894452707666e-05,
      "loss": 2.3789,
      "step": 1818
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.3967232783412032,
      "learning_rate": 7.055808535293334e-05,
      "loss": 2.457,
      "step": 1819
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.387846069427666,
      "learning_rate": 7.017817368780888e-05,
      "loss": 2.418,
      "step": 1820
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4171729908191084,
      "learning_rate": 6.979921036993042e-05,
      "loss": 2.3066,
      "step": 1821
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.3966320504872495,
      "learning_rate": 6.942119623543202e-05,
      "loss": 2.4551,
      "step": 1822
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.41135550019955724,
      "learning_rate": 6.904413211835414e-05,
      "loss": 2.375,
      "step": 1823
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.42070583737892747,
      "learning_rate": 6.866801885064056e-05,
      "loss": 2.4258,
      "step": 1824
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.3930125081459471,
      "learning_rate": 6.829285726213769e-05,
      "loss": 2.4297,
      "step": 1825
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.42477807682372043,
      "learning_rate": 6.79186481805918e-05,
      "loss": 2.3711,
      "step": 1826
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.4268313809320331,
      "learning_rate": 6.754539243164754e-05,
      "loss": 2.2988,
      "step": 1827
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.43702859426393625,
      "learning_rate": 6.717309083884654e-05,
      "loss": 2.498,
      "step": 1828
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.42066246218237013,
      "learning_rate": 6.680174422362468e-05,
      "loss": 2.4629,
      "step": 1829
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.41970876994503564,
      "learning_rate": 6.643135340531136e-05,
      "loss": 2.4297,
      "step": 1830
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.41827900737468343,
      "learning_rate": 6.606191920112664e-05,
      "loss": 2.418,
      "step": 1831
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.40411152836878617,
      "learning_rate": 6.569344242618036e-05,
      "loss": 2.4238,
      "step": 1832
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.41724770237768777,
      "learning_rate": 6.532592389346958e-05,
      "loss": 2.3887,
      "step": 1833
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.42510872824983587,
      "learning_rate": 6.495936441387713e-05,
      "loss": 2.4277,
      "step": 1834
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.4130891922812374,
      "learning_rate": 6.459376479617013e-05,
      "loss": 2.4258,
      "step": 1835
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.4380685016889679,
      "learning_rate": 6.422912584699752e-05,
      "loss": 2.4355,
      "step": 1836
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.3915184020319638,
      "learning_rate": 6.386544837088904e-05,
      "loss": 2.4531,
      "step": 1837
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.41231318498772285,
      "learning_rate": 6.350273317025251e-05,
      "loss": 2.334,
      "step": 1838
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.40551088770859794,
      "learning_rate": 6.314098104537324e-05,
      "loss": 2.3457,
      "step": 1839
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.41294324801132176,
      "learning_rate": 6.278019279441122e-05,
      "loss": 2.4629,
      "step": 1840
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.4189116800280932,
      "learning_rate": 6.242036921339972e-05,
      "loss": 2.377,
      "step": 1841
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.39197241837711333,
      "learning_rate": 6.206151109624402e-05,
      "loss": 2.3809,
      "step": 1842
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.41122997945375034,
      "learning_rate": 6.170361923471868e-05,
      "loss": 2.3945,
      "step": 1843
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.4239791524670697,
      "learning_rate": 6.134669441846691e-05,
      "loss": 2.4141,
      "step": 1844
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.43535742059172416,
      "learning_rate": 6.099073743499772e-05,
      "loss": 2.4316,
      "step": 1845
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.4234623852636729,
      "learning_rate": 6.063574906968511e-05,
      "loss": 2.4805,
      "step": 1846
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.4230945504459543,
      "learning_rate": 6.028173010576582e-05,
      "loss": 2.3594,
      "step": 1847
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.4038358224247526,
      "learning_rate": 5.9928681324337544e-05,
      "loss": 2.2539,
      "step": 1848
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.4065498459582157,
      "learning_rate": 5.957660350435773e-05,
      "loss": 2.4434,
      "step": 1849
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.38691446429256554,
      "learning_rate": 5.922549742264122e-05,
      "loss": 2.2676,
      "step": 1850
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.39767255344196134,
      "learning_rate": 5.8875363853859166e-05,
      "loss": 2.4258,
      "step": 1851
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.40535331478936726,
      "learning_rate": 5.852620357053651e-05,
      "loss": 2.3867,
      "step": 1852
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.423069817683393,
      "learning_rate": 5.8178017343051336e-05,
      "loss": 2.3691,
      "step": 1853
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.4122805223064944,
      "learning_rate": 5.783080593963219e-05,
      "loss": 2.459,
      "step": 1854
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.4329174063783447,
      "learning_rate": 5.748457012635683e-05,
      "loss": 2.3711,
      "step": 1855
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.41219404409282046,
      "learning_rate": 5.713931066715078e-05,
      "loss": 2.373,
      "step": 1856
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.4215858142184679,
      "learning_rate": 5.679502832378497e-05,
      "loss": 2.3867,
      "step": 1857
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.40255427639495855,
      "learning_rate": 5.645172385587482e-05,
      "loss": 2.333,
      "step": 1858
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.43370462905634705,
      "learning_rate": 5.6109398020877834e-05,
      "loss": 2.4609,
      "step": 1859
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.4078301531520449,
      "learning_rate": 5.576805157409265e-05,
      "loss": 2.5,
      "step": 1860
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.4140342306184439,
      "learning_rate": 5.542768526865677e-05,
      "loss": 2.4453,
      "step": 1861
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.4233370709905931,
      "learning_rate": 5.508829985554509e-05,
      "loss": 2.4082,
      "step": 1862
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.44165863865783644,
      "learning_rate": 5.474989608356856e-05,
      "loss": 2.4883,
      "step": 1863
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.4118376043811381,
      "learning_rate": 5.441247469937194e-05,
      "loss": 2.4199,
      "step": 1864
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4143241736434241,
      "learning_rate": 5.407603644743286e-05,
      "loss": 2.418,
      "step": 1865
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.43749966306066734,
      "learning_rate": 5.374058207005944e-05,
      "loss": 2.4023,
      "step": 1866
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4088614334728808,
      "learning_rate": 5.3406112307389066e-05,
      "loss": 2.4004,
      "step": 1867
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4816224368914088,
      "learning_rate": 5.3072627897386926e-05,
      "loss": 2.4941,
      "step": 1868
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4588678446037231,
      "learning_rate": 5.27401295758439e-05,
      "loss": 2.5176,
      "step": 1869
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4076454184865746,
      "learning_rate": 5.2408618076375315e-05,
      "loss": 2.375,
      "step": 1870
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4293078922646919,
      "learning_rate": 5.207809413041914e-05,
      "loss": 2.4844,
      "step": 1871
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.40606989666625865,
      "learning_rate": 5.174855846723459e-05,
      "loss": 2.3184,
      "step": 1872
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4245796933471695,
      "learning_rate": 5.1420011813900104e-05,
      "loss": 2.4121,
      "step": 1873
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.40175172234563844,
      "learning_rate": 5.109245489531211e-05,
      "loss": 2.4336,
      "step": 1874
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4126893505227929,
      "learning_rate": 5.0765888434183446e-05,
      "loss": 2.3418,
      "step": 1875
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.3945466829857134,
      "learning_rate": 5.0440313151041364e-05,
      "loss": 2.4453,
      "step": 1876
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.41051259682630004,
      "learning_rate": 5.011572976422657e-05,
      "loss": 2.3672,
      "step": 1877
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4312866651483132,
      "learning_rate": 4.9792138989890825e-05,
      "loss": 2.5273,
      "step": 1878
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4226690214545823,
      "learning_rate": 4.9469541541996234e-05,
      "loss": 2.3926,
      "step": 1879
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4052604926006738,
      "learning_rate": 4.914793813231305e-05,
      "loss": 2.3652,
      "step": 1880
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.42858090701597623,
      "learning_rate": 4.882732947041818e-05,
      "loss": 2.4355,
      "step": 1881
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4218186928256836,
      "learning_rate": 4.850771626369416e-05,
      "loss": 2.3711,
      "step": 1882
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.42117665560690914,
      "learning_rate": 4.818909921732662e-05,
      "loss": 2.418,
      "step": 1883
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.417029991742264,
      "learning_rate": 4.787147903430383e-05,
      "loss": 2.4199,
      "step": 1884
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4212527343869251,
      "learning_rate": 4.755485641541424e-05,
      "loss": 2.4258,
      "step": 1885
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4148720694226623,
      "learning_rate": 4.723923205924557e-05,
      "loss": 2.4473,
      "step": 1886
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.4031547295316345,
      "learning_rate": 4.6924606662182736e-05,
      "loss": 2.3516,
      "step": 1887
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.43508939488175546,
      "learning_rate": 4.6610980918406596e-05,
      "loss": 2.4473,
      "step": 1888
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.41479035990195107,
      "learning_rate": 4.629835551989276e-05,
      "loss": 2.4746,
      "step": 1889
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.42378701379248107,
      "learning_rate": 4.5986731156409224e-05,
      "loss": 2.3633,
      "step": 1890
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.4237632955957914,
      "learning_rate": 4.567610851551568e-05,
      "loss": 2.3906,
      "step": 1891
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.4031489287526932,
      "learning_rate": 4.536648828256146e-05,
      "loss": 2.3711,
      "step": 1892
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.4029113471089961,
      "learning_rate": 4.505787114068433e-05,
      "loss": 2.3691,
      "step": 1893
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.41223749087709666,
      "learning_rate": 4.4750257770808764e-05,
      "loss": 2.4102,
      "step": 1894
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.4108100273706363,
      "learning_rate": 4.444364885164448e-05,
      "loss": 2.4551,
      "step": 1895
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.4069333038971008,
      "learning_rate": 4.413804505968533e-05,
      "loss": 2.4648,
      "step": 1896
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.38431311100697507,
      "learning_rate": 4.3833447069206944e-05,
      "loss": 2.3945,
      "step": 1897
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.4092275819723983,
      "learning_rate": 4.352985555226635e-05,
      "loss": 2.3262,
      "step": 1898
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.42734474453285143,
      "learning_rate": 4.322727117869951e-05,
      "loss": 2.4922,
      "step": 1899
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.42099940008450376,
      "learning_rate": 4.29256946161205e-05,
      "loss": 2.3984,
      "step": 1900
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.4065053018979364,
      "learning_rate": 4.262512652991968e-05,
      "loss": 2.4199,
      "step": 1901
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.42329877319503406,
      "learning_rate": 4.2325567583262113e-05,
      "loss": 2.3867,
      "step": 1902
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.4385297675021045,
      "learning_rate": 4.2027018437086895e-05,
      "loss": 2.3809,
      "step": 1903
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.4206403097881495,
      "learning_rate": 4.172947975010449e-05,
      "loss": 2.3535,
      "step": 1904
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.407292299616498,
      "learning_rate": 4.143295217879645e-05,
      "loss": 2.3926,
      "step": 1905
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.41792524936673003,
      "learning_rate": 4.113743637741296e-05,
      "loss": 2.2969,
      "step": 1906
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.42853167897127553,
      "learning_rate": 4.084293299797226e-05,
      "loss": 2.3574,
      "step": 1907
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.40897667300826945,
      "learning_rate": 4.054944269025862e-05,
      "loss": 2.3281,
      "step": 1908
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.42322454249049574,
      "learning_rate": 4.025696610182095e-05,
      "loss": 2.4551,
      "step": 1909
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.41501674269518024,
      "learning_rate": 3.996550387797187e-05,
      "loss": 2.3398,
      "step": 1910
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.4208965552474903,
      "learning_rate": 3.9675056661785556e-05,
      "loss": 2.5293,
      "step": 1911
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.4470844714021226,
      "learning_rate": 3.9385625094097154e-05,
      "loss": 2.3672,
      "step": 1912
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.4135774290141522,
      "learning_rate": 3.909720981350034e-05,
      "loss": 2.4551,
      "step": 1913
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.417303417611598,
      "learning_rate": 3.880981145634704e-05,
      "loss": 2.4004,
      "step": 1914
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.44764392002689996,
      "learning_rate": 3.852343065674507e-05,
      "loss": 2.541,
      "step": 1915
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.411854877381274,
      "learning_rate": 3.8238068046557276e-05,
      "loss": 2.457,
      "step": 1916
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.40333373024929847,
      "learning_rate": 3.795372425540006e-05,
      "loss": 2.4336,
      "step": 1917
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.40784403894075927,
      "learning_rate": 3.76703999106418e-05,
      "loss": 2.4141,
      "step": 1918
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.4233378331368735,
      "learning_rate": 3.7388095637401754e-05,
      "loss": 2.457,
      "step": 1919
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.41736137806636847,
      "learning_rate": 3.7106812058548376e-05,
      "loss": 2.5234,
      "step": 1920
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.40025522934655816,
      "learning_rate": 3.682654979469807e-05,
      "loss": 2.4062,
      "step": 1921
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.4330172170644544,
      "learning_rate": 3.654730946421403e-05,
      "loss": 2.5156,
      "step": 1922
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.40990770151859446,
      "learning_rate": 3.6269091683204466e-05,
      "loss": 2.4375,
      "step": 1923
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.4056257859862419,
      "learning_rate": 3.5991897065521693e-05,
      "loss": 2.5215,
      "step": 1924
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.40988197693030226,
      "learning_rate": 3.571572622276026e-05,
      "loss": 2.4141,
      "step": 1925
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.408914496463609,
      "learning_rate": 3.544057976425619e-05,
      "loss": 2.3535,
      "step": 1926
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.41465792939313684,
      "learning_rate": 3.5166458297085146e-05,
      "loss": 2.3691,
      "step": 1927
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.4025541931077644,
      "learning_rate": 3.489336242606111e-05,
      "loss": 2.4473,
      "step": 1928
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.4371563691355653,
      "learning_rate": 3.462129275373577e-05,
      "loss": 2.4141,
      "step": 1929
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.4082904773829414,
      "learning_rate": 3.4350249880395924e-05,
      "loss": 2.4961,
      "step": 1930
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.42395963769519684,
      "learning_rate": 3.408023440406355e-05,
      "loss": 2.4551,
      "step": 1931
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.4037207717111813,
      "learning_rate": 3.381124692049331e-05,
      "loss": 2.5059,
      "step": 1932
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.4088000066519503,
      "learning_rate": 3.354328802317197e-05,
      "loss": 2.4277,
      "step": 1933
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.39747528101840485,
      "learning_rate": 3.327635830331677e-05,
      "loss": 2.3281,
      "step": 1934
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.4355489884873717,
      "learning_rate": 3.3010458349874206e-05,
      "loss": 2.4785,
      "step": 1935
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.4267232288793524,
      "learning_rate": 3.2745588749518775e-05,
      "loss": 2.3926,
      "step": 1936
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.40158736554577934,
      "learning_rate": 3.248175008665161e-05,
      "loss": 2.4375,
      "step": 1937
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.40647876978214337,
      "learning_rate": 3.221894294339911e-05,
      "loss": 2.3867,
      "step": 1938
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.4253485444997097,
      "learning_rate": 3.1957167899611836e-05,
      "loss": 2.4707,
      "step": 1939
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.41770507503742565,
      "learning_rate": 3.169642553286334e-05,
      "loss": 2.4023,
      "step": 1940
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.40599332586235354,
      "learning_rate": 3.143671641844831e-05,
      "loss": 2.3125,
      "step": 1941
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.39685243417220895,
      "learning_rate": 3.117804112938205e-05,
      "loss": 2.3125,
      "step": 1942
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.40354387367298783,
      "learning_rate": 3.092040023639869e-05,
      "loss": 2.4082,
      "step": 1943
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.4074800194137537,
      "learning_rate": 3.066379430795002e-05,
      "loss": 2.3867,
      "step": 1944
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.3876615114395053,
      "learning_rate": 3.040822391020459e-05,
      "loss": 2.3359,
      "step": 1945
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.4257901975748049,
      "learning_rate": 3.0153689607045842e-05,
      "loss": 2.4004,
      "step": 1946
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.4118213440447194,
      "learning_rate": 2.9900191960071545e-05,
      "loss": 2.4531,
      "step": 1947
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.42725136304058303,
      "learning_rate": 2.9647731528591848e-05,
      "loss": 2.3906,
      "step": 1948
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.42291597720632046,
      "learning_rate": 2.9396308869628795e-05,
      "loss": 2.3789,
      "step": 1949
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.4184326938336618,
      "learning_rate": 2.914592453791448e-05,
      "loss": 2.4258,
      "step": 1950
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.42899029966555435,
      "learning_rate": 2.8896579085889994e-05,
      "loss": 2.4766,
      "step": 1951
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.4221240312801532,
      "learning_rate": 2.86482730637046e-05,
      "loss": 2.4805,
      "step": 1952
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.4131138263891258,
      "learning_rate": 2.840100701921383e-05,
      "loss": 2.3711,
      "step": 1953
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.39990874286510913,
      "learning_rate": 2.8154781497978898e-05,
      "loss": 2.4688,
      "step": 1954
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.3937488776319999,
      "learning_rate": 2.7909597043265013e-05,
      "loss": 2.4453,
      "step": 1955
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.405909802671608,
      "learning_rate": 2.7665454196040662e-05,
      "loss": 2.2451,
      "step": 1956
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.413947398749442,
      "learning_rate": 2.7422353494975905e-05,
      "loss": 2.3613,
      "step": 1957
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.40850209194807185,
      "learning_rate": 2.7180295476441573e-05,
      "loss": 2.3594,
      "step": 1958
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.3960822482203752,
      "learning_rate": 2.6939280674508016e-05,
      "loss": 2.4629,
      "step": 1959
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.3970709287728483,
      "learning_rate": 2.669930962094358e-05,
      "loss": 2.377,
      "step": 1960
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.40333024903473425,
      "learning_rate": 2.6460382845214126e-05,
      "loss": 2.4941,
      "step": 1961
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.4199548733991896,
      "learning_rate": 2.6222500874481025e-05,
      "loss": 2.3945,
      "step": 1962
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.4095424065675553,
      "learning_rate": 2.5985664233600827e-05,
      "loss": 2.4297,
      "step": 1963
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.41824323423841964,
      "learning_rate": 2.574987344512336e-05,
      "loss": 2.4102,
      "step": 1964
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.4229319070214064,
      "learning_rate": 2.5515129029290984e-05,
      "loss": 2.334,
      "step": 1965
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.4079168527733058,
      "learning_rate": 2.5281431504037556e-05,
      "loss": 2.4512,
      "step": 1966
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.40733920263052503,
      "learning_rate": 2.504878138498684e-05,
      "loss": 2.3203,
      "step": 1967
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.41929711208815235,
      "learning_rate": 2.48171791854519e-05,
      "loss": 2.5391,
      "step": 1968
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.43122619183315336,
      "learning_rate": 2.4586625416433473e-05,
      "loss": 2.4395,
      "step": 1969
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.40998775187246145,
      "learning_rate": 2.435712058661921e-05,
      "loss": 2.4121,
      "step": 1970
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.4114098978575851,
      "learning_rate": 2.4128665202382327e-05,
      "loss": 2.4043,
      "step": 1971
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.4329047157114394,
      "learning_rate": 2.3901259767780515e-05,
      "loss": 2.4355,
      "step": 1972
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.41577274576556905,
      "learning_rate": 2.367490478455514e-05,
      "loss": 2.4082,
      "step": 1973
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4153537166167028,
      "learning_rate": 2.3449600752129597e-05,
      "loss": 2.3809,
      "step": 1974
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.41426979639691586,
      "learning_rate": 2.3225348167608685e-05,
      "loss": 2.377,
      "step": 1975
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4017847613538476,
      "learning_rate": 2.3002147525777118e-05,
      "loss": 2.3203,
      "step": 1976
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.43182004333916935,
      "learning_rate": 2.2779999319098856e-05,
      "loss": 2.4688,
      "step": 1977
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4376943388260587,
      "learning_rate": 2.255890403771571e-05,
      "loss": 2.4023,
      "step": 1978
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4929329269114405,
      "learning_rate": 2.233886216944614e-05,
      "loss": 2.3418,
      "step": 1979
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4126977066336403,
      "learning_rate": 2.211987419978484e-05,
      "loss": 2.4082,
      "step": 1980
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.42177239302732455,
      "learning_rate": 2.1901940611900705e-05,
      "loss": 2.5098,
      "step": 1981
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4212048902576538,
      "learning_rate": 2.168506188663666e-05,
      "loss": 2.3887,
      "step": 1982
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4467821808198787,
      "learning_rate": 2.1469238502507925e-05,
      "loss": 2.457,
      "step": 1983
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4102314697699932,
      "learning_rate": 2.125447093570154e-05,
      "loss": 2.4043,
      "step": 1984
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4130648944812326,
      "learning_rate": 2.1040759660074793e-05,
      "loss": 2.3164,
      "step": 1985
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.43054006097239855,
      "learning_rate": 2.0828105147154273e-05,
      "loss": 2.3809,
      "step": 1986
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.41703549291138237,
      "learning_rate": 2.061650786613545e-05,
      "loss": 2.4355,
      "step": 1987
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.41610169856482654,
      "learning_rate": 2.040596828388058e-05,
      "loss": 2.4297,
      "step": 1988
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4228132422727743,
      "learning_rate": 2.019648686491865e-05,
      "loss": 2.3438,
      "step": 1989
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4099753842470826,
      "learning_rate": 1.9988064071443767e-05,
      "loss": 2.3633,
      "step": 1990
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4293181046000917,
      "learning_rate": 1.9780700363314253e-05,
      "loss": 2.4609,
      "step": 1991
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4029853568021587,
      "learning_rate": 1.957439619805196e-05,
      "loss": 2.3281,
      "step": 1992
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.41070245706756253,
      "learning_rate": 1.9369152030840554e-05,
      "loss": 2.4199,
      "step": 1993
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4076895838753328,
      "learning_rate": 1.916496831452552e-05,
      "loss": 2.4746,
      "step": 1994
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.43336683830075856,
      "learning_rate": 1.8961845499611998e-05,
      "loss": 2.4297,
      "step": 1995
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.44281428635443615,
      "learning_rate": 1.8759784034264925e-05,
      "loss": 2.3223,
      "step": 1996
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.4107452148237735,
      "learning_rate": 1.855878436430708e-05,
      "loss": 2.3945,
      "step": 1997
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.4299087057976937,
      "learning_rate": 1.835884693321871e-05,
      "loss": 2.4082,
      "step": 1998
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.38820000702268825,
      "learning_rate": 1.8159972182136386e-05,
      "loss": 2.373,
      "step": 1999
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.4268222824655624,
      "learning_rate": 1.7962160549851945e-05,
      "loss": 2.3848,
      "step": 2000
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.43431955105182996,
      "learning_rate": 1.776541247281177e-05,
      "loss": 2.3164,
      "step": 2001
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.39750162686386264,
      "learning_rate": 1.7569728385115224e-05,
      "loss": 2.4102,
      "step": 2002
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.44495559598924683,
      "learning_rate": 1.7375108718514665e-05,
      "loss": 2.4492,
      "step": 2003
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.4070459335167156,
      "learning_rate": 1.7181553902413438e-05,
      "loss": 2.3867,
      "step": 2004
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.4020904365968681,
      "learning_rate": 1.698906436386577e-05,
      "loss": 2.332,
      "step": 2005
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.4193148853689179,
      "learning_rate": 1.679764052757532e-05,
      "loss": 2.3906,
      "step": 2006
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.4229748094220331,
      "learning_rate": 1.6607282815894464e-05,
      "loss": 2.5508,
      "step": 2007
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.4124299057527789,
      "learning_rate": 1.6417991648823405e-05,
      "loss": 2.4707,
      "step": 2008
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.41971679790580435,
      "learning_rate": 1.6229767444008835e-05,
      "loss": 2.4219,
      "step": 2009
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.4022089812194615,
      "learning_rate": 1.604261061674378e-05,
      "loss": 2.4258,
      "step": 2010
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.431684980174039,
      "learning_rate": 1.5856521579965865e-05,
      "loss": 2.4473,
      "step": 2011
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.42371566207113315,
      "learning_rate": 1.5671500744256938e-05,
      "loss": 2.4082,
      "step": 2012
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.3982988063121166,
      "learning_rate": 1.5487548517841953e-05,
      "loss": 2.4043,
      "step": 2013
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.4243064664931957,
      "learning_rate": 1.530466530658814e-05,
      "loss": 2.4551,
      "step": 2014
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.4271580801635476,
      "learning_rate": 1.5122851514004054e-05,
      "loss": 2.3867,
      "step": 2015
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.4112215626608728,
      "learning_rate": 1.4942107541238703e-05,
      "loss": 2.3789,
      "step": 2016
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.41515720291823804,
      "learning_rate": 1.4762433787080809e-05,
      "loss": 2.4492,
      "step": 2017
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.42689880704671346,
      "learning_rate": 1.4583830647957541e-05,
      "loss": 2.4492,
      "step": 2018
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.4093117111871889,
      "learning_rate": 1.4406298517934068e-05,
      "loss": 2.3809,
      "step": 2019
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.4115977406523451,
      "learning_rate": 1.4229837788712562e-05,
      "loss": 2.4805,
      "step": 2020
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.40089834074091046,
      "learning_rate": 1.4054448849631085e-05,
      "loss": 2.2871,
      "step": 2021
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.41638611032522566,
      "learning_rate": 1.3880132087663145e-05,
      "loss": 2.4238,
      "step": 2022
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.42179809389213824,
      "learning_rate": 1.3706887887416419e-05,
      "loss": 2.4785,
      "step": 2023
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.4173267208271537,
      "learning_rate": 1.3534716631132316e-05,
      "loss": 2.3398,
      "step": 2024
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.422554163025764,
      "learning_rate": 1.3363618698684853e-05,
      "loss": 2.3594,
      "step": 2025
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.39157178193579617,
      "learning_rate": 1.3193594467579728e-05,
      "loss": 2.4238,
      "step": 2026
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.41796285244087633,
      "learning_rate": 1.3024644312954026e-05,
      "loss": 2.4043,
      "step": 2027
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.4086643069935261,
      "learning_rate": 1.2856768607574564e-05,
      "loss": 2.3789,
      "step": 2028
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.41210815960017866,
      "learning_rate": 1.2689967721837947e-05,
      "loss": 2.4062,
      "step": 2029
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.4176290104913584,
      "learning_rate": 1.2524242023769006e-05,
      "loss": 2.3496,
      "step": 2030
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.40117569761134647,
      "learning_rate": 1.2359591879020526e-05,
      "loss": 2.4062,
      "step": 2031
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.40523539419890164,
      "learning_rate": 1.2196017650872081e-05,
      "loss": 2.4492,
      "step": 2032
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.42544520563962535,
      "learning_rate": 1.2033519700229367e-05,
      "loss": 2.3555,
      "step": 2033
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.41219164972813893,
      "learning_rate": 1.1872098385623586e-05,
      "loss": 2.3887,
      "step": 2034
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.41265922113877423,
      "learning_rate": 1.1711754063210289e-05,
      "loss": 2.4492,
      "step": 2035
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.3990131935575849,
      "learning_rate": 1.155248708676887e-05,
      "loss": 2.4414,
      "step": 2036
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.4305948099856799,
      "learning_rate": 1.1394297807701737e-05,
      "loss": 2.3105,
      "step": 2037
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.4212458908820652,
      "learning_rate": 1.1237186575033254e-05,
      "loss": 2.4004,
      "step": 2038
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.4278262601835207,
      "learning_rate": 1.1081153735409522e-05,
      "loss": 2.3525,
      "step": 2039
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.398662905200273,
      "learning_rate": 1.0926199633097156e-05,
      "loss": 2.3438,
      "step": 2040
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.4016465057167636,
      "learning_rate": 1.0772324609982787e-05,
      "loss": 2.3379,
      "step": 2041
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.4025150911927196,
      "learning_rate": 1.0619529005571893e-05,
      "loss": 2.3965,
      "step": 2042
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.42278403488786936,
      "learning_rate": 1.0467813156988748e-05,
      "loss": 2.3867,
      "step": 2043
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.4223319569173489,
      "learning_rate": 1.0317177398975031e-05,
      "loss": 2.5059,
      "step": 2044
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.42683037450874,
      "learning_rate": 1.0167622063889326e-05,
      "loss": 2.2949,
      "step": 2045
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.40539007871013993,
      "learning_rate": 1.0019147481706625e-05,
      "loss": 2.3887,
      "step": 2046
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.41846265414888023,
      "learning_rate": 9.871753980017051e-06,
      "loss": 2.4023,
      "step": 2047
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.4124482593278867,
      "learning_rate": 9.725441884025855e-06,
      "loss": 2.3633,
      "step": 2048
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.4084874771501175,
      "learning_rate": 9.580211516551862e-06,
      "loss": 2.4531,
      "step": 2049
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.45847827993060547,
      "learning_rate": 9.436063198027589e-06,
      "loss": 2.418,
      "step": 2050
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.4215154622442065,
      "learning_rate": 9.292997246497959e-06,
      "loss": 2.5137,
      "step": 2051
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.43654870207105134,
      "learning_rate": 9.151013977619693e-06,
      "loss": 2.4883,
      "step": 2052
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.40703487036830677,
      "learning_rate": 9.010113704661038e-06,
      "loss": 2.3125,
      "step": 2053
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.42474109273666344,
      "learning_rate": 8.870296738500316e-06,
      "loss": 2.4863,
      "step": 2054
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.41970618573458696,
      "learning_rate": 8.731563387626095e-06,
      "loss": 2.4395,
      "step": 2055
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.416023569411775,
      "learning_rate": 8.59391395813569e-06,
      "loss": 2.4316,
      "step": 2056
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.40994930481979097,
      "learning_rate": 8.457348753735328e-06,
      "loss": 2.416,
      "step": 2057
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.40888059645443164,
      "learning_rate": 8.321868075738593e-06,
      "loss": 2.3574,
      "step": 2058
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.4112345587717931,
      "learning_rate": 8.187472223066371e-06,
      "loss": 2.4336,
      "step": 2059
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.40663500148045806,
      "learning_rate": 8.054161492246136e-06,
      "loss": 2.3027,
      "step": 2060
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.41523951472212794,
      "learning_rate": 7.921936177411049e-06,
      "loss": 2.3516,
      "step": 2061
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4083210155888962,
      "learning_rate": 7.790796570299463e-06,
      "loss": 2.3828,
      "step": 2062
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.41183863196644405,
      "learning_rate": 7.660742960254207e-06,
      "loss": 2.3848,
      "step": 2063
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4808630357668039,
      "learning_rate": 7.531775634222138e-06,
      "loss": 2.3457,
      "step": 2064
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4127565629514213,
      "learning_rate": 7.403894876753192e-06,
      "loss": 2.3965,
      "step": 2065
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4114326777481178,
      "learning_rate": 7.277100970000061e-06,
      "loss": 2.4492,
      "step": 2066
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.41303787363281325,
      "learning_rate": 7.151394193717408e-06,
      "loss": 2.4141,
      "step": 2067
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.41702043586445764,
      "learning_rate": 7.026774825261151e-06,
      "loss": 2.5039,
      "step": 2068
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4060783808907117,
      "learning_rate": 6.903243139588233e-06,
      "loss": 2.5039,
      "step": 2069
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4609396447519739,
      "learning_rate": 6.780799409255522e-06,
      "loss": 2.4355,
      "step": 2070
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4335532623259515,
      "learning_rate": 6.659443904419637e-06,
      "loss": 2.4453,
      "step": 2071
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4110003769028962,
      "learning_rate": 6.539176892836008e-06,
      "loss": 2.4766,
      "step": 2072
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4109443974828272,
      "learning_rate": 6.4199986398585375e-06,
      "loss": 2.3477,
      "step": 2073
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.3941090510220777,
      "learning_rate": 6.3019094084388884e-06,
      "loss": 2.3984,
      "step": 2074
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4154953689353513,
      "learning_rate": 6.18490945912592e-06,
      "loss": 2.4023,
      "step": 2075
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4136621801867022,
      "learning_rate": 6.068999050065249e-06,
      "loss": 2.3945,
      "step": 2076
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4275571960212731,
      "learning_rate": 5.9541784369983586e-06,
      "loss": 2.3008,
      "step": 2077
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4134184691568077,
      "learning_rate": 5.840447873262433e-06,
      "loss": 2.3027,
      "step": 2078
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4321794447333286,
      "learning_rate": 5.727807609789471e-06,
      "loss": 2.3613,
      "step": 2079
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4334809338015014,
      "learning_rate": 5.616257895105892e-06,
      "loss": 2.4414,
      "step": 2080
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4195661619097815,
      "learning_rate": 5.505798975331933e-06,
      "loss": 2.4355,
      "step": 2081
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.40572264584476714,
      "learning_rate": 5.396431094181198e-06,
      "loss": 2.4336,
      "step": 2082
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.42319037875814536,
      "learning_rate": 5.288154492960107e-06,
      "loss": 2.4785,
      "step": 2083
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.40165495179027727,
      "learning_rate": 5.1809694105671155e-06,
      "loss": 2.4004,
      "step": 2084
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.4035649500824764,
      "learning_rate": 5.074876083492441e-06,
      "loss": 2.3477,
      "step": 2085
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.4012445691217823,
      "learning_rate": 4.96987474581767e-06,
      "loss": 2.3242,
      "step": 2086
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.4180840917728042,
      "learning_rate": 4.865965629214819e-06,
      "loss": 2.457,
      "step": 2087
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.43035928984241595,
      "learning_rate": 4.763148962946218e-06,
      "loss": 2.293,
      "step": 2088
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.4139290461160366,
      "learning_rate": 4.661424973863681e-06,
      "loss": 2.4102,
      "step": 2089
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.43319850476620037,
      "learning_rate": 4.560793886408398e-06,
      "loss": 2.4453,
      "step": 2090
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.4197255110716398,
      "learning_rate": 4.461255922609986e-06,
      "loss": 2.4082,
      "step": 2091
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.4301362244088255,
      "learning_rate": 4.362811302086267e-06,
      "loss": 2.4277,
      "step": 2092
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.40852799945884016,
      "learning_rate": 4.265460242042885e-06,
      "loss": 2.3066,
      "step": 2093
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.44967461408583453,
      "learning_rate": 4.169202957272522e-06,
      "loss": 2.4219,
      "step": 2094
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.41963629298847377,
      "learning_rate": 4.074039660154738e-06,
      "loss": 2.4648,
      "step": 2095
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.44400177130166785,
      "learning_rate": 3.9799705606551325e-06,
      "loss": 2.4023,
      "step": 2096
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.4208980134217279,
      "learning_rate": 3.886995866325294e-06,
      "loss": 2.4492,
      "step": 2097
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.41666636360642967,
      "learning_rate": 3.795115782302072e-06,
      "loss": 2.3457,
      "step": 2098
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.40592048593854874,
      "learning_rate": 3.704330511307197e-06,
      "loss": 2.3906,
      "step": 2099
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.4221554830702615,
      "learning_rate": 3.614640253646828e-06,
      "loss": 2.3457,
      "step": 2100
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.41433474320922153,
      "learning_rate": 3.5260452072110594e-06,
      "loss": 2.3965,
      "step": 2101
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.4300391452308492,
      "learning_rate": 3.4385455674737498e-06,
      "loss": 2.3848,
      "step": 2102
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.4265726176483776,
      "learning_rate": 3.3521415274915256e-06,
      "loss": 2.4316,
      "step": 2103
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.44280965464578154,
      "learning_rate": 3.2668332779041133e-06,
      "loss": 2.4355,
      "step": 2104
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.4004595319255937,
      "learning_rate": 3.1826210069332838e-06,
      "loss": 2.4492,
      "step": 2105
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.43762600782583566,
      "learning_rate": 3.0995049003826324e-06,
      "loss": 2.4609,
      "step": 2106
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.4181475642603128,
      "learning_rate": 3.017485141637355e-06,
      "loss": 2.3652,
      "step": 2107
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.4220239661332557,
      "learning_rate": 2.9365619116636376e-06,
      "loss": 2.3848,
      "step": 2108
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.4238773535861798,
      "learning_rate": 2.856735389008269e-06,
      "loss": 2.4551,
      "step": 2109
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.419529515035068,
      "learning_rate": 2.778005749798307e-06,
      "loss": 2.3828,
      "step": 2110
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.4229216611812852,
      "learning_rate": 2.700373167740744e-06,
      "loss": 2.375,
      "step": 2111
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.409558003171384,
      "learning_rate": 2.62383781412201e-06,
      "loss": 2.2969,
      "step": 2112
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.4195615311163283,
      "learning_rate": 2.5483998578076373e-06,
      "loss": 2.4609,
      "step": 2113
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.4043242283954338,
      "learning_rate": 2.4740594652418736e-06,
      "loss": 2.4434,
      "step": 2114
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.40496715223538415,
      "learning_rate": 2.4008168004472917e-06,
      "loss": 2.4062,
      "step": 2115
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.444204124874282,
      "learning_rate": 2.3286720250246253e-06,
      "loss": 2.4629,
      "step": 2116
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.4174289515425753,
      "learning_rate": 2.2576252981520994e-06,
      "loss": 2.4121,
      "step": 2117
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.4066639093201403,
      "learning_rate": 2.1876767765853233e-06,
      "loss": 2.3574,
      "step": 2118
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.4097774548852568,
      "learning_rate": 2.118826614656788e-06,
      "loss": 2.3535,
      "step": 2119
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.41092053971633585,
      "learning_rate": 2.051074964275701e-06,
      "loss": 2.377,
      "step": 2120
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.41432000189531315,
      "learning_rate": 1.984421974927375e-06,
      "loss": 2.3809,
      "step": 2121
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.39712383608889795,
      "learning_rate": 1.9188677936731734e-06,
      "loss": 2.3574,
      "step": 2122
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.39061435342227435,
      "learning_rate": 1.8544125651501208e-06,
      "loss": 2.418,
      "step": 2123
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.4020487470745629,
      "learning_rate": 1.7910564315704035e-06,
      "loss": 2.3047,
      "step": 2124
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.4312387677633388,
      "learning_rate": 1.7287995327214257e-06,
      "loss": 2.502,
      "step": 2125
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.41986704235483596,
      "learning_rate": 1.6676420059649754e-06,
      "loss": 2.4609,
      "step": 2126
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.4001732901115435,
      "learning_rate": 1.6075839862374485e-06,
      "loss": 2.3457,
      "step": 2127
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.43274105149665415,
      "learning_rate": 1.5486256060492366e-06,
      "loss": 2.541,
      "step": 2128
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.4019511129732767,
      "learning_rate": 1.4907669954844495e-06,
      "loss": 2.3984,
      "step": 2129
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.4110913882997396,
      "learning_rate": 1.434008282200805e-06,
      "loss": 2.4492,
      "step": 2130
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.41845815372149575,
      "learning_rate": 1.3783495914291844e-06,
      "loss": 2.3711,
      "step": 2131
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.4352690268606059,
      "learning_rate": 1.3237910459734104e-06,
      "loss": 2.4375,
      "step": 2132
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.404120207753855,
      "learning_rate": 1.270332766210025e-06,
      "loss": 2.4023,
      "step": 2133
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.40789293515566877,
      "learning_rate": 1.2179748700879012e-06,
      "loss": 2.4629,
      "step": 2134
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.41723852899193026,
      "learning_rate": 1.1667174731280205e-06,
      "loss": 2.3809,
      "step": 2135
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.4165884102159427,
      "learning_rate": 1.1165606884234182e-06,
      "loss": 2.3359,
      "step": 2136
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.4329850291180293,
      "learning_rate": 1.0675046266386268e-06,
      "loss": 2.4297,
      "step": 2137
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.4233413744591337,
      "learning_rate": 1.019549396009567e-06,
      "loss": 2.4434,
      "step": 2138
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.4084877671285245,
      "learning_rate": 9.726951023434348e-07,
      "loss": 2.4062,
      "step": 2139
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.40810612060378887,
      "learning_rate": 9.269418490182591e-07,
      "loss": 2.4805,
      "step": 2140
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.400038523289864,
      "learning_rate": 8.822897369827332e-07,
      "loss": 2.4141,
      "step": 2141
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.41489813024683253,
      "learning_rate": 8.387388647561611e-07,
      "loss": 2.4316,
      "step": 2142
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.42878717768849917,
      "learning_rate": 7.962893284279016e-07,
      "loss": 2.4961,
      "step": 2143
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.4146630809880468,
      "learning_rate": 7.549412216574791e-07,
      "loss": 2.4395,
      "step": 2144
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.42280925836176414,
      "learning_rate": 7.146946356743067e-07,
      "loss": 2.3711,
      "step": 2145
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.4311714876888659,
      "learning_rate": 6.755496592773524e-07,
      "loss": 2.3477,
      "step": 2146
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.4200294826659529,
      "learning_rate": 6.375063788349733e-07,
      "loss": 2.3594,
      "step": 2147
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.4046431089881595,
      "learning_rate": 6.005648782848594e-07,
      "loss": 2.291,
      "step": 2148
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.41338439709535574,
      "learning_rate": 5.647252391337565e-07,
      "loss": 2.3945,
      "step": 2149
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.4150325276056694,
      "learning_rate": 5.299875404572441e-07,
      "loss": 2.3594,
      "step": 2150
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.42925674666750063,
      "learning_rate": 4.963518588996796e-07,
      "loss": 2.5,
      "step": 2151
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.42983228387272016,
      "learning_rate": 4.638182686738657e-07,
      "loss": 2.4688,
      "step": 2152
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.4225666613510629,
      "learning_rate": 4.3238684156110543e-07,
      "loss": 2.4043,
      "step": 2153
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.4032787958363503,
      "learning_rate": 4.020576469108139e-07,
      "loss": 2.3828,
      "step": 2154
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.39679437120082917,
      "learning_rate": 3.7283075164046274e-07,
      "loss": 2.3906,
      "step": 2155
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.4240324010433113,
      "learning_rate": 3.4470622023557995e-07,
      "loss": 2.3789,
      "step": 2156
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.413434199289064,
      "learning_rate": 3.176841147492504e-07,
      "loss": 2.2969,
      "step": 2157
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.404490417182054,
      "learning_rate": 2.9176449480244895e-07,
      "loss": 2.4004,
      "step": 2158
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.3942502976733122,
      "learning_rate": 2.6694741758342967e-07,
      "loss": 2.3809,
      "step": 2159
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.42723991382761517,
      "learning_rate": 2.432329378478926e-07,
      "loss": 2.3984,
      "step": 2160
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.4325080470468073,
      "learning_rate": 2.2062110791892798e-07,
      "loss": 2.3711,
      "step": 2161
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.42550081704527526,
      "learning_rate": 1.9911197768662792e-07,
      "loss": 2.498,
      "step": 2162
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.413198983658783,
      "learning_rate": 1.7870559460814173e-07,
      "loss": 2.4824,
      "step": 2163
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.3904636009760579,
      "learning_rate": 1.5940200370750947e-07,
      "loss": 2.248,
      "step": 2164
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.41890917781405085,
      "learning_rate": 1.4120124757577291e-07,
      "loss": 2.4023,
      "step": 2165
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.40736973498942664,
      "learning_rate": 1.2410336637047603e-07,
      "loss": 2.4922,
      "step": 2166
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.4073843606310897,
      "learning_rate": 1.081083978159425e-07,
      "loss": 2.3691,
      "step": 2167
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.421989946523163,
      "learning_rate": 9.321637720310915e-08,
      "loss": 2.3887,
      "step": 2168
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.40364064417888823,
      "learning_rate": 7.942733738924845e-08,
      "loss": 2.4609,
      "step": 2169
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.40581157364564485,
      "learning_rate": 6.6741308798135e-08,
      "loss": 2.5137,
      "step": 2170
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.42031822621411613,
      "learning_rate": 5.5158319419934546e-08,
      "loss": 2.3652,
      "step": 2171
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.4284202460055003,
      "learning_rate": 4.4678394810981906e-08,
      "loss": 2.3633,
      "step": 2172
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.41313168512486914,
      "learning_rate": 3.5301558093947527e-08,
      "loss": 2.4805,
      "step": 2173
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.4375813663445536,
      "learning_rate": 2.7027829957559925e-08,
      "loss": 2.4316,
      "step": 2174
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.4315881830098248,
      "learning_rate": 1.985722865682771e-08,
      "loss": 2.4277,
      "step": 2175
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.4206853495066232,
      "learning_rate": 1.3789770012762048e-08,
      "loss": 2.3457,
      "step": 2176
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.39881222991380016,
      "learning_rate": 8.825467412376665e-09,
      "loss": 2.3594,
      "step": 2177
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.40341961117565706,
      "learning_rate": 4.96433180879885e-09,
      "loss": 2.4043,
      "step": 2178
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.42310326696591716,
      "learning_rate": 2.206371721158451e-09,
      "loss": 2.5332,
      "step": 2179
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.4046004097215142,
      "learning_rate": 5.515932345323549e-10,
      "loss": 2.4141,
      "step": 2180
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.41262414215219584,
      "learning_rate": 0.0,
      "loss": 2.4785,
      "step": 2181
    },
    {
      "epoch": 1.0,
      "step": 2181,
      "total_flos": 7.408693923838689e+17,
      "train_loss": 2.5682666888755157,
      "train_runtime": 6711.7656,
      "train_samples_per_second": 83.157,
      "train_steps_per_second": 0.325
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 2181,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000,
  "total_flos": 7.408693923838689e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}