{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9998853342506593,
  "eval_steps": 500,
  "global_step": 2180,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 82.95042504892388,
      "learning_rate": 1.5151515151515153e-05,
      "loss": 10.2527,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 82.634932522638,
      "learning_rate": 3.0303030303030306e-05,
      "loss": 10.3101,
      "step": 2
    },
    {
      "epoch": 0.0,
      "grad_norm": 59.38979487176437,
      "learning_rate": 4.545454545454546e-05,
      "loss": 9.4612,
      "step": 3
    },
    {
      "epoch": 0.0,
      "grad_norm": 45.30988225692408,
      "learning_rate": 6.060606060606061e-05,
      "loss": 7.3369,
      "step": 4
    },
    {
      "epoch": 0.0,
      "grad_norm": 20.73490168789013,
      "learning_rate": 7.575757575757576e-05,
      "loss": 6.4832,
      "step": 5
    },
    {
      "epoch": 0.0,
      "grad_norm": 17.60709753388363,
      "learning_rate": 9.090909090909092e-05,
      "loss": 6.0726,
      "step": 6
    },
    {
      "epoch": 0.0,
      "grad_norm": 14.659560795415013,
      "learning_rate": 0.00010606060606060606,
      "loss": 5.8327,
      "step": 7
    },
    {
      "epoch": 0.0,
      "grad_norm": 15.751176871384166,
      "learning_rate": 0.00012121212121212122,
      "loss": 5.8578,
      "step": 8
    },
    {
      "epoch": 0.0,
      "grad_norm": 12.557351640349488,
      "learning_rate": 0.00013636363636363637,
      "loss": 5.5952,
      "step": 9
    },
    {
      "epoch": 0.0,
      "grad_norm": 6.623609285962712,
      "learning_rate": 0.00015151515151515152,
      "loss": 5.337,
      "step": 10
    },
    {
      "epoch": 0.01,
      "grad_norm": 4.471177355492287,
      "learning_rate": 0.00016666666666666666,
      "loss": 5.2032,
      "step": 11
    },
    {
      "epoch": 0.01,
      "grad_norm": 4.1333928108391085,
      "learning_rate": 0.00018181818181818183,
      "loss": 5.1763,
      "step": 12
    },
    {
      "epoch": 0.01,
      "grad_norm": 4.18549508724102,
      "learning_rate": 0.00019696969696969695,
      "loss": 5.0303,
      "step": 13
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.6006917274328964,
      "learning_rate": 0.00021212121212121213,
      "loss": 5.1017,
      "step": 14
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.1435469313041,
      "learning_rate": 0.00022727272727272727,
      "loss": 5.0118,
      "step": 15
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.994855336823533,
      "learning_rate": 0.00024242424242424245,
      "loss": 4.9557,
      "step": 16
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.6131030295208313,
      "learning_rate": 0.00025757575757575756,
      "loss": 4.8665,
      "step": 17
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.4036496870065043,
      "learning_rate": 0.00027272727272727274,
      "loss": 4.849,
      "step": 18
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.346828766867057,
      "learning_rate": 0.0002878787878787879,
      "loss": 4.7954,
      "step": 19
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.0072967735884673,
      "learning_rate": 0.00030303030303030303,
      "loss": 4.5698,
      "step": 20
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.8653457431256537,
      "learning_rate": 0.0003181818181818182,
      "loss": 4.6119,
      "step": 21
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.657378223951846,
      "learning_rate": 0.0003333333333333333,
      "loss": 4.6384,
      "step": 22
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.508574856291383,
      "learning_rate": 0.0003484848484848485,
      "loss": 4.4947,
      "step": 23
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.512618896645684,
      "learning_rate": 0.00036363636363636367,
      "loss": 4.4894,
      "step": 24
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.2561733765065244,
      "learning_rate": 0.0003787878787878788,
      "loss": 4.4936,
      "step": 25
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.3010975917353673,
      "learning_rate": 0.0003939393939393939,
      "loss": 4.3015,
      "step": 26
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.439918891147037,
      "learning_rate": 0.00040909090909090913,
      "loss": 4.2854,
      "step": 27
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.2229129739557523,
      "learning_rate": 0.00042424242424242425,
      "loss": 4.3152,
      "step": 28
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.3207747389141722,
      "learning_rate": 0.0004393939393939394,
      "loss": 4.31,
      "step": 29
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.4141744751584322,
      "learning_rate": 0.00045454545454545455,
      "loss": 4.2469,
      "step": 30
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.3126980101093566,
      "learning_rate": 0.0004696969696969697,
      "loss": 4.1531,
      "step": 31
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.1515487556436643,
      "learning_rate": 0.0004848484848484849,
      "loss": 4.1459,
      "step": 32
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.0839462546825658,
      "learning_rate": 0.0005,
      "loss": 4.1853,
      "step": 33
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.1174369970212654,
      "learning_rate": 0.0005151515151515151,
      "loss": 3.9901,
      "step": 34
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.8991452525125105,
      "learning_rate": 0.0005303030303030302,
      "loss": 4.0768,
      "step": 35
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.9217705652374459,
      "learning_rate": 0.0005454545454545455,
      "loss": 4.0243,
      "step": 36
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.9692396150456398,
      "learning_rate": 0.0005606060606060606,
      "loss": 3.9632,
      "step": 37
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.8947883586802508,
      "learning_rate": 0.0005757575757575758,
      "loss": 3.9983,
      "step": 38
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.7931454653044937,
      "learning_rate": 0.0005909090909090909,
      "loss": 3.8517,
      "step": 39
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.9458623829252094,
      "learning_rate": 0.0006060606060606061,
      "loss": 3.8583,
      "step": 40
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.7957216586548687,
      "learning_rate": 0.0006212121212121212,
      "loss": 3.8007,
      "step": 41
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.7306514455528892,
      "learning_rate": 0.0006363636363636364,
      "loss": 3.8011,
      "step": 42
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.7758085901387844,
      "learning_rate": 0.0006515151515151515,
      "loss": 3.7763,
      "step": 43
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.7763351537262178,
      "learning_rate": 0.0006666666666666666,
      "loss": 3.7814,
      "step": 44
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.6241573491178961,
      "learning_rate": 0.0006818181818181818,
      "loss": 3.7329,
      "step": 45
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.7030848764774981,
      "learning_rate": 0.000696969696969697,
      "loss": 3.6096,
      "step": 46
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.6758237052615907,
      "learning_rate": 0.0007121212121212122,
      "loss": 3.6531,
      "step": 47
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.6140459112609652,
      "learning_rate": 0.0007272727272727273,
      "loss": 3.6726,
      "step": 48
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.6039650058864519,
      "learning_rate": 0.0007424242424242425,
      "loss": 3.5842,
      "step": 49
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.6269049196775377,
      "learning_rate": 0.0007575757575757576,
      "loss": 3.5942,
      "step": 50
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.5902358030487995,
      "learning_rate": 0.0007727272727272727,
      "loss": 3.5121,
      "step": 51
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.5412101536670967,
      "learning_rate": 0.0007878787878787878,
      "loss": 3.6796,
      "step": 52
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.5562674203938032,
      "learning_rate": 0.000803030303030303,
      "loss": 3.552,
      "step": 53
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.6363067536462007,
      "learning_rate": 0.0008181818181818183,
      "loss": 3.6345,
      "step": 54
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.5224737046467328,
      "learning_rate": 0.0008333333333333334,
      "loss": 3.5754,
      "step": 55
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.567462218285191,
      "learning_rate": 0.0008484848484848485,
      "loss": 3.4982,
      "step": 56
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.5107192708394519,
      "learning_rate": 0.0008636363636363636,
      "loss": 3.5486,
      "step": 57
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.4523602848121307,
      "learning_rate": 0.0008787878787878789,
      "loss": 3.5356,
      "step": 58
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.46972095535862,
      "learning_rate": 0.000893939393939394,
      "loss": 3.3634,
      "step": 59
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.43189992399796573,
      "learning_rate": 0.0009090909090909091,
      "loss": 3.3904,
      "step": 60
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.45454507127246835,
      "learning_rate": 0.0009242424242424242,
      "loss": 3.5306,
      "step": 61
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.4333834503064586,
      "learning_rate": 0.0009393939393939394,
      "loss": 3.464,
      "step": 62
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.430370314226551,
      "learning_rate": 0.0009545454545454546,
      "loss": 3.5201,
      "step": 63
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.4070857805816067,
      "learning_rate": 0.0009696969696969698,
      "loss": 3.401,
      "step": 64
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.3902441892380618,
      "learning_rate": 0.000984848484848485,
      "loss": 3.388,
      "step": 65
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.36684798610976865,
      "learning_rate": 0.001,
      "loss": 3.3986,
      "step": 66
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.3748246948706842,
      "learning_rate": 0.0009999994478847943,
      "loss": 3.4062,
      "step": 67
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.38689305793143314,
      "learning_rate": 0.0009999977915403962,
      "loss": 3.4211,
      "step": 68
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.36836414708887383,
      "learning_rate": 0.0009999950309704639,
      "loss": 3.4363,
      "step": 69
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.3842648842329544,
      "learning_rate": 0.000999991166181094,
      "loss": 3.4129,
      "step": 70
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.4139438816846334,
      "learning_rate": 0.0009999861971808216,
      "loss": 3.3786,
      "step": 71
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.3655164012498137,
      "learning_rate": 0.0009999801239806208,
      "loss": 3.3682,
      "step": 72
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.31760042009982875,
      "learning_rate": 0.0009999729465939035,
      "loss": 3.2977,
      "step": 73
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.37437452002795735,
      "learning_rate": 0.0009999646650365212,
      "loss": 3.331,
      "step": 74
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.38632961474556765,
      "learning_rate": 0.0009999552793267634,
      "loss": 3.2532,
      "step": 75
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.350746962296021,
      "learning_rate": 0.0009999447894853577,
      "loss": 3.4019,
      "step": 76
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3330969138225619,
      "learning_rate": 0.0009999331955354708,
      "loss": 3.4745,
      "step": 77
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3450426649810674,
      "learning_rate": 0.0009999204975027073,
      "loss": 3.2767,
      "step": 78
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3414207348247738,
      "learning_rate": 0.0009999066954151103,
      "loss": 3.3253,
      "step": 79
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.32860562975299573,
      "learning_rate": 0.0009998917893031614,
      "loss": 3.3314,
      "step": 80
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.30675058337636746,
      "learning_rate": 0.0009998757791997801,
      "loss": 3.336,
      "step": 81
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3544647815926881,
      "learning_rate": 0.0009998586651403238,
      "loss": 3.3244,
      "step": 82
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3221893415119645,
      "learning_rate": 0.0009998404471625885,
      "loss": 3.2662,
      "step": 83
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3241632979754274,
      "learning_rate": 0.0009998211253068078,
      "loss": 3.4099,
      "step": 84
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3432838076210636,
      "learning_rate": 0.0009998006996156535,
      "loss": 3.2894,
      "step": 85
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.32462687705170473,
      "learning_rate": 0.0009997791701342347,
      "loss": 3.3791,
      "step": 86
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3632567456647841,
      "learning_rate": 0.0009997565369100983,
      "loss": 3.3034,
      "step": 87
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.30859191730940755,
      "learning_rate": 0.0009997327999932291,
      "loss": 3.289,
      "step": 88
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.36227030975870295,
      "learning_rate": 0.000999707959436049,
      "loss": 3.3113,
      "step": 89
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.31358727930263836,
      "learning_rate": 0.0009996820152934176,
      "loss": 3.2404,
      "step": 90
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3714043672867009,
      "learning_rate": 0.000999654967622631,
      "loss": 3.3233,
      "step": 91
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3365585575968304,
      "learning_rate": 0.0009996268164834238,
      "loss": 3.2952,
      "step": 92
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3209548476981024,
      "learning_rate": 0.000999597561937966,
      "loss": 3.2823,
      "step": 93
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.44443599223092217,
      "learning_rate": 0.0009995672040508656,
      "loss": 3.2045,
      "step": 94
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.37018119167061425,
      "learning_rate": 0.0009995357428891662,
      "loss": 3.2339,
      "step": 95
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3997932214917397,
      "learning_rate": 0.0009995031785223491,
      "loss": 3.2935,
      "step": 96
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3476568391682883,
      "learning_rate": 0.000999469511022331,
      "loss": 3.247,
      "step": 97
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.38919076677780207,
      "learning_rate": 0.0009994347404634657,
      "loss": 3.2571,
      "step": 98
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.3348352205240213,
      "learning_rate": 0.0009993988669225423,
      "loss": 3.2614,
      "step": 99
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.3644092409271942,
      "learning_rate": 0.000999361890478786,
      "loss": 3.2118,
      "step": 100
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.36760257059072127,
      "learning_rate": 0.0009993238112138583,
      "loss": 3.2362,
      "step": 101
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.3240184012151109,
      "learning_rate": 0.0009992846292118554,
      "loss": 3.2361,
      "step": 102
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.3269159342841765,
      "learning_rate": 0.000999244344559309,
      "loss": 3.2569,
      "step": 103
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.4242037595737302,
      "learning_rate": 0.0009992029573451869,
      "loss": 3.1982,
      "step": 104
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.39979388743208516,
      "learning_rate": 0.0009991604676608905,
      "loss": 3.1784,
      "step": 105
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.3481968028656099,
      "learning_rate": 0.0009991168756002568,
      "loss": 3.2969,
      "step": 106
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.3313078664749429,
      "learning_rate": 0.0009990721812595574,
      "loss": 3.1431,
      "step": 107
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.40622321803546924,
      "learning_rate": 0.0009990263847374976,
      "loss": 3.2725,
      "step": 108
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.4072712536663883,
      "learning_rate": 0.0009989794861352173,
      "loss": 3.1953,
      "step": 109
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.4257972215354896,
      "learning_rate": 0.0009989314855562905,
      "loss": 3.3183,
      "step": 110
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.35612820816498447,
      "learning_rate": 0.0009988823831067245,
      "loss": 3.2339,
      "step": 111
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.3780899191640233,
      "learning_rate": 0.0009988321788949597,
      "loss": 3.2658,
      "step": 112
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.3585646473716321,
      "learning_rate": 0.0009987808730318709,
      "loss": 3.268,
      "step": 113
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.3308050111655194,
      "learning_rate": 0.0009987284656307644,
      "loss": 3.2384,
      "step": 114
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.38714683287773644,
      "learning_rate": 0.0009986749568073802,
      "loss": 3.2472,
      "step": 115
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.3611136394905656,
      "learning_rate": 0.0009986203466798905,
      "loss": 3.2638,
      "step": 116
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.4028315098952853,
      "learning_rate": 0.0009985646353688996,
      "loss": 3.2306,
      "step": 117
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.37519788342795735,
      "learning_rate": 0.0009985078229974437,
      "loss": 3.2431,
      "step": 118
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.44015629329538253,
      "learning_rate": 0.0009984499096909905,
      "loss": 3.1234,
      "step": 119
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.38116146787362765,
      "learning_rate": 0.0009983908955774397,
      "loss": 3.2417,
      "step": 120
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.35083636950449976,
      "learning_rate": 0.0009983307807871211,
      "loss": 3.2439,
      "step": 121
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.35768596507000633,
      "learning_rate": 0.0009982695654527965,
      "loss": 3.2685,
      "step": 122
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.38409069670430745,
      "learning_rate": 0.0009982072497096571,
      "loss": 3.2577,
      "step": 123
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.325030479031564,
      "learning_rate": 0.000998143833695325,
      "loss": 3.1419,
      "step": 124
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.34186076486867273,
      "learning_rate": 0.0009980793175498517,
      "loss": 3.2149,
      "step": 125
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4305950083005351,
      "learning_rate": 0.000998013701415719,
      "loss": 3.2738,
      "step": 126
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.3523796900377572,
      "learning_rate": 0.0009979469854378372,
      "loss": 3.247,
      "step": 127
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4138432322336785,
      "learning_rate": 0.000997879169763546,
      "loss": 3.2371,
      "step": 128
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4275854926997472,
      "learning_rate": 0.000997810254542614,
      "loss": 3.1366,
      "step": 129
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.45822447879154155,
      "learning_rate": 0.0009977402399272374,
      "loss": 3.3312,
      "step": 130
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.36552121871476567,
      "learning_rate": 0.0009976691260720407,
      "loss": 3.1981,
      "step": 131
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.424912645043032,
      "learning_rate": 0.0009975969131340763,
      "loss": 3.2396,
      "step": 132
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4696508041944591,
      "learning_rate": 0.0009975236012728236,
      "loss": 3.2391,
      "step": 133
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.37162035902748786,
      "learning_rate": 0.0009974491906501886,
      "loss": 3.1937,
      "step": 134
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.3789688972753476,
      "learning_rate": 0.0009973736814305049,
      "loss": 3.183,
      "step": 135
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.477371287222042,
      "learning_rate": 0.0009972970737805312,
      "loss": 3.2383,
      "step": 136
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4192238355119607,
      "learning_rate": 0.0009972193678694525,
      "loss": 3.2077,
      "step": 137
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4097754814835854,
      "learning_rate": 0.0009971405638688794,
      "loss": 3.1663,
      "step": 138
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4063648246679404,
      "learning_rate": 0.0009970606619528475,
      "loss": 3.1732,
      "step": 139
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4571912921126169,
      "learning_rate": 0.000996979662297817,
      "loss": 3.2438,
      "step": 140
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.4358624955959288,
      "learning_rate": 0.0009968975650826721,
      "loss": 3.3695,
      "step": 141
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.40133556665542536,
      "learning_rate": 0.000996814370488722,
      "loss": 3.2588,
      "step": 142
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.47379758831256114,
      "learning_rate": 0.000996730078699698,
      "loss": 3.2042,
      "step": 143
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4033536379862544,
      "learning_rate": 0.0009966446899017558,
      "loss": 3.2582,
      "step": 144
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.38717560919840516,
      "learning_rate": 0.0009965582042834728,
      "loss": 3.2198,
      "step": 145
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.3983717751362564,
      "learning_rate": 0.0009964706220358492,
      "loss": 3.13,
      "step": 146
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.398321585295696,
      "learning_rate": 0.000996381943352307,
      "loss": 3.1809,
      "step": 147
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.38508766872231004,
      "learning_rate": 0.0009962921684286896,
      "loss": 3.1459,
      "step": 148
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4196782794416121,
      "learning_rate": 0.0009962012974632614,
      "loss": 3.3174,
      "step": 149
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.38312506901587917,
      "learning_rate": 0.0009961093306567075,
      "loss": 3.1609,
      "step": 150
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.3834026463850755,
      "learning_rate": 0.0009960162682121328,
      "loss": 3.1727,
      "step": 151
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.37212533331129016,
      "learning_rate": 0.0009959221103350623,
      "loss": 3.2118,
      "step": 152
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.3585062598718845,
      "learning_rate": 0.0009958268572334394,
      "loss": 3.2316,
      "step": 153
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.41745934958868924,
      "learning_rate": 0.0009957305091176274,
      "loss": 3.2481,
      "step": 154
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.34402519353680516,
      "learning_rate": 0.0009956330662004075,
      "loss": 3.1339,
      "step": 155
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.42000843418190076,
      "learning_rate": 0.0009955345286969779,
      "loss": 3.1269,
      "step": 156
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.3405992400300467,
      "learning_rate": 0.0009954348968249551,
      "loss": 3.2557,
      "step": 157
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.39744079187691733,
      "learning_rate": 0.0009953341708043724,
      "loss": 3.3068,
      "step": 158
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.3404108657386337,
      "learning_rate": 0.0009952323508576793,
      "loss": 3.1693,
      "step": 159
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.3950684579015263,
      "learning_rate": 0.0009951294372097406,
      "loss": 3.1515,
      "step": 160
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4482362340996043,
      "learning_rate": 0.0009950254300878378,
      "loss": 3.1808,
      "step": 161
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.38020677113644147,
      "learning_rate": 0.000994920329721666,
      "loss": 3.1741,
      "step": 162
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.3599945744609763,
      "learning_rate": 0.0009948141363433356,
      "loss": 3.1914,
      "step": 163
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.40622969230815925,
      "learning_rate": 0.00099470685018737,
      "loss": 3.0915,
      "step": 164
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.4285939608121229,
      "learning_rate": 0.0009945984714907073,
      "loss": 3.1998,
      "step": 165
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.45402284588934233,
      "learning_rate": 0.000994489000492697,
      "loss": 3.125,
      "step": 166
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.3504381721917374,
      "learning_rate": 0.0009943784374351016,
      "loss": 3.0695,
      "step": 167
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.41291307737428556,
      "learning_rate": 0.0009942667825620951,
      "loss": 3.145,
      "step": 168
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.4121843883779206,
      "learning_rate": 0.0009941540361202634,
      "loss": 3.154,
      "step": 169
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.42824905618839176,
      "learning_rate": 0.0009940401983586022,
      "loss": 3.1346,
      "step": 170
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.45579503995367526,
      "learning_rate": 0.000993925269528518,
      "loss": 3.1215,
      "step": 171
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.4267455863227672,
      "learning_rate": 0.0009938092498838265,
      "loss": 3.1676,
      "step": 172
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.47980485147618696,
      "learning_rate": 0.0009936921396807524,
      "loss": 3.1425,
      "step": 173
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.40237071602296964,
      "learning_rate": 0.0009935739391779292,
      "loss": 3.1554,
      "step": 174
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.43339561257680276,
      "learning_rate": 0.000993454648636398,
      "loss": 3.0667,
      "step": 175
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.45262843009609055,
      "learning_rate": 0.0009933342683196074,
      "loss": 3.2084,
      "step": 176
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.42919683421217986,
      "learning_rate": 0.0009932127984934125,
      "loss": 3.1305,
      "step": 177
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.42736970339734914,
      "learning_rate": 0.0009930902394260745,
      "loss": 3.1811,
      "step": 178
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.47906720240389955,
      "learning_rate": 0.0009929665913882607,
      "loss": 3.1476,
      "step": 179
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.45727421220655834,
      "learning_rate": 0.0009928418546530425,
      "loss": 3.1152,
      "step": 180
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.3964710652365098,
      "learning_rate": 0.0009927160294958964,
      "loss": 3.1728,
      "step": 181
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.43066772330863295,
      "learning_rate": 0.000992589116194702,
      "loss": 3.0938,
      "step": 182
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.4278315161148297,
      "learning_rate": 0.000992461115029743,
      "loss": 3.2027,
      "step": 183
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.4165564813405439,
      "learning_rate": 0.000992332026283704,
      "loss": 3.1369,
      "step": 184
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.44490859123358745,
      "learning_rate": 0.0009922018502416736,
      "loss": 3.1791,
      "step": 185
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4297239385990805,
      "learning_rate": 0.0009920705871911395,
      "loss": 3.0978,
      "step": 186
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4164329344297113,
      "learning_rate": 0.0009919382374219915,
      "loss": 3.0626,
      "step": 187
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4240943922812887,
      "learning_rate": 0.0009918048012265187,
      "loss": 3.1073,
      "step": 188
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.37505297434649765,
      "learning_rate": 0.0009916702788994097,
      "loss": 3.0991,
      "step": 189
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.39210701995667224,
      "learning_rate": 0.0009915346707377519,
      "loss": 3.0966,
      "step": 190
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.38874774590113353,
      "learning_rate": 0.0009913979770410305,
      "loss": 3.0343,
      "step": 191
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4378272803448782,
      "learning_rate": 0.0009912601981111285,
      "loss": 3.1417,
      "step": 192
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.41803064162729764,
      "learning_rate": 0.0009911213342523248,
      "loss": 3.141,
      "step": 193
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.38800768847334605,
      "learning_rate": 0.000990981385771295,
      "loss": 3.1103,
      "step": 194
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4158892914115033,
      "learning_rate": 0.00099084035297711,
      "loss": 3.2183,
      "step": 195
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.43483103388982164,
      "learning_rate": 0.000990698236181235,
      "loss": 3.1441,
      "step": 196
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.400079061193393,
      "learning_rate": 0.0009905550356975293,
      "loss": 3.134,
      "step": 197
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4373421005675195,
      "learning_rate": 0.0009904107518422457,
      "loss": 3.08,
      "step": 198
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4658531688683174,
      "learning_rate": 0.0009902653849340295,
      "loss": 3.1103,
      "step": 199
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.5014574119656803,
      "learning_rate": 0.0009901189352939177,
      "loss": 3.2289,
      "step": 200
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4554518699498125,
      "learning_rate": 0.0009899714032453387,
      "loss": 3.2423,
      "step": 201
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4694856298039221,
      "learning_rate": 0.000989822789114111,
      "loss": 3.148,
      "step": 202
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.41830359322807736,
      "learning_rate": 0.0009896730932284434,
      "loss": 3.1455,
      "step": 203
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.44864494604252647,
      "learning_rate": 0.0009895223159189332,
      "loss": 3.1112,
      "step": 204
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4187237689782655,
      "learning_rate": 0.0009893704575185663,
      "loss": 3.215,
      "step": 205
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.38269763029338416,
      "learning_rate": 0.000989217518362716,
      "loss": 3.1615,
      "step": 206
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4429232685444736,
      "learning_rate": 0.0009890634987891425,
      "loss": 3.1956,
      "step": 207
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.5060709084581992,
      "learning_rate": 0.0009889083991379917,
      "loss": 3.1378,
      "step": 208
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4049674656253958,
      "learning_rate": 0.0009887522197517954,
      "loss": 3.0508,
      "step": 209
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.40092360523144455,
      "learning_rate": 0.0009885949609754693,
      "loss": 3.1287,
      "step": 210
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.508749976861702,
      "learning_rate": 0.000988436623156314,
      "loss": 3.1596,
      "step": 211
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.45643647222065215,
      "learning_rate": 0.0009882772066440114,
      "loss": 3.1289,
      "step": 212
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.39709749740612094,
      "learning_rate": 0.0009881167117906276,
      "loss": 3.1457,
      "step": 213
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.5008607848796507,
      "learning_rate": 0.0009879551389506084,
      "loss": 3.1286,
      "step": 214
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4718674449530359,
      "learning_rate": 0.0009877924884807814,
      "loss": 3.0666,
      "step": 215
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.45993404115378045,
      "learning_rate": 0.000987628760740354,
      "loss": 3.0883,
      "step": 216
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.41001673991185883,
      "learning_rate": 0.0009874639560909118,
      "loss": 3.1061,
      "step": 217
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.45676599715528154,
      "learning_rate": 0.0009872980748964202,
      "loss": 3.0723,
      "step": 218
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4774796859660166,
      "learning_rate": 0.000987131117523221,
      "loss": 3.1656,
      "step": 219
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.39048001670056587,
      "learning_rate": 0.000986963084340033,
      "loss": 3.1426,
      "step": 220
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4231483499082633,
      "learning_rate": 0.0009867939757179508,
      "loss": 3.1827,
      "step": 221
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4727524817393217,
      "learning_rate": 0.0009866237920304443,
      "loss": 3.106,
      "step": 222
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4194575418604502,
      "learning_rate": 0.0009864525336533577,
      "loss": 3.1018,
      "step": 223
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.5006044290698525,
      "learning_rate": 0.000986280200964908,
      "loss": 3.0744,
      "step": 224
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4401855145250986,
      "learning_rate": 0.0009861067943456856,
      "loss": 3.08,
      "step": 225
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.44489706477495233,
      "learning_rate": 0.000985932314178652,
      "loss": 3.103,
      "step": 226
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.40791050146457725,
      "learning_rate": 0.00098575676084914,
      "loss": 3.0859,
      "step": 227
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.42377460089817603,
      "learning_rate": 0.0009855801347448518,
      "loss": 3.0935,
      "step": 228
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.45592980379171116,
      "learning_rate": 0.0009854024362558596,
      "loss": 3.1125,
      "step": 229
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.4118101691213073,
      "learning_rate": 0.0009852236657746035,
      "loss": 3.1589,
      "step": 230
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5272881338571506,
      "learning_rate": 0.0009850438236958911,
      "loss": 3.0957,
      "step": 231
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.4490047956941026,
      "learning_rate": 0.0009848629104168966,
      "loss": 3.1208,
      "step": 232
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.43782874471630884,
      "learning_rate": 0.00098468092633716,
      "loss": 3.0995,
      "step": 233
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.48116080637997155,
      "learning_rate": 0.0009844978718585855,
      "loss": 3.0628,
      "step": 234
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.542738686660602,
      "learning_rate": 0.0009843137473854423,
      "loss": 3.0773,
      "step": 235
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5344688972610916,
      "learning_rate": 0.000984128553324362,
      "loss": 3.1078,
      "step": 236
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.45444094037694366,
      "learning_rate": 0.0009839422900843383,
      "loss": 3.0386,
      "step": 237
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.493470439418435,
      "learning_rate": 0.0009837549580767261,
      "loss": 3.0623,
      "step": 238
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5418050329289771,
      "learning_rate": 0.0009835665577152411,
      "loss": 3.0467,
      "step": 239
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5055816469702586,
      "learning_rate": 0.000983377089415958,
      "loss": 3.1215,
      "step": 240
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.4359793703137916,
      "learning_rate": 0.0009831865535973102,
      "loss": 3.097,
      "step": 241
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5920888473542507,
      "learning_rate": 0.0009829949506800885,
      "loss": 3.1075,
      "step": 242
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5313870573864672,
      "learning_rate": 0.0009828022810874405,
      "loss": 3.1538,
      "step": 243
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.49934752960555806,
      "learning_rate": 0.0009826085452448693,
      "loss": 3.1054,
      "step": 244
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.43099905853553383,
      "learning_rate": 0.000982413743580233,
      "loss": 3.0721,
      "step": 245
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.43121090914085797,
      "learning_rate": 0.0009822178765237436,
      "loss": 3.0757,
      "step": 246
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.481990293258324,
      "learning_rate": 0.0009820209445079654,
      "loss": 3.096,
      "step": 247
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5061323287985197,
      "learning_rate": 0.0009818229479678158,
      "loss": 3.1012,
      "step": 248
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.4785333831749571,
      "learning_rate": 0.0009816238873405615,
      "loss": 3.1351,
      "step": 249
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.46682488420438106,
      "learning_rate": 0.0009814237630658207,
      "loss": 3.1114,
      "step": 250
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.5072455535157702,
      "learning_rate": 0.00098122257558556,
      "loss": 3.2007,
      "step": 251
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.40041882998300415,
      "learning_rate": 0.0009810203253440937,
      "loss": 3.123,
      "step": 252
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4496159830126479,
      "learning_rate": 0.0009808170127880837,
      "loss": 3.1186,
      "step": 253
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.484003873950594,
      "learning_rate": 0.000980612638366538,
      "loss": 3.0687,
      "step": 254
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.42442397592498443,
      "learning_rate": 0.0009804072025308096,
      "loss": 3.168,
      "step": 255
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.46377527750237707,
      "learning_rate": 0.000980200705734595,
      "loss": 3.1004,
      "step": 256
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4446499437423701,
      "learning_rate": 0.0009799931484339344,
      "loss": 3.0519,
      "step": 257
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.5041640794359902,
      "learning_rate": 0.0009797845310872103,
      "loss": 3.1724,
      "step": 258
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.47835857384331953,
      "learning_rate": 0.0009795748541551457,
      "loss": 3.1003,
      "step": 259
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.434898596894168,
      "learning_rate": 0.000979364118100804,
      "loss": 3.1418,
      "step": 260
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4318498174627635,
      "learning_rate": 0.0009791523233895875,
      "loss": 3.0687,
      "step": 261
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4318970309259732,
      "learning_rate": 0.0009789394704892364,
      "loss": 3.0405,
      "step": 262
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.476703679735708,
      "learning_rate": 0.0009787255598698282,
      "loss": 3.1871,
      "step": 263
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.41057241553203194,
      "learning_rate": 0.0009785105920037758,
      "loss": 3.0853,
      "step": 264
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4835283589151547,
      "learning_rate": 0.0009782945673658275,
      "loss": 3.0942,
      "step": 265
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.5385887655656103,
      "learning_rate": 0.0009780774864330654,
      "loss": 3.1635,
      "step": 266
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4195330117826838,
      "learning_rate": 0.000977859349684904,
      "loss": 3.1021,
      "step": 267
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.43818416797584986,
      "learning_rate": 0.00097764015760309,
      "loss": 3.0344,
      "step": 268
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.5162035278358195,
      "learning_rate": 0.0009774199106717004,
      "loss": 3.1493,
      "step": 269
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4567388393472109,
      "learning_rate": 0.0009771986093771417,
      "loss": 3.1545,
      "step": 270
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.49344034923911934,
      "learning_rate": 0.0009769762542081496,
      "loss": 3.1004,
      "step": 271
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.5034562936667282,
      "learning_rate": 0.000976752845655786,
      "loss": 3.0405,
      "step": 272
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.44110073582390935,
      "learning_rate": 0.0009765283842134411,
      "loss": 3.0355,
      "step": 273
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.3962732003629882,
      "learning_rate": 0.0009763028703768282,
      "loss": 3.0247,
      "step": 274
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4200951475097383,
      "learning_rate": 0.0009760763046439862,
      "loss": 2.9413,
      "step": 275
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4394458881335293,
      "learning_rate": 0.0009758486875152766,
      "loss": 3.0569,
      "step": 276
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4576215314085917,
      "learning_rate": 0.0009756200194933829,
      "loss": 3.0147,
      "step": 277
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4543258327166807,
      "learning_rate": 0.0009753903010833094,
      "loss": 3.0853,
      "step": 278
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.5150532889800652,
      "learning_rate": 0.0009751595327923803,
      "loss": 3.0914,
      "step": 279
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4791776603488495,
      "learning_rate": 0.0009749277151302382,
      "loss": 3.0906,
      "step": 280
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.5305733772036967,
      "learning_rate": 0.0009746948486088435,
      "loss": 3.1035,
      "step": 281
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4976368794364665,
      "learning_rate": 0.0009744609337424727,
      "loss": 3.0905,
      "step": 282
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.5990072746458743,
      "learning_rate": 0.0009742259710477177,
      "loss": 3.0403,
      "step": 283
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.457180281105584,
      "learning_rate": 0.0009739899610434841,
      "loss": 3.0107,
      "step": 284
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.5691034753532734,
      "learning_rate": 0.0009737529042509913,
      "loss": 3.1911,
      "step": 285
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.43104504062352855,
      "learning_rate": 0.0009735148011937693,
      "loss": 3.0662,
      "step": 286
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4464040256436296,
      "learning_rate": 0.00097327565239766,
      "loss": 3.0768,
      "step": 287
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.48121574776448583,
      "learning_rate": 0.0009730354583908136,
      "loss": 3.0723,
      "step": 288
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4492604734314535,
      "learning_rate": 0.0009727942197036895,
      "loss": 3.1109,
      "step": 289
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4751256904836382,
      "learning_rate": 0.0009725519368690539,
      "loss": 3.0962,
      "step": 290
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.48966108126015545,
      "learning_rate": 0.0009723086104219787,
      "loss": 3.0486,
      "step": 291
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4415086536993296,
      "learning_rate": 0.0009720642408998409,
      "loss": 3.0539,
      "step": 292
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4955858288652964,
      "learning_rate": 0.0009718188288423211,
      "loss": 2.9687,
      "step": 293
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.5635385639569106,
      "learning_rate": 0.0009715723747914022,
      "loss": 3.0044,
      "step": 294
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5040733177428176,
      "learning_rate": 0.0009713248792913685,
      "loss": 2.9573,
      "step": 295
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.48763789572095345,
      "learning_rate": 0.0009710763428888037,
      "loss": 3.0118,
      "step": 296
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5089192549992846,
      "learning_rate": 0.0009708267661325909,
      "loss": 3.0716,
      "step": 297
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5704575486009696,
      "learning_rate": 0.0009705761495739107,
      "loss": 3.074,
      "step": 298
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4982231702246244,
      "learning_rate": 0.0009703244937662399,
      "loss": 2.9961,
      "step": 299
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.47686210036848165,
      "learning_rate": 0.0009700717992653505,
      "loss": 3.0401,
      "step": 300
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5043821939077378,
      "learning_rate": 0.0009698180666293083,
      "loss": 3.0293,
      "step": 301
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.51108815454753,
      "learning_rate": 0.000969563296418472,
      "loss": 3.0199,
      "step": 302
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.48029497523757336,
      "learning_rate": 0.0009693074891954914,
      "loss": 3.1206,
      "step": 303
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.48929534763175214,
      "learning_rate": 0.0009690506455253072,
      "loss": 3.1588,
      "step": 304
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4651349084870595,
      "learning_rate": 0.0009687927659751481,
      "loss": 3.0654,
      "step": 305
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4412143183011205,
      "learning_rate": 0.0009685338511145312,
      "loss": 3.0085,
      "step": 306
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5047303894510224,
      "learning_rate": 0.0009682739015152598,
      "loss": 2.9846,
      "step": 307
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4439234300306221,
      "learning_rate": 0.0009680129177514226,
      "loss": 3.0199,
      "step": 308
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5053855594822741,
      "learning_rate": 0.0009677509003993915,
      "loss": 2.9622,
      "step": 309
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4714928209978853,
      "learning_rate": 0.0009674878500378221,
      "loss": 3.0119,
      "step": 310
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5206857170514964,
      "learning_rate": 0.0009672237672476505,
      "loss": 3.0394,
      "step": 311
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5157904091378976,
      "learning_rate": 0.0009669586526120935,
      "loss": 3.1681,
      "step": 312
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5589361778593485,
      "learning_rate": 0.0009666925067166459,
      "loss": 3.0563,
      "step": 313
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4883117092623467,
      "learning_rate": 0.000966425330149081,
      "loss": 2.8973,
      "step": 314
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4655464739020218,
      "learning_rate": 0.0009661571234994475,
      "loss": 2.9918,
      "step": 315
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.48524124973975213,
      "learning_rate": 0.0009658878873600691,
      "loss": 3.0725,
      "step": 316
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4600346257725194,
      "learning_rate": 0.0009656176223255438,
      "loss": 3.0072,
      "step": 317
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4899746598711084,
      "learning_rate": 0.000965346328992741,
      "loss": 3.0979,
      "step": 318
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4682045574289672,
      "learning_rate": 0.0009650740079608014,
      "loss": 3.0828,
      "step": 319
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.46047176594216327,
      "learning_rate": 0.0009648006598311353,
      "loss": 3.123,
      "step": 320
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4388428568052404,
      "learning_rate": 0.0009645262852074214,
      "loss": 3.0697,
      "step": 321
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4985649784084978,
      "learning_rate": 0.0009642508846956053,
      "loss": 2.916,
      "step": 322
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4865522533091172,
      "learning_rate": 0.0009639744589038983,
      "loss": 3.0701,
      "step": 323
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4701109428139839,
      "learning_rate": 0.0009636970084427759,
      "loss": 2.9631,
      "step": 324
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4687897608554863,
      "learning_rate": 0.0009634185339249766,
      "loss": 3.0297,
      "step": 325
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.48690208768786464,
      "learning_rate": 0.0009631390359655003,
      "loss": 3.0389,
      "step": 326
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4822703859374579,
      "learning_rate": 0.0009628585151816074,
      "loss": 3.0992,
      "step": 327
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.49027420014184586,
      "learning_rate": 0.0009625769721928172,
      "loss": 3.1263,
      "step": 328
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.48041370623938345,
      "learning_rate": 0.0009622944076209061,
      "loss": 3.0158,
      "step": 329
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.44305260781128986,
      "learning_rate": 0.0009620108220899071,
      "loss": 3.0836,
      "step": 330
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.49600057755910937,
      "learning_rate": 0.0009617262162261075,
      "loss": 3.1382,
      "step": 331
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.47018596626971704,
      "learning_rate": 0.0009614405906580486,
      "loss": 3.0862,
      "step": 332
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.487893258555253,
      "learning_rate": 0.000961153946016523,
      "loss": 3.0112,
      "step": 333
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.5346228381478916,
      "learning_rate": 0.000960866282934574,
      "loss": 2.9964,
      "step": 334
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.5057431347484991,
      "learning_rate": 0.0009605776020474945,
      "loss": 3.1034,
      "step": 335
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.4982220444519792,
      "learning_rate": 0.0009602879039928249,
      "loss": 3.0489,
      "step": 336
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.5537396879593695,
      "learning_rate": 0.0009599971894103521,
      "loss": 2.9787,
      "step": 337
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.40627706080294745,
      "learning_rate": 0.0009597054589421077,
      "loss": 3.0086,
      "step": 338
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4879160651165343,
      "learning_rate": 0.0009594127132323669,
      "loss": 3.0945,
      "step": 339
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.46797694459842665,
      "learning_rate": 0.0009591189529276474,
      "loss": 3.0399,
      "step": 340
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4696602589181308,
      "learning_rate": 0.0009588241786767072,
      "loss": 3.0719,
      "step": 341
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4843098525339636,
      "learning_rate": 0.0009585283911305436,
      "loss": 3.0296,
      "step": 342
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4798243772095176,
      "learning_rate": 0.000958231590942392,
      "loss": 3.098,
      "step": 343
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.498609086823684,
      "learning_rate": 0.0009579337787677238,
      "loss": 3.0521,
      "step": 344
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4765578621322359,
      "learning_rate": 0.0009576349552642456,
      "loss": 2.9481,
      "step": 345
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4454407074455648,
      "learning_rate": 0.0009573351210918975,
      "loss": 3.0238,
      "step": 346
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4988400567288936,
      "learning_rate": 0.0009570342769128514,
      "loss": 2.9955,
      "step": 347
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.43168145214126674,
      "learning_rate": 0.0009567324233915099,
      "loss": 3.1314,
      "step": 348
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.5143204791193541,
      "learning_rate": 0.0009564295611945047,
      "loss": 2.9786,
      "step": 349
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4048829632590379,
      "learning_rate": 0.000956125690990695,
      "loss": 3.0386,
      "step": 350
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.606095112929325,
      "learning_rate": 0.0009558208134511665,
      "loss": 3.0629,
      "step": 351
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4550869204413142,
      "learning_rate": 0.0009555149292492289,
      "loss": 3.0376,
      "step": 352
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.47566921266912415,
      "learning_rate": 0.0009552080390604159,
      "loss": 3.1736,
      "step": 353
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.55080372234207,
      "learning_rate": 0.0009549001435624823,
      "loss": 2.9837,
      "step": 354
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4509504477067053,
      "learning_rate": 0.0009545912434354029,
      "loss": 3.0398,
      "step": 355
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.5338822513329282,
      "learning_rate": 0.0009542813393613721,
      "loss": 2.9519,
      "step": 356
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.5441582279188419,
      "learning_rate": 0.0009539704320248006,
      "loss": 3.0605,
      "step": 357
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.509224581576941,
      "learning_rate": 0.0009536585221123151,
      "loss": 2.9538,
      "step": 358
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4377590413705248,
      "learning_rate": 0.0009533456103127565,
      "loss": 2.9679,
      "step": 359
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.4429655378689488,
      "learning_rate": 0.000953031697317178,
      "loss": 3.0538,
      "step": 360
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5679104034208167,
      "learning_rate": 0.0009527167838188445,
      "loss": 2.9194,
      "step": 361
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.47529767192616873,
      "learning_rate": 0.0009524008705132299,
      "loss": 2.9776,
      "step": 362
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5534471815840221,
      "learning_rate": 0.0009520839580980166,
      "loss": 3.0446,
      "step": 363
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5286943923618481,
      "learning_rate": 0.0009517660472730929,
      "loss": 2.9446,
      "step": 364
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.47775327960168995,
      "learning_rate": 0.0009514471387405526,
      "loss": 2.9894,
      "step": 365
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5809941342275982,
      "learning_rate": 0.0009511272332046926,
      "loss": 2.9459,
      "step": 366
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5070919461332507,
      "learning_rate": 0.0009508063313720119,
      "loss": 3.0021,
      "step": 367
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.48152230628043413,
      "learning_rate": 0.0009504844339512095,
      "loss": 3.0805,
      "step": 368
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5054527516998113,
      "learning_rate": 0.0009501615416531835,
      "loss": 3.0184,
      "step": 369
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5146853715938834,
      "learning_rate": 0.0009498376551910285,
      "loss": 3.0452,
      "step": 370
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.43134805674947385,
      "learning_rate": 0.0009495127752800352,
      "loss": 3.0313,
      "step": 371
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5356726184825262,
      "learning_rate": 0.0009491869026376882,
      "loss": 3.0163,
      "step": 372
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.4899367322241046,
      "learning_rate": 0.0009488600379836648,
      "loss": 3.1271,
      "step": 373
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.4726827829894805,
      "learning_rate": 0.0009485321820398321,
      "loss": 3.0836,
      "step": 374
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.4663226272027926,
      "learning_rate": 0.0009482033355302475,
      "loss": 3.0792,
      "step": 375
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.6016945641109074,
      "learning_rate": 0.0009478734991811556,
      "loss": 2.9813,
      "step": 376
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.4881397438118027,
      "learning_rate": 0.0009475426737209871,
      "loss": 3.0804,
      "step": 377
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5529403080632629,
      "learning_rate": 0.000947210859880357,
      "loss": 3.0023,
      "step": 378
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5875267291466769,
      "learning_rate": 0.0009468780583920631,
      "loss": 2.9256,
      "step": 379
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5394188300653993,
      "learning_rate": 0.0009465442699910846,
      "loss": 3.0018,
      "step": 380
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5765294864804436,
      "learning_rate": 0.0009462094954145801,
      "loss": 2.997,
      "step": 381
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.48345874381753895,
      "learning_rate": 0.0009458737354018859,
      "loss": 2.9406,
      "step": 382
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6035973913974746,
      "learning_rate": 0.000945536990694515,
      "loss": 3.0295,
      "step": 383
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.5062688177104546,
      "learning_rate": 0.0009451992620361551,
      "loss": 3.0511,
      "step": 384
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.5180825486609733,
      "learning_rate": 0.0009448605501726664,
      "loss": 3.0151,
      "step": 385
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6046632514050444,
      "learning_rate": 0.000944520855852081,
      "loss": 2.9617,
      "step": 386
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.47843443440018185,
      "learning_rate": 0.0009441801798246002,
      "loss": 2.983,
      "step": 387
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4715912159681642,
      "learning_rate": 0.0009438385228425939,
      "loss": 3.0215,
      "step": 388
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6146005902208077,
      "learning_rate": 0.0009434958856605982,
      "loss": 3.0772,
      "step": 389
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4609760270508822,
      "learning_rate": 0.0009431522690353137,
      "loss": 2.928,
      "step": 390
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.5329345662254492,
      "learning_rate": 0.0009428076737256044,
      "loss": 2.9899,
      "step": 391
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.46676657691266404,
      "learning_rate": 0.0009424621004924954,
      "loss": 2.9449,
      "step": 392
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.44187559540363164,
      "learning_rate": 0.0009421155500991719,
      "loss": 3.0253,
      "step": 393
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.571003803057975,
      "learning_rate": 0.0009417680233109767,
      "loss": 3.0754,
      "step": 394
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.5305659030984635,
      "learning_rate": 0.000941419520895409,
      "loss": 3.002,
      "step": 395
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.5029704997484273,
      "learning_rate": 0.0009410700436221229,
      "loss": 3.0294,
      "step": 396
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4970159440525706,
      "learning_rate": 0.0009407195922629252,
      "loss": 2.9632,
      "step": 397
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4394418129773029,
      "learning_rate": 0.000940368167591774,
      "loss": 2.9151,
      "step": 398
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.5395774624072547,
      "learning_rate": 0.0009400157703847769,
      "loss": 2.9488,
      "step": 399
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4660245997485381,
      "learning_rate": 0.0009396624014201895,
      "loss": 3.048,
      "step": 400
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.47705779740329024,
      "learning_rate": 0.000939308061478413,
      "loss": 2.9699,
      "step": 401
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.45402379450769487,
      "learning_rate": 0.0009389527513419935,
      "loss": 3.0008,
      "step": 402
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.47358943094409517,
      "learning_rate": 0.0009385964717956195,
      "loss": 2.8775,
      "step": 403
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.6162999419541931,
      "learning_rate": 0.0009382392236261201,
      "loss": 3.0417,
      "step": 404
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.4987551168026171,
      "learning_rate": 0.0009378810076224644,
      "loss": 2.9643,
      "step": 405
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.548870078424536,
      "learning_rate": 0.0009375218245757582,
      "loss": 2.9958,
      "step": 406
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.518639074187955,
      "learning_rate": 0.0009371616752792432,
      "loss": 2.9839,
      "step": 407
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.5406094243266623,
      "learning_rate": 0.0009368005605282949,
      "loss": 2.9244,
      "step": 408
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.46251386339698414,
      "learning_rate": 0.0009364384811204212,
      "loss": 2.9786,
      "step": 409
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.4933066238767263,
      "learning_rate": 0.00093607543785526,
      "loss": 2.9733,
      "step": 410
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.548651226793732,
      "learning_rate": 0.0009357114315345787,
      "loss": 2.9623,
      "step": 411
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.4158226446837171,
      "learning_rate": 0.0009353464629622705,
      "loss": 2.8826,
      "step": 412
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.4228662108018181,
      "learning_rate": 0.0009349805329443544,
      "loss": 2.9811,
      "step": 413
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.5157418778471298,
      "learning_rate": 0.0009346136422889724,
      "loss": 3.0147,
      "step": 414
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.46735735665049033,
      "learning_rate": 0.0009342457918063882,
      "loss": 2.943,
      "step": 415
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.6376252592860742,
      "learning_rate": 0.0009338769823089853,
      "loss": 3.0437,
      "step": 416
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.49325825946383267,
      "learning_rate": 0.0009335072146112648,
      "loss": 2.9208,
      "step": 417
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.5596290314171709,
      "learning_rate": 0.0009331364895298444,
      "loss": 2.8705,
      "step": 418
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.4884054625815647,
      "learning_rate": 0.0009327648078834559,
      "loss": 2.9132,
      "step": 419
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.5496974839242864,
      "learning_rate": 0.0009323921704929434,
      "loss": 2.9552,
      "step": 420
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.49132631497978096,
      "learning_rate": 0.0009320185781812623,
      "loss": 2.9904,
      "step": 421
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.43657765190006026,
      "learning_rate": 0.0009316440317734762,
      "loss": 2.9211,
      "step": 422
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.48661805476874975,
      "learning_rate": 0.0009312685320967565,
      "loss": 2.9624,
      "step": 423
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.5135715364325427,
      "learning_rate": 0.0009308920799803793,
      "loss": 2.9397,
      "step": 424
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.5213564984984786,
      "learning_rate": 0.0009305146762557246,
      "loss": 2.9539,
      "step": 425
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.45148383956790444,
      "learning_rate": 0.0009301363217562736,
      "loss": 2.9351,
      "step": 426
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5960017161681906,
      "learning_rate": 0.0009297570173176074,
      "loss": 2.9276,
      "step": 427
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.511075499664819,
      "learning_rate": 0.000929376763777405,
      "loss": 2.9691,
      "step": 428
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5224780437181922,
      "learning_rate": 0.0009289955619754413,
      "loss": 3.0003,
      "step": 429
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5341936886872939,
      "learning_rate": 0.0009286134127535859,
      "loss": 2.925,
      "step": 430
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5729715396293762,
      "learning_rate": 0.0009282303169558,
      "loss": 2.9691,
      "step": 431
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5219776108137949,
      "learning_rate": 0.0009278462754281359,
      "loss": 2.9531,
      "step": 432
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5639021131909823,
      "learning_rate": 0.0009274612890187342,
      "loss": 3.0339,
      "step": 433
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.423981207360482,
      "learning_rate": 0.0009270753585778222,
      "loss": 2.9041,
      "step": 434
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.4879316860560613,
      "learning_rate": 0.0009266884849577124,
      "loss": 2.8857,
      "step": 435
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.533230155055393,
      "learning_rate": 0.0009263006690127998,
      "loss": 3.0521,
      "step": 436
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5419517537515356,
      "learning_rate": 0.0009259119115995609,
      "loss": 2.9755,
      "step": 437
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.49860727195367804,
      "learning_rate": 0.0009255222135765511,
      "loss": 2.9616,
      "step": 438
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.539691726375911,
      "learning_rate": 0.0009251315758044032,
      "loss": 2.9118,
      "step": 439
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5233911902091123,
      "learning_rate": 0.0009247399991458255,
      "loss": 2.8851,
      "step": 440
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.49625909044493643,
      "learning_rate": 0.0009243474844655996,
      "loss": 2.9513,
      "step": 441
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5674325585191178,
      "learning_rate": 0.0009239540326305791,
      "loss": 2.9316,
      "step": 442
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5620663517264144,
      "learning_rate": 0.0009235596445096864,
      "loss": 3.0614,
      "step": 443
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.4770836891464476,
      "learning_rate": 0.0009231643209739127,
      "loss": 2.9604,
      "step": 444
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5883505727472255,
      "learning_rate": 0.0009227680628963145,
      "loss": 2.8912,
      "step": 445
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.4960038185401185,
      "learning_rate": 0.000922370871152012,
      "loss": 2.9507,
      "step": 446
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.4875655421509081,
      "learning_rate": 0.0009219727466181877,
      "loss": 2.9655,
      "step": 447
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5346241963009257,
      "learning_rate": 0.0009215736901740841,
      "loss": 2.9841,
      "step": 448
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5570461562419412,
      "learning_rate": 0.0009211737027010016,
      "loss": 2.9582,
      "step": 449
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5428447838624275,
      "learning_rate": 0.0009207727850822971,
      "loss": 2.9631,
      "step": 450
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5153435768317739,
      "learning_rate": 0.0009203709382033814,
      "loss": 2.9474,
      "step": 451
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5616424856636266,
      "learning_rate": 0.0009199681629517173,
      "loss": 2.9879,
      "step": 452
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5561590624389096,
      "learning_rate": 0.0009195644602168184,
      "loss": 2.8772,
      "step": 453
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5261545733785727,
      "learning_rate": 0.0009191598308902464,
      "loss": 2.9756,
      "step": 454
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.503695859923239,
      "learning_rate": 0.0009187542758656091,
      "loss": 2.9651,
      "step": 455
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.4611058162423113,
      "learning_rate": 0.0009183477960385591,
      "loss": 2.9308,
      "step": 456
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5110891052157471,
      "learning_rate": 0.0009179403923067912,
      "loss": 2.9121,
      "step": 457
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5572457689919895,
      "learning_rate": 0.0009175320655700406,
      "loss": 3.0012,
      "step": 458
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.56575456099697,
      "learning_rate": 0.0009171228167300805,
      "loss": 2.9347,
      "step": 459
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.6737909709752703,
      "learning_rate": 0.0009167126466907215,
      "loss": 2.9121,
      "step": 460
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.48306566069640283,
      "learning_rate": 0.0009163015563578074,
      "loss": 2.8241,
      "step": 461
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.551315896441015,
      "learning_rate": 0.0009158895466392158,
      "loss": 2.9989,
      "step": 462
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5239093149442633,
      "learning_rate": 0.0009154766184448535,
      "loss": 2.9659,
      "step": 463
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5679546114613675,
      "learning_rate": 0.0009150627726866568,
      "loss": 2.964,
      "step": 464
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5776425199961028,
      "learning_rate": 0.000914648010278587,
      "loss": 2.9475,
      "step": 465
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5897599367569054,
      "learning_rate": 0.0009142323321366315,
      "loss": 2.8971,
      "step": 466
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5177650589375873,
      "learning_rate": 0.0009138157391787986,
      "loss": 2.9064,
      "step": 467
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.4374348835989601,
      "learning_rate": 0.0009133982323251177,
      "loss": 2.9857,
      "step": 468
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.6088452751788686,
      "learning_rate": 0.0009129798124976365,
      "loss": 2.9426,
      "step": 469
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5926389954790133,
      "learning_rate": 0.0009125604806204187,
      "loss": 2.8814,
      "step": 470
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.4672703058624501,
      "learning_rate": 0.0009121402376195421,
      "loss": 2.855,
      "step": 471
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5610911037218453,
      "learning_rate": 0.0009117190844230972,
      "loss": 2.999,
      "step": 472
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5097476947768206,
      "learning_rate": 0.0009112970219611841,
      "loss": 2.9025,
      "step": 473
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5063098850604952,
      "learning_rate": 0.0009108740511659115,
      "loss": 2.9227,
      "step": 474
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5370771029988459,
      "learning_rate": 0.0009104501729713935,
      "loss": 2.941,
      "step": 475
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5959185849794054,
      "learning_rate": 0.0009100253883137488,
      "loss": 2.994,
      "step": 476
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.4937170152698863,
      "learning_rate": 0.0009095996981310974,
      "loss": 2.9104,
      "step": 477
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.47650997879531587,
      "learning_rate": 0.0009091731033635596,
      "loss": 2.9115,
      "step": 478
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.44011750443479547,
      "learning_rate": 0.0009087456049532529,
      "loss": 2.9221,
      "step": 479
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5322455264120257,
      "learning_rate": 0.0009083172038442914,
      "loss": 2.9359,
      "step": 480
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5183598599185936,
      "learning_rate": 0.0009078879009827817,
      "loss": 2.9428,
      "step": 481
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.4618647837081506,
      "learning_rate": 0.0009074576973168223,
      "loss": 2.9797,
      "step": 482
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5147828479886168,
      "learning_rate": 0.0009070265937965015,
      "loss": 2.9793,
      "step": 483
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.4684521134489715,
      "learning_rate": 0.0009065945913738942,
      "loss": 2.9277,
      "step": 484
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5429982829901041,
      "learning_rate": 0.0009061616910030609,
      "loss": 2.9257,
      "step": 485
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.45291350473876096,
      "learning_rate": 0.0009057278936400453,
      "loss": 2.8675,
      "step": 486
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.49644894337956574,
      "learning_rate": 0.0009052932002428715,
      "loss": 2.9112,
      "step": 487
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.4798836530615159,
      "learning_rate": 0.0009048576117715435,
      "loss": 2.9464,
      "step": 488
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.4981773949039495,
      "learning_rate": 0.0009044211291880407,
      "loss": 2.9835,
      "step": 489
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5600084614536893,
      "learning_rate": 0.000903983753456318,
      "loss": 2.9778,
      "step": 490
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5101647323119486,
      "learning_rate": 0.0009035454855423026,
      "loss": 2.8529,
      "step": 491
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5780686512312957,
      "learning_rate": 0.0009031063264138922,
      "loss": 2.9457,
      "step": 492
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4913676655656347,
      "learning_rate": 0.0009026662770409522,
      "loss": 2.9559,
      "step": 493
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4990109684403697,
      "learning_rate": 0.0009022253383953147,
      "loss": 2.9019,
      "step": 494
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5877748599338133,
      "learning_rate": 0.0009017835114507753,
      "loss": 2.9429,
      "step": 495
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5177379832570983,
      "learning_rate": 0.0009013407971830914,
      "loss": 2.985,
      "step": 496
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5932959048307527,
      "learning_rate": 0.0009008971965699801,
      "loss": 2.8916,
      "step": 497
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.54932480626174,
      "learning_rate": 0.0009004527105911163,
      "loss": 2.9476,
      "step": 498
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4845103221930796,
      "learning_rate": 0.0009000073402281295,
      "loss": 2.8539,
      "step": 499
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5472696408301263,
      "learning_rate": 0.0008995610864646028,
      "loss": 2.9767,
      "step": 500
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.515946712565081,
      "learning_rate": 0.0008991139502860703,
      "loss": 2.9315,
      "step": 501
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.46194685756295556,
      "learning_rate": 0.0008986659326800146,
      "loss": 2.9608,
      "step": 502
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4560765574370238,
      "learning_rate": 0.0008982170346358651,
      "loss": 2.8346,
      "step": 503
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5391918532572942,
      "learning_rate": 0.0008977672571449956,
      "loss": 2.9439,
      "step": 504
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4814760570041407,
      "learning_rate": 0.0008973166012007217,
      "loss": 2.827,
      "step": 505
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5629457114946768,
      "learning_rate": 0.0008968650677982998,
      "loss": 2.9719,
      "step": 506
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5593845257197965,
      "learning_rate": 0.0008964126579349236,
      "loss": 2.8864,
      "step": 507
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.46519104813690254,
      "learning_rate": 0.0008959593726097226,
      "loss": 2.8429,
      "step": 508
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4840023861659385,
      "learning_rate": 0.0008955052128237596,
      "loss": 2.9331,
      "step": 509
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5558219699545366,
      "learning_rate": 0.0008950501795800288,
      "loss": 2.9016,
      "step": 510
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4857175065946673,
      "learning_rate": 0.0008945942738834532,
      "loss": 2.9583,
      "step": 511
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.46442689539646187,
      "learning_rate": 0.0008941374967408826,
      "loss": 2.9313,
      "step": 512
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4920680577558812,
      "learning_rate": 0.0008936798491610916,
      "loss": 2.9322,
      "step": 513
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.42071657053145184,
      "learning_rate": 0.0008932213321547768,
      "loss": 2.8092,
      "step": 514
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5423075015977806,
      "learning_rate": 0.0008927619467345554,
      "loss": 2.9191,
      "step": 515
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.46214326428447566,
      "learning_rate": 0.0008923016939149615,
      "loss": 2.9774,
      "step": 516
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5368441064573027,
      "learning_rate": 0.0008918405747124458,
      "loss": 2.9746,
      "step": 517
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.47725237881777255,
      "learning_rate": 0.0008913785901453721,
      "loss": 2.8924,
      "step": 518
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4587232357368762,
      "learning_rate": 0.000890915741234015,
      "loss": 2.9563,
      "step": 519
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5025733386528658,
      "learning_rate": 0.0008904520290005582,
      "loss": 2.8566,
      "step": 520
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5062215102612683,
      "learning_rate": 0.000889987454469092,
      "loss": 2.9385,
      "step": 521
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.46271562433176705,
      "learning_rate": 0.0008895220186656111,
      "loss": 2.8672,
      "step": 522
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4447670480633681,
      "learning_rate": 0.0008890557226180122,
      "loss": 2.9219,
      "step": 523
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5131916806570509,
      "learning_rate": 0.0008885885673560921,
      "loss": 2.8686,
      "step": 524
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.472971672961692,
      "learning_rate": 0.0008881205539115444,
      "loss": 2.834,
      "step": 525
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5203357089744916,
      "learning_rate": 0.0008876516833179589,
      "loss": 2.8396,
      "step": 526
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5795193057589354,
      "learning_rate": 0.0008871819566108177,
      "loss": 2.8938,
      "step": 527
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4828578545983392,
      "learning_rate": 0.000886711374827494,
      "loss": 2.8184,
      "step": 528
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5005724587827719,
      "learning_rate": 0.0008862399390072491,
      "loss": 2.9117,
      "step": 529
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.543022044362364,
      "learning_rate": 0.0008857676501912305,
      "loss": 2.7959,
      "step": 530
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4921165011571337,
      "learning_rate": 0.0008852945094224697,
      "loss": 2.9149,
      "step": 531
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.46669093263366496,
      "learning_rate": 0.0008848205177458795,
      "loss": 2.907,
      "step": 532
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5449088043083928,
      "learning_rate": 0.0008843456762082518,
      "loss": 2.9024,
      "step": 533
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4735115730561581,
      "learning_rate": 0.0008838699858582557,
      "loss": 2.9188,
      "step": 534
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.499556618409815,
      "learning_rate": 0.0008833934477464347,
      "loss": 2.9121,
      "step": 535
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5592446376142547,
      "learning_rate": 0.0008829160629252045,
      "loss": 2.8401,
      "step": 536
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5053968364268197,
      "learning_rate": 0.0008824378324488509,
      "loss": 2.9381,
      "step": 537
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5803014676195123,
      "learning_rate": 0.0008819587573735268,
      "loss": 2.9658,
      "step": 538
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5655567818068991,
      "learning_rate": 0.0008814788387572513,
      "loss": 2.9185,
      "step": 539
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5394021744156271,
      "learning_rate": 0.0008809980776599053,
      "loss": 2.8238,
      "step": 540
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5006668242505091,
      "learning_rate": 0.0008805164751432312,
      "loss": 2.9115,
      "step": 541
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.549900650742081,
      "learning_rate": 0.0008800340322708292,
      "loss": 2.878,
      "step": 542
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.4434311475917405,
      "learning_rate": 0.0008795507501081555,
      "loss": 2.929,
      "step": 543
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5049435314266008,
      "learning_rate": 0.0008790666297225196,
      "loss": 2.8983,
      "step": 544
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5258503290310728,
      "learning_rate": 0.0008785816721830829,
      "loss": 2.8395,
      "step": 545
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5226439862806438,
      "learning_rate": 0.0008780958785608546,
      "loss": 2.9605,
      "step": 546
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.47502288491554423,
      "learning_rate": 0.0008776092499286912,
      "loss": 2.9058,
      "step": 547
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.49777311686968306,
      "learning_rate": 0.0008771217873612929,
      "loss": 2.8013,
      "step": 548
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5040476517317467,
      "learning_rate": 0.0008766334919352017,
      "loss": 2.8891,
      "step": 549
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5734027619323142,
      "learning_rate": 0.0008761443647287987,
      "loss": 2.8978,
      "step": 550
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5591583484093245,
      "learning_rate": 0.0008756544068223026,
      "loss": 2.9268,
      "step": 551
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5590005958740321,
      "learning_rate": 0.0008751636192977659,
      "loss": 2.9581,
      "step": 552
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.4989723022744546,
      "learning_rate": 0.0008746720032390737,
      "loss": 2.8166,
      "step": 553
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.4524663721859508,
      "learning_rate": 0.0008741795597319408,
      "loss": 2.876,
      "step": 554
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5340577235221757,
      "learning_rate": 0.0008736862898639095,
      "loss": 2.8726,
      "step": 555
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.46073229195296406,
      "learning_rate": 0.0008731921947243468,
      "loss": 2.8697,
      "step": 556
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5008000922809575,
      "learning_rate": 0.0008726972754044427,
      "loss": 2.9267,
      "step": 557
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.4447115743712347,
      "learning_rate": 0.0008722015329972069,
      "loss": 2.8616,
      "step": 558
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.45839536662578106,
      "learning_rate": 0.0008717049685974672,
      "loss": 2.8847,
      "step": 559
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5138574329331402,
      "learning_rate": 0.0008712075833018665,
      "loss": 2.927,
      "step": 560
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.4617875992822275,
      "learning_rate": 0.0008707093782088608,
      "loss": 2.8355,
      "step": 561
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.4768214484564255,
      "learning_rate": 0.0008702103544187167,
      "loss": 2.9338,
      "step": 562
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5092198530538106,
      "learning_rate": 0.0008697105130335085,
      "loss": 2.8922,
      "step": 563
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.4909867955599156,
      "learning_rate": 0.0008692098551571164,
      "loss": 2.8805,
      "step": 564
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.48400549737164544,
      "learning_rate": 0.0008687083818952235,
      "loss": 2.844,
      "step": 565
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5312485094846079,
      "learning_rate": 0.0008682060943553143,
      "loss": 2.8628,
      "step": 566
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5284169290059281,
      "learning_rate": 0.0008677029936466707,
      "loss": 2.8693,
      "step": 567
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5596873208407769,
      "learning_rate": 0.0008671990808803711,
      "loss": 2.8628,
      "step": 568
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5127726821195686,
      "learning_rate": 0.0008666943571692871,
      "loss": 2.8931,
      "step": 569
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5412302489200196,
      "learning_rate": 0.0008661888236280813,
      "loss": 2.8495,
      "step": 570
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5648066434836707,
      "learning_rate": 0.0008656824813732045,
      "loss": 2.8212,
      "step": 571
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.530080954397923,
      "learning_rate": 0.000865175331522894,
      "loss": 2.9589,
      "step": 572
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.49007327673633844,
      "learning_rate": 0.0008646673751971703,
      "loss": 2.9213,
      "step": 573
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.47956842689479867,
      "learning_rate": 0.000864158613517835,
      "loss": 2.8258,
      "step": 574
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.563080037087868,
      "learning_rate": 0.0008636490476084681,
      "loss": 2.8938,
      "step": 575
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.48614569752706366,
      "learning_rate": 0.0008631386785944264,
      "loss": 2.8999,
      "step": 576
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.47949260744852895,
      "learning_rate": 0.0008626275076028397,
      "loss": 2.9086,
      "step": 577
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5452426340004343,
      "learning_rate": 0.0008621155357626091,
      "loss": 2.7912,
      "step": 578
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5467956749933328,
      "learning_rate": 0.0008616027642044042,
      "loss": 2.8611,
      "step": 579
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5134651071394092,
      "learning_rate": 0.000861089194060661,
      "loss": 2.8905,
      "step": 580
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.472677589381949,
      "learning_rate": 0.000860574826465579,
      "loss": 2.7825,
      "step": 581
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.48018765852806505,
      "learning_rate": 0.0008600596625551191,
      "loss": 2.8662,
      "step": 582
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5128764288735578,
      "learning_rate": 0.0008595437034670006,
      "loss": 2.8177,
      "step": 583
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.4966513345718583,
      "learning_rate": 0.0008590269503406985,
      "loss": 2.9176,
      "step": 584
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5082739810555716,
      "learning_rate": 0.0008585094043174423,
      "loss": 2.8508,
      "step": 585
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.48652094508387944,
      "learning_rate": 0.0008579910665402118,
      "loss": 2.8479,
      "step": 586
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.48352095303386156,
      "learning_rate": 0.000857471938153736,
      "loss": 2.8137,
      "step": 587
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.46689576961775836,
      "learning_rate": 0.0008569520203044892,
      "loss": 2.8206,
      "step": 588
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5249930996254872,
      "learning_rate": 0.0008564313141406901,
      "loss": 2.8354,
      "step": 589
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5101748353606734,
      "learning_rate": 0.0008559098208122973,
      "loss": 2.8479,
      "step": 590
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.4737361503886194,
      "learning_rate": 0.0008553875414710089,
      "loss": 2.8718,
      "step": 591
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.4695780598746661,
      "learning_rate": 0.0008548644772702579,
      "loss": 2.834,
      "step": 592
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5442563661083804,
      "learning_rate": 0.0008543406293652116,
      "loss": 2.8823,
      "step": 593
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.4769846167337409,
      "learning_rate": 0.0008538159989127671,
      "loss": 2.8325,
      "step": 594
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.48928408239321125,
      "learning_rate": 0.0008532905870715505,
      "loss": 2.9471,
      "step": 595
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5755923681979764,
      "learning_rate": 0.0008527643950019131,
      "loss": 2.9084,
      "step": 596
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5150173270935033,
      "learning_rate": 0.0008522374238659296,
      "loss": 2.9108,
      "step": 597
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.49089687158658574,
      "learning_rate": 0.0008517096748273951,
      "loss": 2.8406,
      "step": 598
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5384211751976695,
      "learning_rate": 0.0008511811490518227,
      "loss": 2.93,
      "step": 599
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.4549451269314346,
      "learning_rate": 0.0008506518477064405,
      "loss": 2.7794,
      "step": 600
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.4935542960309496,
      "learning_rate": 0.0008501217719601903,
      "loss": 2.8384,
      "step": 601
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.47367721409710756,
      "learning_rate": 0.0008495909229837233,
      "loss": 2.8738,
      "step": 602
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.4822635411013927,
      "learning_rate": 0.000849059301949399,
      "loss": 2.8853,
      "step": 603
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.5211203299307834,
      "learning_rate": 0.0008485269100312812,
      "loss": 2.8763,
      "step": 604
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.5093788703853647,
      "learning_rate": 0.0008479937484051368,
      "loss": 2.9627,
      "step": 605
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.49379817177711127,
      "learning_rate": 0.0008474598182484323,
      "loss": 2.9097,
      "step": 606
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.5055896442532648,
      "learning_rate": 0.0008469251207403317,
      "loss": 2.8566,
      "step": 607
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.48998652758761935,
      "learning_rate": 0.0008463896570616934,
      "loss": 2.8092,
      "step": 608
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.4643446399451955,
      "learning_rate": 0.0008458534283950678,
      "loss": 2.9923,
      "step": 609
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.47750788765136326,
      "learning_rate": 0.0008453164359246952,
      "loss": 2.8214,
      "step": 610
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.5387199911323801,
      "learning_rate": 0.0008447786808365022,
      "loss": 2.8333,
      "step": 611
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.537020119273398,
      "learning_rate": 0.0008442401643181,
      "loss": 2.8287,
      "step": 612
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.45996197632489566,
      "learning_rate": 0.0008437008875587811,
      "loss": 2.8405,
      "step": 613
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.46585158868278714,
      "learning_rate": 0.0008431608517495171,
      "loss": 2.8358,
      "step": 614
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.5324745662763111,
      "learning_rate": 0.0008426200580829561,
      "loss": 2.9404,
      "step": 615
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.4897854216836938,
      "learning_rate": 0.0008420785077534195,
      "loss": 2.7592,
      "step": 616
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.5143061610832548,
      "learning_rate": 0.0008415362019569001,
      "loss": 2.8299,
      "step": 617
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.4901690572264902,
      "learning_rate": 0.0008409931418910591,
      "loss": 2.782,
      "step": 618
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.5710877923272081,
      "learning_rate": 0.0008404493287552232,
      "loss": 2.9038,
      "step": 619
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.5387796073193574,
      "learning_rate": 0.0008399047637503825,
      "loss": 2.8469,
      "step": 620
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.46356672629268736,
      "learning_rate": 0.0008393594480791875,
      "loss": 2.8451,
      "step": 621
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.4857247723265462,
      "learning_rate": 0.0008388133829459463,
      "loss": 2.8477,
      "step": 622
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5028032907022912,
      "learning_rate": 0.0008382665695566227,
      "loss": 2.8071,
      "step": 623
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5352764699826724,
      "learning_rate": 0.0008377190091188324,
      "loss": 2.8124,
      "step": 624
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5154671314728527,
      "learning_rate": 0.0008371707028418413,
      "loss": 2.8065,
      "step": 625
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5525779227319001,
      "learning_rate": 0.0008366216519365621,
      "loss": 2.81,
      "step": 626
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.4999272063558071,
      "learning_rate": 0.0008360718576155525,
      "loss": 2.9294,
      "step": 627
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.538437972979967,
      "learning_rate": 0.0008355213210930118,
      "loss": 2.802,
      "step": 628
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.4779442710284314,
      "learning_rate": 0.0008349700435847778,
      "loss": 2.9181,
      "step": 629
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5112276129707731,
      "learning_rate": 0.0008344180263083256,
      "loss": 2.8876,
      "step": 630
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5121208689408047,
      "learning_rate": 0.000833865270482764,
      "loss": 2.9277,
      "step": 631
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5073415586581523,
      "learning_rate": 0.0008333117773288324,
      "loss": 2.9644,
      "step": 632
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.516536914907916,
      "learning_rate": 0.0008327575480688985,
      "loss": 2.907,
      "step": 633
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5365473704448307,
      "learning_rate": 0.000832202583926956,
      "loss": 2.8206,
      "step": 634
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5366448186943982,
      "learning_rate": 0.0008316468861286217,
      "loss": 2.83,
      "step": 635
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5228990718651447,
      "learning_rate": 0.0008310904559011323,
      "loss": 2.8893,
      "step": 636
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.46544827337040146,
      "learning_rate": 0.0008305332944733419,
      "loss": 2.7929,
      "step": 637
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5332647947062249,
      "learning_rate": 0.0008299754030757202,
      "loss": 2.8962,
      "step": 638
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5373153750438453,
      "learning_rate": 0.0008294167829403481,
      "loss": 2.7954,
      "step": 639
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.513076059645814,
      "learning_rate": 0.0008288574353009164,
      "loss": 2.8701,
      "step": 640
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.47984399848374937,
      "learning_rate": 0.0008282973613927225,
      "loss": 2.7395,
      "step": 641
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.4847363450816728,
      "learning_rate": 0.0008277365624526675,
      "loss": 2.8574,
      "step": 642
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.4910899907700678,
      "learning_rate": 0.0008271750397192541,
      "loss": 2.8263,
      "step": 643
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.5443766929577677,
      "learning_rate": 0.0008266127944325832,
      "loss": 2.8328,
      "step": 644
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.5039112413589106,
      "learning_rate": 0.0008260498278343513,
      "loss": 2.915,
      "step": 645
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.45117555991793873,
      "learning_rate": 0.0008254861411678485,
      "loss": 2.802,
      "step": 646
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.4620612469446592,
      "learning_rate": 0.0008249217356779544,
      "loss": 2.7939,
      "step": 647
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.5213350459237406,
      "learning_rate": 0.0008243566126111363,
      "loss": 2.8363,
      "step": 648
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.46127740901909686,
      "learning_rate": 0.0008237907732154466,
      "loss": 2.776,
      "step": 649
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.5195423408551074,
      "learning_rate": 0.0008232242187405194,
      "loss": 2.8347,
      "step": 650
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.4597719384843304,
      "learning_rate": 0.000822656950437568,
      "loss": 2.7871,
      "step": 651
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.4744744191905495,
      "learning_rate": 0.0008220889695593823,
      "loss": 2.8932,
      "step": 652
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.4592569922712311,
      "learning_rate": 0.0008215202773603259,
      "loss": 2.8186,
      "step": 653
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.4888379498693485,
      "learning_rate": 0.0008209508750963328,
      "loss": 2.8816,
      "step": 654
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.4608427644037051,
      "learning_rate": 0.0008203807640249062,
      "loss": 2.8588,
      "step": 655
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.5427147262691617,
      "learning_rate": 0.0008198099454051136,
      "loss": 2.8434,
      "step": 656
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.5064096594473994,
      "learning_rate": 0.0008192384204975857,
      "loss": 2.8643,
      "step": 657
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.4837547195674473,
      "learning_rate": 0.000818666190564513,
      "loss": 2.8047,
      "step": 658
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.44277891757582555,
      "learning_rate": 0.0008180932568696426,
      "loss": 2.852,
      "step": 659
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.52358325472259,
      "learning_rate": 0.0008175196206782764,
      "loss": 2.8352,
      "step": 660
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.4997652840704003,
      "learning_rate": 0.0008169452832572675,
      "loss": 2.8574,
      "step": 661
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.46535861244240406,
      "learning_rate": 0.0008163702458750173,
      "loss": 2.7891,
      "step": 662
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.5844228916475839,
      "learning_rate": 0.0008157945098014734,
      "loss": 2.8538,
      "step": 663
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.5340395984470981,
      "learning_rate": 0.0008152180763081267,
      "loss": 2.8357,
      "step": 664
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.4773983930271618,
      "learning_rate": 0.0008146409466680076,
      "loss": 2.8222,
      "step": 665
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5307602408186971,
      "learning_rate": 0.0008140631221556845,
      "loss": 2.8009,
      "step": 666
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.43860393901215583,
      "learning_rate": 0.0008134846040472599,
      "loss": 2.8659,
      "step": 667
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5107311724742075,
      "learning_rate": 0.0008129053936203688,
      "loss": 2.9087,
      "step": 668
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5191816808666726,
      "learning_rate": 0.0008123254921541745,
      "loss": 2.8502,
      "step": 669
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.51020382429763,
      "learning_rate": 0.0008117449009293668,
      "loss": 2.8447,
      "step": 670
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.4703843692795738,
      "learning_rate": 0.0008111636212281586,
      "loss": 2.8351,
      "step": 671
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.46621448944448807,
      "learning_rate": 0.0008105816543342833,
      "loss": 2.7883,
      "step": 672
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.4301977863025668,
      "learning_rate": 0.0008099990015329919,
      "loss": 2.8082,
      "step": 673
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5173946426936311,
      "learning_rate": 0.0008094156641110504,
      "loss": 2.7677,
      "step": 674
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5981599780929721,
      "learning_rate": 0.0008088316433567369,
      "loss": 2.8339,
      "step": 675
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.6234006489255799,
      "learning_rate": 0.0008082469405598378,
      "loss": 2.9027,
      "step": 676
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5224969823883701,
      "learning_rate": 0.0008076615570116468,
      "loss": 2.8669,
      "step": 677
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.6091701591643708,
      "learning_rate": 0.0008070754940049603,
      "loss": 2.7997,
      "step": 678
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5965877845222998,
      "learning_rate": 0.0008064887528340756,
      "loss": 2.878,
      "step": 679
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.4939727269291862,
      "learning_rate": 0.0008059013347947874,
      "loss": 2.7794,
      "step": 680
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.532980097448278,
      "learning_rate": 0.0008053132411843857,
      "loss": 2.7978,
      "step": 681
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.556732847287261,
      "learning_rate": 0.0008047244733016521,
      "loss": 2.8827,
      "step": 682
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5175738146243685,
      "learning_rate": 0.0008041350324468573,
      "loss": 2.8155,
      "step": 683
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5526779343389316,
      "learning_rate": 0.0008035449199217583,
      "loss": 2.7721,
      "step": 684
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5096027320085115,
      "learning_rate": 0.0008029541370295957,
      "loss": 2.877,
      "step": 685
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5215402888330627,
      "learning_rate": 0.0008023626850750903,
      "loss": 2.8386,
      "step": 686
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5357029520261977,
      "learning_rate": 0.0008017705653644406,
      "loss": 2.8333,
      "step": 687
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5481223895987873,
      "learning_rate": 0.0008011777792053195,
      "loss": 2.821,
      "step": 688
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.541773595807566,
      "learning_rate": 0.0008005843279068725,
      "loss": 2.7868,
      "step": 689
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5333916070138732,
      "learning_rate": 0.000799990212779713,
      "loss": 2.76,
      "step": 690
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.6896986380917894,
      "learning_rate": 0.0007993954351359214,
      "loss": 2.8998,
      "step": 691
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5138560292942024,
      "learning_rate": 0.0007987999962890406,
      "loss": 2.7831,
      "step": 692
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5006192365144864,
      "learning_rate": 0.0007982038975540742,
      "loss": 2.8001,
      "step": 693
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5706754312956506,
      "learning_rate": 0.0007976071402474826,
      "loss": 2.9346,
      "step": 694
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5819806935647719,
      "learning_rate": 0.0007970097256871811,
      "loss": 2.8146,
      "step": 695
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5298625281963546,
      "learning_rate": 0.0007964116551925364,
      "loss": 2.8184,
      "step": 696
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.4808085412662244,
      "learning_rate": 0.0007958129300843637,
      "loss": 2.7149,
      "step": 697
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5279065705275813,
      "learning_rate": 0.0007952135516849239,
      "loss": 2.818,
      "step": 698
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.49558916945973513,
      "learning_rate": 0.0007946135213179207,
      "loss": 2.784,
      "step": 699
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5285076010156654,
      "learning_rate": 0.0007940128403084977,
      "loss": 2.8013,
      "step": 700
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.49178833068753275,
      "learning_rate": 0.0007934115099832355,
      "loss": 2.766,
      "step": 701
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.45489547648663475,
      "learning_rate": 0.0007928095316701483,
      "loss": 2.7464,
      "step": 702
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5672774198786119,
      "learning_rate": 0.0007922069066986819,
      "loss": 2.8402,
      "step": 703
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5559238096112596,
      "learning_rate": 0.0007916036363997097,
      "loss": 2.8086,
      "step": 704
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.49014582917174837,
      "learning_rate": 0.0007909997221055308,
      "loss": 2.8316,
      "step": 705
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.44636188694556145,
      "learning_rate": 0.0007903951651498658,
      "loss": 2.7963,
      "step": 706
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.47986388320100326,
      "learning_rate": 0.0007897899668678557,
      "loss": 2.8437,
      "step": 707
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.4959767047311583,
      "learning_rate": 0.0007891841285960566,
      "loss": 2.819,
      "step": 708
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.49856630474601016,
      "learning_rate": 0.0007885776516724388,
      "loss": 2.7593,
      "step": 709
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.46537743342048815,
      "learning_rate": 0.0007879705374363831,
      "loss": 2.8627,
      "step": 710
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.46719387680375635,
      "learning_rate": 0.000787362787228677,
      "loss": 2.8295,
      "step": 711
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.4771797067030745,
      "learning_rate": 0.0007867544023915134,
      "loss": 2.7843,
      "step": 712
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5077335914469979,
      "learning_rate": 0.0007861453842684861,
      "loss": 2.837,
      "step": 713
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.4859065777537034,
      "learning_rate": 0.0007855357342045882,
      "loss": 2.9189,
      "step": 714
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.47599378592576,
      "learning_rate": 0.0007849254535462074,
      "loss": 2.729,
      "step": 715
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.4792105033293628,
      "learning_rate": 0.0007843145436411252,
      "loss": 2.7473,
      "step": 716
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5094603109719453,
      "learning_rate": 0.0007837030058385117,
      "loss": 2.7893,
      "step": 717
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5159291742558351,
      "learning_rate": 0.0007830908414889246,
      "loss": 2.8468,
      "step": 718
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5226777221205205,
      "learning_rate": 0.0007824780519443046,
      "loss": 2.8741,
      "step": 719
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.48165745332515575,
      "learning_rate": 0.0007818646385579735,
      "loss": 2.8219,
      "step": 720
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.492731216786435,
      "learning_rate": 0.0007812506026846307,
      "loss": 2.8265,
      "step": 721
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5212723481310344,
      "learning_rate": 0.0007806359456803504,
      "loss": 2.8288,
      "step": 722
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5168373653618235,
      "learning_rate": 0.0007800206689025785,
      "loss": 2.8805,
      "step": 723
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.50933421230669,
      "learning_rate": 0.0007794047737101297,
      "loss": 2.7754,
      "step": 724
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.516459366923144,
      "learning_rate": 0.0007787882614631843,
      "loss": 2.8378,
      "step": 725
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.4814820725937623,
      "learning_rate": 0.0007781711335232856,
      "loss": 2.7943,
      "step": 726
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.4875185583230406,
      "learning_rate": 0.0007775533912533363,
      "loss": 2.7584,
      "step": 727
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.48607259538361763,
      "learning_rate": 0.0007769350360175962,
      "loss": 2.7943,
      "step": 728
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5750716834562372,
      "learning_rate": 0.0007763160691816784,
      "loss": 2.7651,
      "step": 729
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.48445857712796325,
      "learning_rate": 0.000775696492112547,
      "loss": 2.8346,
      "step": 730
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.4924724269823932,
      "learning_rate": 0.0007750763061785137,
      "loss": 2.6975,
      "step": 731
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5184944074242422,
      "learning_rate": 0.000774455512749235,
      "loss": 2.7624,
      "step": 732
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.6235652530453863,
      "learning_rate": 0.0007738341131957085,
      "loss": 2.7976,
      "step": 733
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.46422224574631493,
      "learning_rate": 0.000773212108890271,
      "loss": 2.785,
      "step": 734
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5318321963574243,
      "learning_rate": 0.0007725895012065947,
      "loss": 2.812,
      "step": 735
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5907337544025071,
      "learning_rate": 0.0007719662915196844,
      "loss": 2.7859,
      "step": 736
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5053319064383238,
      "learning_rate": 0.0007713424812058736,
      "loss": 2.8105,
      "step": 737
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5016101292225251,
      "learning_rate": 0.0007707180716428237,
      "loss": 2.8497,
      "step": 738
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5635667122980256,
      "learning_rate": 0.0007700930642095184,
      "loss": 2.8426,
      "step": 739
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5347229459146409,
      "learning_rate": 0.0007694674602862621,
      "loss": 2.7543,
      "step": 740
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.49840886650055877,
      "learning_rate": 0.0007688412612546769,
      "loss": 2.809,
      "step": 741
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.49061312191402545,
      "learning_rate": 0.0007682144684976983,
      "loss": 2.7986,
      "step": 742
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.4796430732364091,
      "learning_rate": 0.0007675870833995739,
      "loss": 2.7881,
      "step": 743
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5085767803752446,
      "learning_rate": 0.0007669591073458592,
      "loss": 2.8191,
      "step": 744
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.4960853465685693,
      "learning_rate": 0.0007663305417234146,
      "loss": 2.7734,
      "step": 745
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.4725389933963003,
      "learning_rate": 0.0007657013879204022,
      "loss": 2.7513,
      "step": 746
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5086340202810876,
      "learning_rate": 0.0007650716473262842,
      "loss": 2.8695,
      "step": 747
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.4913173591153497,
      "learning_rate": 0.0007644413213318177,
      "loss": 2.7421,
      "step": 748
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5117193247039085,
      "learning_rate": 0.0007638104113290531,
      "loss": 2.9339,
      "step": 749
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5502113295847995,
      "learning_rate": 0.0007631789187113303,
      "loss": 2.8683,
      "step": 750
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.4831331978157491,
      "learning_rate": 0.000762546844873276,
      "loss": 2.7499,
      "step": 751
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.468882911032173,
      "learning_rate": 0.0007619141912108007,
      "loss": 2.7906,
      "step": 752
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.4899884078767972,
      "learning_rate": 0.000761280959121095,
      "loss": 2.8205,
      "step": 753
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.570115786694176,
      "learning_rate": 0.0007606471500026273,
      "loss": 2.893,
      "step": 754
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5258819361717191,
      "learning_rate": 0.0007600127652551401,
      "loss": 2.8404,
      "step": 755
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5330489775108016,
      "learning_rate": 0.0007593778062796472,
      "loss": 2.8236,
      "step": 756
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5157248399249127,
      "learning_rate": 0.000758742274478431,
      "loss": 2.774,
      "step": 757
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5030996059317102,
      "learning_rate": 0.0007581061712550381,
      "loss": 2.7986,
      "step": 758
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5510672987815763,
      "learning_rate": 0.0007574694980142779,
      "loss": 2.8002,
      "step": 759
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.4892304524594696,
      "learning_rate": 0.0007568322561622183,
      "loss": 2.7387,
      "step": 760
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.43620102638706626,
      "learning_rate": 0.0007561944471061826,
      "loss": 2.8067,
      "step": 761
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5249551128598866,
      "learning_rate": 0.0007555560722547475,
      "loss": 2.8052,
      "step": 762
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5314704328210782,
      "learning_rate": 0.0007549171330177387,
      "loss": 2.7803,
      "step": 763
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5083259243551354,
      "learning_rate": 0.0007542776308062285,
      "loss": 2.8439,
      "step": 764
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.49486921119685484,
      "learning_rate": 0.0007536375670325325,
      "loss": 2.8783,
      "step": 765
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.4948632029244516,
      "learning_rate": 0.0007529969431102063,
      "loss": 2.8038,
      "step": 766
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5613854379150421,
      "learning_rate": 0.000752355760454043,
      "loss": 2.7952,
      "step": 767
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.48408993039962017,
      "learning_rate": 0.0007517140204800693,
      "loss": 2.8285,
      "step": 768
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5613325063670752,
      "learning_rate": 0.0007510717246055425,
      "loss": 2.8986,
      "step": 769
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5057024413578123,
      "learning_rate": 0.0007504288742489482,
      "loss": 2.9194,
      "step": 770
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5182327909106127,
      "learning_rate": 0.0007497854708299963,
      "loss": 2.7443,
      "step": 771
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.42623461117590616,
      "learning_rate": 0.0007491415157696178,
      "loss": 2.7663,
      "step": 772
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.4616439913926983,
      "learning_rate": 0.0007484970104899623,
      "loss": 2.758,
      "step": 773
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5216346476601743,
      "learning_rate": 0.0007478519564143945,
      "loss": 2.7734,
      "step": 774
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5263024721605336,
      "learning_rate": 0.000747206354967491,
      "loss": 2.8226,
      "step": 775
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.49211891503665545,
      "learning_rate": 0.0007465602075750373,
      "loss": 2.8352,
      "step": 776
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.4640334355314008,
      "learning_rate": 0.0007459135156640247,
      "loss": 2.8302,
      "step": 777
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.4916355653630351,
      "learning_rate": 0.0007452662806626468,
      "loss": 2.9638,
      "step": 778
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5682294532098234,
      "learning_rate": 0.0007446185040002967,
      "loss": 2.7034,
      "step": 779
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5083832030962198,
      "learning_rate": 0.0007439701871075642,
      "loss": 2.8328,
      "step": 780
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5496820839619448,
      "learning_rate": 0.0007433213314162313,
      "loss": 2.8676,
      "step": 781
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.4942563061066859,
      "learning_rate": 0.0007426719383592705,
      "loss": 2.7837,
      "step": 782
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.4696144837532974,
      "learning_rate": 0.000742022009370841,
      "loss": 2.7719,
      "step": 783
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5210832302973146,
      "learning_rate": 0.0007413715458862855,
      "loss": 2.7627,
      "step": 784
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.4761344198378059,
      "learning_rate": 0.0007407205493421272,
      "loss": 2.806,
      "step": 785
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5221206952994478,
      "learning_rate": 0.0007400690211760661,
      "loss": 2.7719,
      "step": 786
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.48563540014368656,
      "learning_rate": 0.0007394169628269771,
      "loss": 2.7514,
      "step": 787
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.48413737878074503,
      "learning_rate": 0.0007387643757349051,
      "loss": 2.7659,
      "step": 788
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5118877841303338,
      "learning_rate": 0.0007381112613410635,
      "loss": 2.947,
      "step": 789
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5316034920848375,
      "learning_rate": 0.0007374576210878298,
      "loss": 2.8874,
      "step": 790
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.4883297464158908,
      "learning_rate": 0.0007368034564187425,
      "loss": 2.8233,
      "step": 791
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.513939458135119,
      "learning_rate": 0.0007361487687784989,
      "loss": 2.7089,
      "step": 792
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5785424945432845,
      "learning_rate": 0.0007354935596129513,
      "loss": 2.8671,
      "step": 793
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.48353704239395495,
      "learning_rate": 0.000734837830369103,
      "loss": 2.6387,
      "step": 794
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5005110059498641,
      "learning_rate": 0.0007341815824951066,
      "loss": 2.8208,
      "step": 795
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.46936386034848254,
      "learning_rate": 0.0007335248174402597,
      "loss": 2.8644,
      "step": 796
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.46979639946920604,
      "learning_rate": 0.0007328675366550023,
      "loss": 2.8314,
      "step": 797
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5113553778805662,
      "learning_rate": 0.0007322097415909134,
      "loss": 2.8458,
      "step": 798
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.4583682893078192,
      "learning_rate": 0.0007315514337007071,
      "loss": 2.7059,
      "step": 799
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.511531165607695,
      "learning_rate": 0.0007308926144382312,
      "loss": 2.7719,
      "step": 800
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5340364889597687,
      "learning_rate": 0.0007302332852584619,
      "loss": 2.7535,
      "step": 801
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5270216020215914,
      "learning_rate": 0.0007295734476175018,
      "loss": 2.8255,
      "step": 802
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.4695692919781961,
      "learning_rate": 0.0007289131029725768,
      "loss": 2.7913,
      "step": 803
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.48772634714980395,
      "learning_rate": 0.0007282522527820319,
      "loss": 2.7327,
      "step": 804
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.4908532542810394,
      "learning_rate": 0.000727590898505329,
      "loss": 2.759,
      "step": 805
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5392196973957384,
      "learning_rate": 0.0007269290416030429,
      "loss": 2.7883,
      "step": 806
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5229476344661595,
      "learning_rate": 0.000726266683536859,
      "loss": 2.7925,
      "step": 807
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.46506825598686913,
      "learning_rate": 0.0007256038257695687,
      "loss": 2.7616,
      "step": 808
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5105977210525918,
      "learning_rate": 0.0007249404697650678,
      "loss": 2.7546,
      "step": 809
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5221766178000213,
      "learning_rate": 0.0007242766169883518,
      "loss": 2.7515,
      "step": 810
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5520794117426239,
      "learning_rate": 0.0007236122689055138,
      "loss": 2.8096,
      "step": 811
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5449218556525243,
      "learning_rate": 0.0007229474269837401,
      "loss": 2.8004,
      "step": 812
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.526682352936027,
      "learning_rate": 0.0007222820926913085,
      "loss": 2.7215,
      "step": 813
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.580858902393497,
      "learning_rate": 0.0007216162674975833,
      "loss": 2.8152,
      "step": 814
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5564683137969968,
      "learning_rate": 0.0007209499528730138,
      "loss": 2.7984,
      "step": 815
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5172723636230036,
      "learning_rate": 0.0007202831502891294,
      "loss": 2.8307,
      "step": 816
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.44820234752374694,
      "learning_rate": 0.0007196158612185375,
      "loss": 2.6601,
      "step": 817
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5291399878434985,
      "learning_rate": 0.0007189480871349201,
      "loss": 2.7593,
      "step": 818
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.48846075033892566,
      "learning_rate": 0.0007182798295130299,
      "loss": 2.7638,
      "step": 819
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5031815366441419,
      "learning_rate": 0.0007176110898286878,
      "loss": 2.8129,
      "step": 820
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5315462339416552,
      "learning_rate": 0.0007169418695587791,
      "loss": 2.8152,
      "step": 821
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5302833561195777,
      "learning_rate": 0.0007162721701812506,
      "loss": 2.8827,
      "step": 822
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.45982473679298547,
      "learning_rate": 0.0007156019931751072,
      "loss": 2.7817,
      "step": 823
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5074921998828663,
      "learning_rate": 0.0007149313400204082,
      "loss": 2.6997,
      "step": 824
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5159064353389416,
      "learning_rate": 0.0007142602121982653,
      "loss": 2.7705,
      "step": 825
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5180972511567747,
      "learning_rate": 0.0007135886111908379,
      "loss": 2.8564,
      "step": 826
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.49085156382430956,
      "learning_rate": 0.0007129165384813303,
      "loss": 2.802,
      "step": 827
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.4730921860189482,
      "learning_rate": 0.0007122439955539888,
      "loss": 2.7001,
      "step": 828
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5292831592244239,
      "learning_rate": 0.0007115709838940983,
      "loss": 2.7195,
      "step": 829
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5413440021833602,
      "learning_rate": 0.0007108975049879785,
      "loss": 2.8672,
      "step": 830
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.4740637709041407,
      "learning_rate": 0.0007102235603229814,
      "loss": 2.7726,
      "step": 831
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.4788801287316272,
      "learning_rate": 0.000709549151387487,
      "loss": 2.7526,
      "step": 832
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.4772345087746463,
      "learning_rate": 0.0007088742796709013,
      "loss": 2.7133,
      "step": 833
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.47261956559665125,
      "learning_rate": 0.000708198946663652,
      "loss": 2.7813,
      "step": 834
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5610903768467415,
      "learning_rate": 0.0007075231538571856,
      "loss": 2.8754,
      "step": 835
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.48776587880542527,
      "learning_rate": 0.0007068469027439641,
      "loss": 2.8114,
      "step": 836
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.49431864674560283,
      "learning_rate": 0.0007061701948174613,
      "loss": 2.8075,
      "step": 837
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.48851198234487625,
      "learning_rate": 0.0007054930315721606,
      "loss": 2.8488,
      "step": 838
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5604073171311356,
      "learning_rate": 0.0007048154145035501,
      "loss": 2.8551,
      "step": 839
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5194267496909427,
      "learning_rate": 0.0007041373451081207,
      "loss": 2.7335,
      "step": 840
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.4783931437057692,
      "learning_rate": 0.0007034588248833621,
      "loss": 2.8199,
      "step": 841
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.6279616789683025,
      "learning_rate": 0.0007027798553277595,
      "loss": 2.722,
      "step": 842
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5679575025331579,
      "learning_rate": 0.0007021004379407909,
      "loss": 2.7367,
      "step": 843
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5842541136276735,
      "learning_rate": 0.0007014205742229227,
      "loss": 2.7953,
      "step": 844
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5240874226653945,
      "learning_rate": 0.0007007402656756072,
      "loss": 2.8248,
      "step": 845
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.4993236489277078,
      "learning_rate": 0.0007000595138012797,
      "loss": 2.7417,
      "step": 846
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5514996901209408,
      "learning_rate": 0.0006993783201033535,
      "loss": 2.7423,
      "step": 847
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5344575439985406,
      "learning_rate": 0.0006986966860862182,
      "loss": 2.6965,
      "step": 848
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5739598688379087,
      "learning_rate": 0.000698014613255236,
      "loss": 2.7276,
      "step": 849
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5219557469078394,
      "learning_rate": 0.0006973321031167382,
      "loss": 2.7379,
      "step": 850
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5638436367299189,
      "learning_rate": 0.0006966491571780216,
      "loss": 2.8022,
      "step": 851
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5493939763772003,
      "learning_rate": 0.0006959657769473453,
      "loss": 2.7537,
      "step": 852
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5205268381254381,
      "learning_rate": 0.000695281963933928,
      "loss": 2.7773,
      "step": 853
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5078383199467512,
      "learning_rate": 0.0006945977196479438,
      "loss": 2.7682,
      "step": 854
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5014949245052864,
      "learning_rate": 0.0006939130456005196,
      "loss": 2.759,
      "step": 855
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5029223070598458,
      "learning_rate": 0.0006932279433037311,
      "loss": 2.806,
      "step": 856
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.4709328628183458,
      "learning_rate": 0.0006925424142705997,
      "loss": 2.6671,
      "step": 857
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5076559891877385,
      "learning_rate": 0.0006918564600150896,
      "loss": 2.7781,
      "step": 858
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.479546724105253,
      "learning_rate": 0.0006911700820521042,
      "loss": 2.7367,
      "step": 859
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5092651438817656,
      "learning_rate": 0.0006904832818974818,
      "loss": 2.8519,
      "step": 860
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5618203714233453,
      "learning_rate": 0.0006897960610679939,
      "loss": 2.7845,
      "step": 861
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.48106458305246025,
      "learning_rate": 0.0006891084210813407,
      "loss": 2.8059,
      "step": 862
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.48366679677231933,
      "learning_rate": 0.0006884203634561483,
      "loss": 2.7889,
      "step": 863
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5492583624585266,
      "learning_rate": 0.0006877318897119651,
      "loss": 2.7834,
      "step": 864
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5054741274256389,
      "learning_rate": 0.0006870430013692579,
      "loss": 2.8286,
      "step": 865
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.48903380592737233,
      "learning_rate": 0.0006863536999494101,
      "loss": 2.78,
      "step": 866
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.4987610224213302,
      "learning_rate": 0.0006856639869747167,
      "loss": 2.799,
      "step": 867
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5381791610275435,
      "learning_rate": 0.0006849738639683818,
      "loss": 2.7685,
      "step": 868
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.49956599991520023,
      "learning_rate": 0.000684283332454515,
      "loss": 2.7864,
      "step": 869
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.466848144196578,
      "learning_rate": 0.0006835923939581281,
      "loss": 2.7065,
      "step": 870
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.48405128109984186,
      "learning_rate": 0.0006829010500051318,
      "loss": 2.7659,
      "step": 871
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5353280000892553,
      "learning_rate": 0.0006822093021223321,
      "loss": 2.7429,
      "step": 872
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.524535401854029,
      "learning_rate": 0.0006815171518374268,
      "loss": 2.8012,
      "step": 873
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5331439117843774,
      "learning_rate": 0.0006808246006790031,
      "loss": 2.7336,
      "step": 874
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5187697764424821,
      "learning_rate": 0.0006801316501765329,
      "loss": 2.7901,
      "step": 875
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5115009666948729,
      "learning_rate": 0.0006794383018603704,
      "loss": 2.7998,
      "step": 876
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5265446854041065,
      "learning_rate": 0.0006787445572617481,
      "loss": 2.809,
      "step": 877
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5165138048260781,
      "learning_rate": 0.0006780504179127734,
      "loss": 2.7829,
      "step": 878
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5388697163620849,
      "learning_rate": 0.0006773558853464265,
      "loss": 2.7535,
      "step": 879
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.49606269040249723,
      "learning_rate": 0.000676660961096555,
      "loss": 2.8486,
      "step": 880
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.48285254073520323,
      "learning_rate": 0.000675965646697872,
      "loss": 2.8458,
      "step": 881
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5037317036941886,
      "learning_rate": 0.0006752699436859519,
      "loss": 2.6753,
      "step": 882
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5331738557233788,
      "learning_rate": 0.0006745738535972279,
      "loss": 2.7633,
      "step": 883
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5059399351796149,
      "learning_rate": 0.0006738773779689874,
      "loss": 2.8242,
      "step": 884
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.49336863688376975,
      "learning_rate": 0.0006731805183393696,
      "loss": 2.6581,
      "step": 885
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5338208080333934,
      "learning_rate": 0.0006724832762473618,
      "loss": 2.7471,
      "step": 886
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.512212145987671,
      "learning_rate": 0.0006717856532327956,
      "loss": 2.7511,
      "step": 887
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5089148809429237,
      "learning_rate": 0.0006710876508363444,
      "loss": 2.7457,
      "step": 888
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5354632741946636,
      "learning_rate": 0.0006703892705995189,
      "loss": 2.6854,
      "step": 889
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5304678838742364,
      "learning_rate": 0.0006696905140646647,
      "loss": 2.7535,
      "step": 890
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.533702012090405,
      "learning_rate": 0.0006689913827749581,
      "loss": 2.8708,
      "step": 891
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5368945292993025,
      "learning_rate": 0.0006682918782744032,
      "loss": 2.6945,
      "step": 892
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5129913038785266,
      "learning_rate": 0.0006675920021078282,
      "loss": 2.7662,
      "step": 893
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5239537318193066,
      "learning_rate": 0.0006668917558208823,
      "loss": 2.751,
      "step": 894
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.49404334394218646,
      "learning_rate": 0.0006661911409600321,
      "loss": 2.744,
      "step": 895
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5279168802712604,
      "learning_rate": 0.0006654901590725577,
      "loss": 2.7897,
      "step": 896
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.47740648051606127,
      "learning_rate": 0.0006647888117065507,
      "loss": 2.776,
      "step": 897
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.49574907735898144,
      "learning_rate": 0.0006640871004109086,
      "loss": 2.7985,
      "step": 898
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.48851802670010686,
      "learning_rate": 0.000663385026735334,
      "loss": 2.7733,
      "step": 899
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.4901231714809981,
      "learning_rate": 0.0006626825922303285,
      "loss": 2.7551,
      "step": 900
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5340245179409802,
      "learning_rate": 0.0006619797984471915,
      "loss": 2.8051,
      "step": 901
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5176702974142926,
      "learning_rate": 0.0006612766469380158,
      "loss": 2.6531,
      "step": 902
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.515304315905886,
      "learning_rate": 0.0006605731392556833,
      "loss": 2.7889,
      "step": 903
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5508704003690635,
      "learning_rate": 0.0006598692769538637,
      "loss": 2.784,
      "step": 904
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.4822902267852383,
      "learning_rate": 0.0006591650615870091,
      "loss": 2.7872,
      "step": 905
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.4859115343419366,
      "learning_rate": 0.0006584604947103514,
      "loss": 2.7676,
      "step": 906
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5085655835784982,
      "learning_rate": 0.0006577555778798993,
      "loss": 2.7351,
      "step": 907
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.4860661074427596,
      "learning_rate": 0.0006570503126524336,
      "loss": 2.7096,
      "step": 908
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.4699952016210587,
      "learning_rate": 0.0006563447005855054,
      "loss": 2.7034,
      "step": 909
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5060565648840971,
      "learning_rate": 0.000655638743237431,
      "loss": 2.7263,
      "step": 910
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.47594623453155094,
      "learning_rate": 0.0006549324421672894,
      "loss": 2.7143,
      "step": 911
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.51332013498744,
      "learning_rate": 0.0006542257989349194,
      "loss": 2.7109,
      "step": 912
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.49432391712052376,
      "learning_rate": 0.0006535188151009142,
      "loss": 2.7343,
      "step": 913
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5386518205007791,
      "learning_rate": 0.0006528114922266204,
      "loss": 2.7886,
      "step": 914
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5499047415726279,
      "learning_rate": 0.0006521038318741327,
      "loss": 2.7817,
      "step": 915
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5244798311024138,
      "learning_rate": 0.0006513958356062912,
      "loss": 2.8404,
      "step": 916
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5413925971366917,
      "learning_rate": 0.0006506875049866781,
      "loss": 2.6724,
      "step": 917
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5134299712873807,
      "learning_rate": 0.0006499788415796137,
      "loss": 2.7241,
      "step": 918
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.4569887968329291,
      "learning_rate": 0.0006492698469501532,
      "loss": 2.7116,
      "step": 919
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5224365872653551,
      "learning_rate": 0.0006485605226640837,
      "loss": 2.7662,
      "step": 920
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5322607243526167,
      "learning_rate": 0.00064785087028792,
      "loss": 2.7222,
      "step": 921
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.49290865595176075,
      "learning_rate": 0.0006471408913889019,
      "loss": 2.8255,
      "step": 922
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5333278998203554,
      "learning_rate": 0.0006464305875349892,
      "loss": 2.7839,
      "step": 923
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5121321016797032,
      "learning_rate": 0.000645719960294861,
      "loss": 2.7508,
      "step": 924
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.48885744251733626,
      "learning_rate": 0.0006450090112379092,
      "loss": 2.7846,
      "step": 925
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5120561873544219,
      "learning_rate": 0.0006442977419342371,
      "loss": 2.7736,
      "step": 926
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5306422198088276,
      "learning_rate": 0.000643586153954655,
      "loss": 2.7874,
      "step": 927
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.565391511598094,
      "learning_rate": 0.0006428742488706772,
      "loss": 2.7837,
      "step": 928
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5100710447487015,
      "learning_rate": 0.0006421620282545182,
      "loss": 2.7277,
      "step": 929
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5068331248190641,
      "learning_rate": 0.0006414494936790892,
      "loss": 2.7277,
      "step": 930
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5268686984396896,
      "learning_rate": 0.0006407366467179951,
      "loss": 2.8227,
      "step": 931
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.6331680104171693,
      "learning_rate": 0.0006400234889455301,
      "loss": 2.7951,
      "step": 932
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5286557193661298,
      "learning_rate": 0.0006393100219366755,
      "loss": 2.8292,
      "step": 933
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.47977126524528535,
      "learning_rate": 0.0006385962472670953,
      "loss": 2.8411,
      "step": 934
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5306774229454788,
      "learning_rate": 0.0006378821665131328,
      "loss": 2.8441,
      "step": 935
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5531360143509272,
      "learning_rate": 0.0006371677812518072,
      "loss": 2.7508,
      "step": 936
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.49015902487161006,
      "learning_rate": 0.0006364530930608107,
      "loss": 2.767,
      "step": 937
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.48197842791490547,
      "learning_rate": 0.0006357381035185038,
      "loss": 2.7212,
      "step": 938
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.4960997386760046,
      "learning_rate": 0.0006350228142039131,
      "loss": 2.756,
      "step": 939
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.4889530258340255,
      "learning_rate": 0.000634307226696727,
      "loss": 2.783,
      "step": 940
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5057503592741923,
      "learning_rate": 0.0006335913425772926,
      "loss": 2.7917,
      "step": 941
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.4943948350428404,
      "learning_rate": 0.0006328751634266117,
      "loss": 2.6598,
      "step": 942
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.46633811615065784,
      "learning_rate": 0.0006321586908263382,
      "loss": 2.8788,
      "step": 943
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5027357980409337,
      "learning_rate": 0.0006314419263587732,
      "loss": 2.8161,
      "step": 944
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.4780403947059295,
      "learning_rate": 0.0006307248716068637,
      "loss": 2.7568,
      "step": 945
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5354898944258344,
      "learning_rate": 0.0006300075281541964,
      "loss": 2.782,
      "step": 946
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.4825693139234744,
      "learning_rate": 0.0006292898975849966,
      "loss": 2.6409,
      "step": 947
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.4953320693853845,
      "learning_rate": 0.000628571981484123,
      "loss": 2.7598,
      "step": 948
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.49583074441018976,
      "learning_rate": 0.0006278537814370654,
      "loss": 2.731,
      "step": 949
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.506943671479962,
      "learning_rate": 0.0006271352990299406,
      "loss": 2.7261,
      "step": 950
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.49590862978012457,
      "learning_rate": 0.0006264165358494885,
      "loss": 2.6804,
      "step": 951
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.4836761407691934,
      "learning_rate": 0.0006256974934830694,
      "loss": 2.7867,
      "step": 952
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.49074157533365576,
      "learning_rate": 0.0006249781735186606,
      "loss": 2.7114,
      "step": 953
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.484774695388245,
      "learning_rate": 0.0006242585775448518,
      "loss": 2.7948,
      "step": 954
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.4941148175026909,
      "learning_rate": 0.0006235387071508427,
      "loss": 2.692,
      "step": 955
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.4847084319877268,
      "learning_rate": 0.0006228185639264384,
      "loss": 2.7059,
      "step": 956
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5071978236179299,
      "learning_rate": 0.0006220981494620475,
      "loss": 2.7394,
      "step": 957
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5052658765389426,
      "learning_rate": 0.000621377465348677,
      "loss": 2.7638,
      "step": 958
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5064093323515364,
      "learning_rate": 0.0006206565131779293,
      "loss": 2.7689,
      "step": 959
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5266116971099526,
      "learning_rate": 0.0006199352945419994,
      "loss": 2.7573,
      "step": 960
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.52258713353477,
      "learning_rate": 0.00061921381103367,
      "loss": 2.7787,
      "step": 961
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.492513899615317,
      "learning_rate": 0.0006184920642463094,
      "loss": 2.7568,
      "step": 962
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.4940183946813311,
      "learning_rate": 0.0006177700557738672,
      "loss": 2.7518,
      "step": 963
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.48677685476444593,
      "learning_rate": 0.0006170477872108706,
      "loss": 2.6772,
      "step": 964
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.4939143055623651,
      "learning_rate": 0.0006163252601524216,
      "loss": 2.7732,
      "step": 965
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5095451104144231,
      "learning_rate": 0.0006156024761941925,
      "loss": 2.679,
      "step": 966
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.513846128660273,
      "learning_rate": 0.000614879436932424,
      "loss": 2.7193,
      "step": 967
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.4724434249408676,
      "learning_rate": 0.0006141561439639196,
      "loss": 2.7792,
      "step": 968
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.48717747250736454,
      "learning_rate": 0.0006134325988860433,
      "loss": 2.794,
      "step": 969
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5029007019676133,
      "learning_rate": 0.0006127088032967165,
      "loss": 2.7591,
      "step": 970
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.4787567617208102,
      "learning_rate": 0.0006119847587944131,
      "loss": 2.784,
      "step": 971
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.4639459141945223,
      "learning_rate": 0.0006112604669781572,
      "loss": 2.7821,
      "step": 972
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.4803397577434123,
      "learning_rate": 0.0006105359294475188,
      "loss": 2.7296,
      "step": 973
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5274444754035275,
      "learning_rate": 0.0006098111478026107,
      "loss": 2.7894,
      "step": 974
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.49882630644473697,
      "learning_rate": 0.0006090861236440848,
      "loss": 2.7505,
      "step": 975
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5118656383712906,
      "learning_rate": 0.0006083608585731282,
      "loss": 2.7583,
      "step": 976
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5382357072396842,
      "learning_rate": 0.0006076353541914609,
      "loss": 2.7801,
      "step": 977
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.46786372948033095,
      "learning_rate": 0.0006069096121013307,
      "loss": 2.7018,
      "step": 978
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5176454688066614,
      "learning_rate": 0.0006061836339055105,
      "loss": 2.7784,
      "step": 979
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5218061053688355,
      "learning_rate": 0.0006054574212072948,
      "loss": 2.7018,
      "step": 980
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.4683031931534088,
      "learning_rate": 0.0006047309756104958,
      "loss": 2.7693,
      "step": 981
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.48629138149265483,
      "learning_rate": 0.00060400429871944,
      "loss": 2.6775,
      "step": 982
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.557362827835192,
      "learning_rate": 0.0006032773921389654,
      "loss": 2.7742,
      "step": 983
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5254113038041371,
      "learning_rate": 0.0006025502574744162,
      "loss": 2.78,
      "step": 984
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5059265191460307,
      "learning_rate": 0.000601822896331641,
      "loss": 2.7791,
      "step": 985
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.47105009173924467,
      "learning_rate": 0.0006010953103169883,
      "loss": 2.6798,
      "step": 986
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5561196580147597,
      "learning_rate": 0.0006003675010373034,
      "loss": 2.8296,
      "step": 987
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5165045770334074,
      "learning_rate": 0.0005996394700999246,
      "loss": 2.9029,
      "step": 988
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5101993421533687,
      "learning_rate": 0.0005989112191126794,
      "loss": 2.7209,
      "step": 989
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5177171252845812,
      "learning_rate": 0.0005981827496838822,
      "loss": 2.7911,
      "step": 990
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.4724719435665852,
      "learning_rate": 0.0005974540634223286,
      "loss": 2.786,
      "step": 991
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5191707403352941,
      "learning_rate": 0.0005967251619372939,
      "loss": 2.6762,
      "step": 992
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5349647184014108,
      "learning_rate": 0.0005959960468385284,
      "loss": 2.6603,
      "step": 993
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5276285975099707,
      "learning_rate": 0.0005952667197362542,
      "loss": 2.742,
      "step": 994
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.48761272423611945,
      "learning_rate": 0.0005945371822411621,
      "loss": 2.7236,
      "step": 995
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.4951388158867274,
      "learning_rate": 0.0005938074359644063,
      "loss": 2.7032,
      "step": 996
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5179825452776126,
      "learning_rate": 0.0005930774825176034,
      "loss": 2.7732,
      "step": 997
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5261671815856594,
      "learning_rate": 0.0005923473235128268,
      "loss": 2.6949,
      "step": 998
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5454529618310321,
      "learning_rate": 0.0005916169605626042,
      "loss": 2.7585,
      "step": 999
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5171200520944219,
      "learning_rate": 0.0005908863952799134,
      "loss": 2.7531,
      "step": 1000
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5108482656003888,
      "learning_rate": 0.0005901556292781793,
      "loss": 2.6341,
      "step": 1001
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5133915239424481,
      "learning_rate": 0.0005894246641712698,
      "loss": 2.7207,
      "step": 1002
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5178007832270196,
      "learning_rate": 0.0005886935015734931,
      "loss": 2.6931,
      "step": 1003
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5261630751402555,
      "learning_rate": 0.0005879621430995928,
      "loss": 2.7794,
      "step": 1004
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5800813607505337,
      "learning_rate": 0.0005872305903647455,
      "loss": 2.7248,
      "step": 1005
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.47443508986314176,
      "learning_rate": 0.0005864988449845569,
      "loss": 2.6529,
      "step": 1006
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5070692657788186,
      "learning_rate": 0.0005857669085750578,
      "loss": 2.698,
      "step": 1007
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.49710768733498933,
      "learning_rate": 0.0005850347827527013,
      "loss": 2.8145,
      "step": 1008
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.48794295035729135,
      "learning_rate": 0.0005843024691343584,
      "loss": 2.7522,
      "step": 1009
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5372183075893122,
      "learning_rate": 0.000583569969337315,
      "loss": 2.6309,
      "step": 1010
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5148698851559652,
      "learning_rate": 0.0005828372849792686,
      "loss": 2.7922,
      "step": 1011
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5253495243002605,
      "learning_rate": 0.0005821044176783234,
      "loss": 2.6397,
      "step": 1012
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5615607894519717,
      "learning_rate": 0.0005813713690529886,
      "loss": 2.8245,
      "step": 1013
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5584846261092475,
      "learning_rate": 0.0005806381407221729,
      "loss": 2.7152,
      "step": 1014
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.6014166606226927,
      "learning_rate": 0.0005799047343051826,
      "loss": 2.7203,
      "step": 1015
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.525573548062308,
      "learning_rate": 0.0005791711514217171,
      "loss": 2.6897,
      "step": 1016
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5312153749005265,
      "learning_rate": 0.0005784373936918654,
      "loss": 2.809,
      "step": 1017
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5143882775165995,
      "learning_rate": 0.0005777034627361025,
      "loss": 2.662,
      "step": 1018
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5359573245878332,
      "learning_rate": 0.0005769693601752864,
      "loss": 2.6911,
      "step": 1019
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5109889887406951,
      "learning_rate": 0.0005762350876306537,
      "loss": 2.7018,
      "step": 1020
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5172131414778152,
      "learning_rate": 0.0005755006467238168,
      "loss": 2.6924,
      "step": 1021
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5485499278181196,
      "learning_rate": 0.0005747660390767593,
      "loss": 2.8091,
      "step": 1022
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.48054642686647003,
      "learning_rate": 0.0005740312663118338,
      "loss": 2.6736,
      "step": 1023
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.521044664523183,
      "learning_rate": 0.0005732963300517568,
      "loss": 2.7532,
      "step": 1024
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5361701837440671,
      "learning_rate": 0.0005725612319196064,
      "loss": 2.8393,
      "step": 1025
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.49084106343502676,
      "learning_rate": 0.000571825973538818,
      "loss": 2.7183,
      "step": 1026
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5375516542212618,
      "learning_rate": 0.0005710905565331811,
      "loss": 2.7335,
      "step": 1027
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5053029384847482,
      "learning_rate": 0.0005703549825268353,
      "loss": 2.7108,
      "step": 1028
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.519682072044458,
      "learning_rate": 0.0005696192531442667,
      "loss": 2.7367,
      "step": 1029
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.4757472288509061,
      "learning_rate": 0.000568883370010305,
      "loss": 2.7173,
      "step": 1030
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5392330537381186,
      "learning_rate": 0.0005681473347501192,
      "loss": 2.7214,
      "step": 1031
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5315838547724138,
      "learning_rate": 0.0005674111489892144,
      "loss": 2.7917,
      "step": 1032
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.448318881403438,
      "learning_rate": 0.0005666748143534282,
      "loss": 2.6982,
      "step": 1033
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.4972833521635573,
      "learning_rate": 0.0005659383324689266,
      "loss": 2.8145,
      "step": 1034
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.4916731453327063,
      "learning_rate": 0.0005652017049622007,
      "loss": 2.7421,
      "step": 1035
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5171637772665034,
      "learning_rate": 0.0005644649334600641,
      "loss": 2.7596,
      "step": 1036
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.542883455271621,
      "learning_rate": 0.0005637280195896474,
      "loss": 2.7189,
      "step": 1037
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5282758428090761,
      "learning_rate": 0.0005629909649783961,
      "loss": 2.6737,
      "step": 1038
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5089528080633915,
      "learning_rate": 0.0005622537712540664,
      "loss": 2.6951,
      "step": 1039
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5243099019740807,
      "learning_rate": 0.0005615164400447218,
      "loss": 2.7161,
      "step": 1040
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.509150957834291,
      "learning_rate": 0.0005607789729787294,
      "loss": 2.7112,
      "step": 1041
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5631580259985867,
      "learning_rate": 0.0005600413716847564,
      "loss": 2.7171,
      "step": 1042
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5872546894665344,
      "learning_rate": 0.000559303637791766,
      "loss": 2.7637,
      "step": 1043
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.45929386151150653,
      "learning_rate": 0.0005585657729290151,
      "loss": 2.6886,
      "step": 1044
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.44537439497508813,
      "learning_rate": 0.000557827778726049,
      "loss": 2.7289,
      "step": 1045
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5381913357471398,
      "learning_rate": 0.0005570896568126993,
      "loss": 2.8449,
      "step": 1046
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5504937001251007,
      "learning_rate": 0.0005563514088190788,
      "loss": 2.7537,
      "step": 1047
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5667706482242284,
      "learning_rate": 0.0005556130363755798,
      "loss": 2.7856,
      "step": 1048
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5138881491458973,
      "learning_rate": 0.0005548745411128688,
      "loss": 2.7844,
      "step": 1049
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5078894393261245,
      "learning_rate": 0.0005541359246618835,
      "loss": 2.6914,
      "step": 1050
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5473828440912689,
      "learning_rate": 0.0005533971886538293,
      "loss": 2.7733,
      "step": 1051
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5122145383119249,
      "learning_rate": 0.000552658334720176,
      "loss": 2.7788,
      "step": 1052
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5094814064577552,
      "learning_rate": 0.0005519193644926535,
      "loss": 2.7697,
      "step": 1053
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5311239360105956,
      "learning_rate": 0.0005511802796032485,
      "loss": 2.6993,
      "step": 1054
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5056310115445127,
      "learning_rate": 0.0005504410816842009,
      "loss": 2.7289,
      "step": 1055
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.4949936441053868,
      "learning_rate": 0.0005497017723680009,
      "loss": 2.7502,
      "step": 1056
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5454563029138563,
      "learning_rate": 0.0005489623532873836,
      "loss": 2.7846,
      "step": 1057
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5354216455781943,
      "learning_rate": 0.0005482228260753273,
      "loss": 2.7601,
      "step": 1058
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.48572818637266924,
      "learning_rate": 0.0005474831923650488,
      "loss": 2.7872,
      "step": 1059
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5263026450656437,
      "learning_rate": 0.00054674345379,
      "loss": 2.6232,
      "step": 1060
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5000018369139888,
      "learning_rate": 0.000546003611983865,
      "loss": 2.6328,
      "step": 1061
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5556285101513226,
      "learning_rate": 0.0005452636685805552,
      "loss": 2.7592,
      "step": 1062
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.45809746005774116,
      "learning_rate": 0.0005445236252142066,
      "loss": 2.755,
      "step": 1063
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.4946177166899814,
      "learning_rate": 0.000543783483519176,
      "loss": 2.7223,
      "step": 1064
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5428958050953088,
      "learning_rate": 0.0005430432451300374,
      "loss": 2.728,
      "step": 1065
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.4984862220837191,
      "learning_rate": 0.0005423029116815781,
      "loss": 2.6564,
      "step": 1066
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5202225535340897,
      "learning_rate": 0.0005415624848087959,
      "loss": 2.6913,
      "step": 1067
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.4840020784650056,
      "learning_rate": 0.000540821966146894,
      "loss": 2.7645,
      "step": 1068
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.537218186589317,
      "learning_rate": 0.0005400813573312793,
      "loss": 2.6545,
      "step": 1069
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.515707831667649,
      "learning_rate": 0.0005393406599975572,
      "loss": 2.749,
      "step": 1070
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5549236832823138,
      "learning_rate": 0.0005385998757815287,
      "loss": 2.7562,
      "step": 1071
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5343596595593862,
      "learning_rate": 0.0005378590063191867,
      "loss": 2.6751,
      "step": 1072
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.4858988524039984,
      "learning_rate": 0.0005371180532467124,
      "loss": 2.6543,
      "step": 1073
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5207563675106729,
      "learning_rate": 0.000536377018200472,
      "loss": 2.7075,
      "step": 1074
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5016061501970053,
      "learning_rate": 0.0005356359028170118,
      "loss": 2.7058,
      "step": 1075
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5389093829141083,
      "learning_rate": 0.0005348947087330564,
      "loss": 2.6559,
      "step": 1076
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.500205939366958,
      "learning_rate": 0.0005341534375855037,
      "loss": 2.7282,
      "step": 1077
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5106159799557759,
      "learning_rate": 0.0005334120910114222,
      "loss": 2.6785,
      "step": 1078
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.504326286873783,
      "learning_rate": 0.0005326706706480467,
      "loss": 2.6254,
      "step": 1079
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5253973148032602,
      "learning_rate": 0.0005319291781327749,
      "loss": 2.7623,
      "step": 1080
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5000257783006888,
      "learning_rate": 0.0005311876151031642,
      "loss": 2.6727,
      "step": 1081
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5172388003338478,
      "learning_rate": 0.0005304459831969274,
      "loss": 2.7316,
      "step": 1082
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5201798177167883,
      "learning_rate": 0.0005297042840519294,
      "loss": 2.7758,
      "step": 1083
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5309435318399293,
      "learning_rate": 0.0005289625193061838,
      "loss": 2.7189,
      "step": 1084
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.527008962765747,
      "learning_rate": 0.0005282206905978489,
      "loss": 2.7294,
      "step": 1085
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5050458147191232,
      "learning_rate": 0.0005274787995652246,
      "loss": 2.6612,
      "step": 1086
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5502926357148964,
      "learning_rate": 0.000526736847846748,
      "loss": 2.7704,
      "step": 1087
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.490455214150219,
      "learning_rate": 0.0005259948370809901,
      "loss": 2.7215,
      "step": 1088
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5347934406932487,
      "learning_rate": 0.0005252527689066533,
      "loss": 2.775,
      "step": 1089
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5154514383005036,
      "learning_rate": 0.0005245106449625654,
      "loss": 2.7741,
      "step": 1090
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.49788471680941243,
      "learning_rate": 0.0005237684668876785,
      "loss": 2.6886,
      "step": 1091
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5622040490861864,
      "learning_rate": 0.0005230262363210637,
      "loss": 2.7918,
      "step": 1092
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5033855334705318,
      "learning_rate": 0.0005222839549019079,
      "loss": 2.7837,
      "step": 1093
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.48777337194218956,
      "learning_rate": 0.0005215416242695108,
      "loss": 2.712,
      "step": 1094
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5174186221509732,
      "learning_rate": 0.0005207992460632804,
      "loss": 2.8,
      "step": 1095
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5142965927082324,
      "learning_rate": 0.0005200568219227299,
      "loss": 2.7445,
      "step": 1096
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5322244992730195,
      "learning_rate": 0.000519314353487474,
      "loss": 2.7588,
      "step": 1097
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.48693592767653204,
      "learning_rate": 0.0005185718423972251,
      "loss": 2.6827,
      "step": 1098
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5046692545615626,
      "learning_rate": 0.0005178292902917898,
      "loss": 2.7289,
      "step": 1099
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.47978454683003274,
      "learning_rate": 0.0005170866988110656,
      "loss": 2.7013,
      "step": 1100
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.4788540630749357,
      "learning_rate": 0.0005163440695950362,
      "loss": 2.6882,
      "step": 1101
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.4612893735815785,
      "learning_rate": 0.0005156014042837695,
      "loss": 2.617,
      "step": 1102
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.4898027943481741,
      "learning_rate": 0.0005148587045174128,
      "loss": 2.7188,
      "step": 1103
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5748413551545054,
      "learning_rate": 0.0005141159719361891,
      "loss": 2.6455,
      "step": 1104
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5042890521101759,
      "learning_rate": 0.0005133732081803945,
      "loss": 2.7683,
      "step": 1105
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.467371098882922,
      "learning_rate": 0.0005126304148903936,
      "loss": 2.7084,
      "step": 1106
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.49812638741410353,
      "learning_rate": 0.0005118875937066161,
      "loss": 2.6525,
      "step": 1107
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5431472257240424,
      "learning_rate": 0.0005111447462695537,
      "loss": 2.6888,
      "step": 1108
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.522739308694978,
      "learning_rate": 0.0005104018742197557,
      "loss": 2.7076,
      "step": 1109
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5522937972499815,
      "learning_rate": 0.0005096589791978261,
      "loss": 2.7346,
      "step": 1110
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5031886324169182,
      "learning_rate": 0.0005089160628444192,
      "loss": 2.7199,
      "step": 1111
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5084434276140318,
      "learning_rate": 0.0005081731268002371,
      "loss": 2.7622,
      "step": 1112
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.511271211085511,
      "learning_rate": 0.0005074301727060243,
      "loss": 2.7239,
      "step": 1113
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.544946721292002,
      "learning_rate": 0.0005066872022025663,
      "loss": 2.7618,
      "step": 1114
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5619258028846829,
      "learning_rate": 0.0005059442169306844,
      "loss": 2.6509,
      "step": 1115
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5295161880183932,
      "learning_rate": 0.0005052012185312321,
      "loss": 2.7507,
      "step": 1116
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5068540600601226,
      "learning_rate": 0.0005044582086450925,
      "loss": 2.7624,
      "step": 1117
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5268615789640267,
      "learning_rate": 0.0005037151889131737,
      "loss": 2.6579,
      "step": 1118
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5153348593645998,
      "learning_rate": 0.0005029721609764059,
      "loss": 2.7871,
      "step": 1119
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5100418837135368,
      "learning_rate": 0.000502229126475737,
      "loss": 2.6463,
      "step": 1120
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5443509651515888,
      "learning_rate": 0.0005014860870521293,
      "loss": 2.6746,
      "step": 1121
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5419846571794396,
      "learning_rate": 0.0005007430443465569,
      "loss": 2.7415,
      "step": 1122
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5175078596247342,
      "learning_rate": 0.0005,
      "loss": 2.6247,
      "step": 1123
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5179433140743496,
      "learning_rate": 0.0004992569556534432,
      "loss": 2.6846,
      "step": 1124
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4946276392672176,
      "learning_rate": 0.0004985139129478707,
      "loss": 2.6575,
      "step": 1125
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.48893495857371877,
      "learning_rate": 0.0004977708735242633,
      "loss": 2.7387,
      "step": 1126
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4955502765030083,
      "learning_rate": 0.0004970278390235942,
      "loss": 2.7535,
      "step": 1127
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5168723682375538,
      "learning_rate": 0.0004962848110868262,
      "loss": 2.8069,
      "step": 1128
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5428503966348986,
      "learning_rate": 0.0004955417913549074,
      "loss": 2.7472,
      "step": 1129
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5306641995533713,
      "learning_rate": 0.0004947987814687679,
      "loss": 2.6359,
      "step": 1130
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.48015854367251193,
      "learning_rate": 0.0004940557830693157,
      "loss": 2.6359,
      "step": 1131
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.48039658976659516,
      "learning_rate": 0.0004933127977974338,
      "loss": 2.6904,
      "step": 1132
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4853648270056312,
      "learning_rate": 0.0004925698272939757,
      "loss": 2.6597,
      "step": 1133
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5279285976556198,
      "learning_rate": 0.0004918268731997632,
      "loss": 2.7264,
      "step": 1134
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5161431000228389,
      "learning_rate": 0.0004910839371555809,
      "loss": 2.7227,
      "step": 1135
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.49553525141278476,
      "learning_rate": 0.0004903410208021739,
      "loss": 2.6899,
      "step": 1136
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5059000734394524,
      "learning_rate": 0.0004895981257802443,
      "loss": 2.7795,
      "step": 1137
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5350253735217294,
      "learning_rate": 0.0004888552537304463,
      "loss": 2.7082,
      "step": 1138
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5019712060738322,
      "learning_rate": 0.00048811240629338394,
      "loss": 2.6921,
      "step": 1139
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.523206090444028,
      "learning_rate": 0.00048736958510960663,
      "loss": 2.6801,
      "step": 1140
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5164272224817171,
      "learning_rate": 0.00048662679181960564,
      "loss": 2.738,
      "step": 1141
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5830602270625335,
      "learning_rate": 0.00048588402806381094,
      "loss": 2.7607,
      "step": 1142
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5298383467699622,
      "learning_rate": 0.0004851412954825874,
      "loss": 2.717,
      "step": 1143
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.512246394627895,
      "learning_rate": 0.00048439859571623034,
      "loss": 2.6335,
      "step": 1144
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5011379483814702,
      "learning_rate": 0.00048365593040496373,
      "loss": 2.6826,
      "step": 1145
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5281686013861873,
      "learning_rate": 0.00048291330118893443,
      "loss": 2.7771,
      "step": 1146
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5102472914604643,
      "learning_rate": 0.0004821707097082102,
      "loss": 2.6045,
      "step": 1147
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5091022957421559,
      "learning_rate": 0.0004814281576027749,
      "loss": 2.8107,
      "step": 1148
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5303007055945778,
      "learning_rate": 0.000480685646512526,
      "loss": 2.7708,
      "step": 1149
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.49490887220412944,
      "learning_rate": 0.00047994317807727025,
      "loss": 2.6633,
      "step": 1150
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.513011380391448,
      "learning_rate": 0.00047920075393671974,
      "loss": 2.7299,
      "step": 1151
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5059021296300894,
      "learning_rate": 0.0004784583757304893,
      "loss": 2.679,
      "step": 1152
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.48525591061607676,
      "learning_rate": 0.00047771604509809214,
      "loss": 2.6396,
      "step": 1153
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5659968954298682,
      "learning_rate": 0.0004769737636789364,
      "loss": 2.7725,
      "step": 1154
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5279178618523368,
      "learning_rate": 0.00047623153311232157,
      "loss": 2.7942,
      "step": 1155
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5146934688183075,
      "learning_rate": 0.0004754893550374346,
      "loss": 2.6455,
      "step": 1156
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5003048276789507,
      "learning_rate": 0.00047474723109334685,
      "loss": 2.7766,
      "step": 1157
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5297617102579024,
      "learning_rate": 0.00047400516291900993,
      "loss": 2.6946,
      "step": 1158
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5477355727573853,
      "learning_rate": 0.0004732631521532522,
      "loss": 2.7106,
      "step": 1159
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.545802547383211,
      "learning_rate": 0.0004725212004347755,
      "loss": 2.7187,
      "step": 1160
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5231349493200539,
      "learning_rate": 0.00047177930940215095,
      "loss": 2.5814,
      "step": 1161
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5099141050656486,
      "learning_rate": 0.00047103748069381624,
      "loss": 2.6898,
      "step": 1162
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.4930609535375999,
      "learning_rate": 0.0004702957159480707,
      "loss": 2.5817,
      "step": 1163
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5110460153161239,
      "learning_rate": 0.00046955401680307267,
      "loss": 2.7144,
      "step": 1164
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5264240278823245,
      "learning_rate": 0.0004688123848968359,
      "loss": 2.6491,
      "step": 1165
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.4861388795181152,
      "learning_rate": 0.00046807082186722516,
      "loss": 2.7457,
      "step": 1166
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.538320758121501,
      "learning_rate": 0.0004673293293519535,
      "loss": 2.7267,
      "step": 1167
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5141531601284485,
      "learning_rate": 0.00046658790898857806,
      "loss": 2.6588,
      "step": 1168
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5393189514744423,
      "learning_rate": 0.0004658465624144963,
      "loss": 2.6439,
      "step": 1169
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5075406485654212,
      "learning_rate": 0.0004651052912669438,
      "loss": 2.6696,
      "step": 1170
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5084751938532551,
      "learning_rate": 0.0004643640971829883,
      "loss": 2.6247,
      "step": 1171
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5521555175386781,
      "learning_rate": 0.0004636229817995281,
      "loss": 2.7402,
      "step": 1172
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.4957649974272555,
      "learning_rate": 0.0004628819467532876,
      "loss": 2.6356,
      "step": 1173
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5433136295282748,
      "learning_rate": 0.00046214099368081335,
      "loss": 2.6873,
      "step": 1174
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.4912787340676148,
      "learning_rate": 0.0004614001242184714,
      "loss": 2.6935,
      "step": 1175
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5217062443217886,
      "learning_rate": 0.000460659340002443,
      "loss": 2.7332,
      "step": 1176
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5442638613752073,
      "learning_rate": 0.00045991864266872073,
      "loss": 2.7246,
      "step": 1177
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5493579234487449,
      "learning_rate": 0.00045917803385310595,
      "loss": 2.6759,
      "step": 1178
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5076552213401087,
      "learning_rate": 0.00045843751519120417,
      "loss": 2.6131,
      "step": 1179
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5152874355491169,
      "learning_rate": 0.00045769708831842193,
      "loss": 2.6546,
      "step": 1180
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5248816597544856,
      "learning_rate": 0.00045695675486996266,
      "loss": 2.7054,
      "step": 1181
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5289329890099478,
      "learning_rate": 0.00045621651648082405,
      "loss": 2.777,
      "step": 1182
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5524923456607232,
      "learning_rate": 0.00045547637478579356,
      "loss": 2.7582,
      "step": 1183
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5140880299388174,
      "learning_rate": 0.0004547363314194449,
      "loss": 2.7499,
      "step": 1184
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5260468960845509,
      "learning_rate": 0.000453996388016135,
      "loss": 2.7139,
      "step": 1185
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5697662426989528,
      "learning_rate": 0.0004532565462099999,
      "loss": 2.7058,
      "step": 1186
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5446586048220051,
      "learning_rate": 0.0004525168076349513,
      "loss": 2.7671,
      "step": 1187
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5154600575004109,
      "learning_rate": 0.0004517771739246729,
      "loss": 2.7239,
      "step": 1188
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5092276381940357,
      "learning_rate": 0.0004510376467126165,
      "loss": 2.6469,
      "step": 1189
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5340423602240948,
      "learning_rate": 0.0004502982276319992,
      "loss": 2.7083,
      "step": 1190
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5021036478189213,
      "learning_rate": 0.0004495589183157991,
      "loss": 2.7102,
      "step": 1191
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5278837389171002,
      "learning_rate": 0.0004488197203967517,
      "loss": 2.6987,
      "step": 1192
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5239167410416333,
      "learning_rate": 0.0004480806355073467,
      "loss": 2.7199,
      "step": 1193
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5537995592297151,
      "learning_rate": 0.000447341665279824,
      "loss": 2.6851,
      "step": 1194
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.4983249639717384,
      "learning_rate": 0.0004466028113461708,
      "loss": 2.6985,
      "step": 1195
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.4929675781049322,
      "learning_rate": 0.0004458640753381167,
      "loss": 2.6313,
      "step": 1196
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5089935520278261,
      "learning_rate": 0.0004451254588871313,
      "loss": 2.733,
      "step": 1197
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5451378729112433,
      "learning_rate": 0.0004443869636244203,
      "loss": 2.7469,
      "step": 1198
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5320232607590252,
      "learning_rate": 0.0004436485911809212,
      "loss": 2.6481,
      "step": 1199
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5317670289041627,
      "learning_rate": 0.00044291034318730087,
      "loss": 2.7457,
      "step": 1200
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5156506956502293,
      "learning_rate": 0.0004421722212739511,
      "loss": 2.7613,
      "step": 1201
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5272163387861835,
      "learning_rate": 0.0004414342270709848,
      "loss": 2.7265,
      "step": 1202
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5275098425681611,
      "learning_rate": 0.00044069636220823397,
      "loss": 2.5964,
      "step": 1203
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5240106078741427,
      "learning_rate": 0.0004399586283152437,
      "loss": 2.7432,
      "step": 1204
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.49511084326793237,
      "learning_rate": 0.0004392210270212706,
      "loss": 2.6584,
      "step": 1205
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5208526920810025,
      "learning_rate": 0.00043848355995527825,
      "loss": 2.7275,
      "step": 1206
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5191648086883246,
      "learning_rate": 0.00043774622874593374,
      "loss": 2.7275,
      "step": 1207
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.4976784460103935,
      "learning_rate": 0.000437009035021604,
      "loss": 2.749,
      "step": 1208
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.48846678466292937,
      "learning_rate": 0.00043627198041035274,
      "loss": 2.6654,
      "step": 1209
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.48387609147398325,
      "learning_rate": 0.00043553506653993597,
      "loss": 2.6456,
      "step": 1210
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.530094160468025,
      "learning_rate": 0.0004347982950377992,
      "loss": 2.7032,
      "step": 1211
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5136700628059662,
      "learning_rate": 0.0004340616675310735,
      "loss": 2.73,
      "step": 1212
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5234312619642245,
      "learning_rate": 0.00043332518564657193,
      "loss": 2.7077,
      "step": 1213
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5549987779837912,
      "learning_rate": 0.0004325888510107856,
      "loss": 2.7254,
      "step": 1214
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.536956217722958,
      "learning_rate": 0.0004318526652498809,
      "loss": 2.6648,
      "step": 1215
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5266642997036883,
      "learning_rate": 0.00043111662998969523,
      "loss": 2.6836,
      "step": 1216
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5205419535470811,
      "learning_rate": 0.0004303807468557335,
      "loss": 2.7394,
      "step": 1217
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5162685973116814,
      "learning_rate": 0.0004296450174731648,
      "loss": 2.6234,
      "step": 1218
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.47005263283225,
      "learning_rate": 0.0004289094434668188,
      "loss": 2.7119,
      "step": 1219
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.4856691255220598,
      "learning_rate": 0.00042817402646118185,
      "loss": 2.5816,
      "step": 1220
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5175805949037477,
      "learning_rate": 0.0004274387680803936,
      "loss": 2.7353,
      "step": 1221
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5211854619477375,
      "learning_rate": 0.00042670366994824327,
      "loss": 2.7057,
      "step": 1222
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5238526713577384,
      "learning_rate": 0.0004259687336881663,
      "loss": 2.7197,
      "step": 1223
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5130864930776108,
      "learning_rate": 0.0004252339609232408,
      "loss": 2.7056,
      "step": 1224
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.4709359184046822,
      "learning_rate": 0.0004244993532761834,
      "loss": 2.6671,
      "step": 1225
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5006742495645969,
      "learning_rate": 0.00042376491236934634,
      "loss": 2.6794,
      "step": 1226
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5306017315906422,
      "learning_rate": 0.0004230306398247136,
      "loss": 2.7454,
      "step": 1227
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.4902486378456207,
      "learning_rate": 0.0004222965372638976,
      "loss": 2.7657,
      "step": 1228
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.48884495835320135,
      "learning_rate": 0.0004215626063081348,
      "loss": 2.7078,
      "step": 1229
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5369922979048567,
      "learning_rate": 0.000420828848578283,
      "loss": 2.6114,
      "step": 1230
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5651909198412906,
      "learning_rate": 0.0004200952656948175,
      "loss": 2.705,
      "step": 1231
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5091114658115851,
      "learning_rate": 0.0004193618592778272,
      "loss": 2.763,
      "step": 1232
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5391752109391068,
      "learning_rate": 0.0004186286309470116,
      "loss": 2.7214,
      "step": 1233
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5266451505884581,
      "learning_rate": 0.0004178955823216767,
      "loss": 2.6838,
      "step": 1234
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5079695365655955,
      "learning_rate": 0.00041716271502073137,
      "loss": 2.6945,
      "step": 1235
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5115712255812047,
      "learning_rate": 0.000416430030662685,
      "loss": 2.6174,
      "step": 1236
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.47719038433776967,
      "learning_rate": 0.00041569753086564173,
      "loss": 2.615,
      "step": 1237
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.48504822455139973,
      "learning_rate": 0.0004149652172472988,
      "loss": 2.6736,
      "step": 1238
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5027945111144037,
      "learning_rate": 0.00041423309142494234,
      "loss": 2.7765,
      "step": 1239
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5087264453689427,
      "learning_rate": 0.0004135011550154433,
      "loss": 2.6747,
      "step": 1240
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.4963248329607529,
      "learning_rate": 0.0004127694096352546,
      "loss": 2.6672,
      "step": 1241
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.49245570308527503,
      "learning_rate": 0.00041203785690040743,
      "loss": 2.6622,
      "step": 1242
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.511324616946125,
      "learning_rate": 0.00041130649842650694,
      "loss": 2.7645,
      "step": 1243
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.506264914707819,
      "learning_rate": 0.00041057533582873016,
      "loss": 2.5778,
      "step": 1244
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5247516709137736,
      "learning_rate": 0.0004098443707218208,
      "loss": 2.7934,
      "step": 1245
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5121379808649518,
      "learning_rate": 0.00040911360472008673,
      "loss": 2.6048,
      "step": 1246
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5213086279626764,
      "learning_rate": 0.0004083830394373959,
      "loss": 2.5874,
      "step": 1247
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5002438181372962,
      "learning_rate": 0.00040765267648717324,
      "loss": 2.6725,
      "step": 1248
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5331152391687023,
      "learning_rate": 0.00040692251748239677,
      "loss": 2.7529,
      "step": 1249
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.4922541965967873,
      "learning_rate": 0.00040619256403559383,
      "loss": 2.6033,
      "step": 1250
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5153929595123157,
      "learning_rate": 0.000405462817758838,
      "loss": 2.718,
      "step": 1251
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5362437889613196,
      "learning_rate": 0.0004047332802637457,
      "loss": 2.6772,
      "step": 1252
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.537617974282814,
      "learning_rate": 0.00040400395316147157,
      "loss": 2.6029,
      "step": 1253
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5302414105894713,
      "learning_rate": 0.00040327483806270627,
      "loss": 2.7248,
      "step": 1254
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5411024000098807,
      "learning_rate": 0.0004025459365776715,
      "loss": 2.7704,
      "step": 1255
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5108937202580804,
      "learning_rate": 0.00040181725031611794,
      "loss": 2.6939,
      "step": 1256
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5417170269937848,
      "learning_rate": 0.0004010887808873206,
      "loss": 2.5715,
      "step": 1257
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5265344253058065,
      "learning_rate": 0.00040036052990007553,
      "loss": 2.7045,
      "step": 1258
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.49242870611440315,
      "learning_rate": 0.0003996324989626967,
      "loss": 2.7156,
      "step": 1259
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5262843821454762,
      "learning_rate": 0.00039890468968301166,
      "loss": 2.6944,
      "step": 1260
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5062871500981689,
      "learning_rate": 0.0003981771036683591,
      "loss": 2.688,
      "step": 1261
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5613320881888487,
      "learning_rate": 0.00039744974252558385,
      "loss": 2.6829,
      "step": 1262
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.528478583718188,
      "learning_rate": 0.00039672260786103463,
      "loss": 2.6247,
      "step": 1263
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.4864969756943371,
      "learning_rate": 0.00039599570128055994,
      "loss": 2.6176,
      "step": 1264
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5663985879252396,
      "learning_rate": 0.0003952690243895044,
      "loss": 2.8156,
      "step": 1265
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5275365590680847,
      "learning_rate": 0.0003945425787927054,
      "loss": 2.7165,
      "step": 1266
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5493667599593538,
      "learning_rate": 0.00039381636609448975,
      "loss": 2.6462,
      "step": 1267
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5611665351308049,
      "learning_rate": 0.0003930903878986693,
      "loss": 2.6689,
      "step": 1268
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5404360719910868,
      "learning_rate": 0.00039236464580853916,
      "loss": 2.6527,
      "step": 1269
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.505174874369067,
      "learning_rate": 0.0003916391414268718,
      "loss": 2.6826,
      "step": 1270
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5639900241376554,
      "learning_rate": 0.00039091387635591536,
      "loss": 2.7014,
      "step": 1271
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.49449134587345966,
      "learning_rate": 0.0003901888521973894,
      "loss": 2.5968,
      "step": 1272
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5143663600475958,
      "learning_rate": 0.0003894640705524813,
      "loss": 2.684,
      "step": 1273
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.48680340727192584,
      "learning_rate": 0.00038873953302184284,
      "loss": 2.5858,
      "step": 1274
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5056531725024779,
      "learning_rate": 0.000388015241205587,
      "loss": 2.668,
      "step": 1275
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.522206055277458,
      "learning_rate": 0.00038729119670328355,
      "loss": 2.6914,
      "step": 1276
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.4971525583490198,
      "learning_rate": 0.00038656740111395665,
      "loss": 2.6147,
      "step": 1277
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.47069233433337143,
      "learning_rate": 0.00038584385603608053,
      "loss": 2.6734,
      "step": 1278
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.48432379552169946,
      "learning_rate": 0.00038512056306757615,
      "loss": 2.5903,
      "step": 1279
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.506206348808886,
      "learning_rate": 0.0003843975238058075,
      "loss": 2.6175,
      "step": 1280
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.545245898521085,
      "learning_rate": 0.00038367473984757863,
      "loss": 2.6534,
      "step": 1281
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.49604806389875744,
      "learning_rate": 0.0003829522127891296,
      "loss": 2.6956,
      "step": 1282
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5158270886662941,
      "learning_rate": 0.0003822299442261329,
      "loss": 2.6417,
      "step": 1283
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5059214299850134,
      "learning_rate": 0.00038150793575369063,
      "loss": 2.675,
      "step": 1284
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5224810009016285,
      "learning_rate": 0.0003807861889663299,
      "loss": 2.7552,
      "step": 1285
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5195794970194385,
      "learning_rate": 0.0003800647054580006,
      "loss": 2.6491,
      "step": 1286
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5148188704483899,
      "learning_rate": 0.00037934348682207064,
      "loss": 2.695,
      "step": 1287
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5278647166587669,
      "learning_rate": 0.00037862253465132306,
      "loss": 2.6972,
      "step": 1288
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.4892436913639856,
      "learning_rate": 0.00037790185053795245,
      "loss": 2.7493,
      "step": 1289
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5352080394436993,
      "learning_rate": 0.0003771814360735616,
      "loss": 2.7055,
      "step": 1290
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5373346606741812,
      "learning_rate": 0.00037646129284915755,
      "loss": 2.6753,
      "step": 1291
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5249264201475462,
      "learning_rate": 0.00037574142245514825,
      "loss": 2.7573,
      "step": 1292
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.48280529815600787,
      "learning_rate": 0.0003750218264813393,
      "loss": 2.7079,
      "step": 1293
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5196471508521048,
      "learning_rate": 0.0003743025065169305,
      "loss": 2.6802,
      "step": 1294
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.4964173378444257,
      "learning_rate": 0.0003735834641505116,
      "loss": 2.6752,
      "step": 1295
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5162613483990264,
      "learning_rate": 0.00037286470097005954,
      "loss": 2.7268,
      "step": 1296
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5018651588395292,
      "learning_rate": 0.0003721462185629347,
      "loss": 2.7103,
      "step": 1297
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5507834983005782,
      "learning_rate": 0.00037142801851587707,
      "loss": 2.6567,
      "step": 1298
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5154756264521153,
      "learning_rate": 0.00037071010241500357,
      "loss": 2.7171,
      "step": 1299
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5029247817534002,
      "learning_rate": 0.00036999247184580383,
      "loss": 2.6644,
      "step": 1300
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.50994069274581,
      "learning_rate": 0.00036927512839313636,
      "loss": 2.6381,
      "step": 1301
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.536276373105891,
      "learning_rate": 0.0003685580736412268,
      "loss": 2.67,
      "step": 1302
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5475948271066223,
      "learning_rate": 0.000367841309173662,
      "loss": 2.5782,
      "step": 1303
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.49993086294162664,
      "learning_rate": 0.0003671248365733883,
      "loss": 2.6694,
      "step": 1304
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5278557948829261,
      "learning_rate": 0.0003664086574227075,
      "loss": 2.5884,
      "step": 1305
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5540280963053137,
      "learning_rate": 0.000365692773303273,
      "loss": 2.6333,
      "step": 1306
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5372922526398834,
      "learning_rate": 0.00036497718579608696,
      "loss": 2.6449,
      "step": 1307
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5214245722065235,
      "learning_rate": 0.0003642618964814964,
      "loss": 2.6123,
      "step": 1308
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5352511013531179,
      "learning_rate": 0.00036354690693918946,
      "loss": 2.6486,
      "step": 1309
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5453198872038278,
      "learning_rate": 0.00036283221874819284,
      "loss": 2.7267,
      "step": 1310
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5472038402482459,
      "learning_rate": 0.0003621178334868672,
      "loss": 2.6638,
      "step": 1311
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5259870423086725,
      "learning_rate": 0.00036140375273290476,
      "loss": 2.7166,
      "step": 1312
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5231931408709464,
      "learning_rate": 0.0003606899780633245,
      "loss": 2.6994,
      "step": 1313
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5223921064242383,
      "learning_rate": 0.0003599765110544699,
      "loss": 2.722,
      "step": 1314
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5271834978796934,
      "learning_rate": 0.0003592633532820052,
      "loss": 2.681,
      "step": 1315
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5082674225390232,
      "learning_rate": 0.0003585505063209109,
      "loss": 2.6802,
      "step": 1316
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.4986074496910536,
      "learning_rate": 0.00035783797174548194,
      "loss": 2.6523,
      "step": 1317
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5051620073045062,
      "learning_rate": 0.00035712575112932277,
      "loss": 2.6989,
      "step": 1318
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5160093602678967,
      "learning_rate": 0.000356413846045345,
      "loss": 2.6912,
      "step": 1319
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5213383514155379,
      "learning_rate": 0.000355702258065763,
      "loss": 2.584,
      "step": 1320
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5340990832988344,
      "learning_rate": 0.0003549909887620909,
      "loss": 2.6896,
      "step": 1321
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5067360663370191,
      "learning_rate": 0.00035428003970513914,
      "loss": 2.6335,
      "step": 1322
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5079787276852579,
      "learning_rate": 0.00035356941246501085,
      "loss": 2.6712,
      "step": 1323
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.485577544885249,
      "learning_rate": 0.0003528591086110984,
      "loss": 2.6764,
      "step": 1324
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5382166800900163,
      "learning_rate": 0.00035214912971208,
      "loss": 2.7166,
      "step": 1325
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5438082787138756,
      "learning_rate": 0.0003514394773359163,
      "loss": 2.6632,
      "step": 1326
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5175385265452299,
      "learning_rate": 0.0003507301530498469,
      "loss": 2.7192,
      "step": 1327
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5447481287357926,
      "learning_rate": 0.00035002115842038646,
      "loss": 2.734,
      "step": 1328
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.536057808840742,
      "learning_rate": 0.00034931249501332195,
      "loss": 2.7175,
      "step": 1329
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5099243388229495,
      "learning_rate": 0.00034860416439370885,
      "loss": 2.7495,
      "step": 1330
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5319499433274613,
      "learning_rate": 0.0003478961681258674,
      "loss": 2.7375,
      "step": 1331
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5339495623948575,
      "learning_rate": 0.0003471885077733796,
      "loss": 2.6261,
      "step": 1332
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5046021856215106,
      "learning_rate": 0.0003464811848990859,
      "loss": 2.7132,
      "step": 1333
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5003942937815565,
      "learning_rate": 0.00034577420106508063,
      "loss": 2.6119,
      "step": 1334
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5299897735958715,
      "learning_rate": 0.0003450675578327105,
      "loss": 2.7272,
      "step": 1335
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5343157362738072,
      "learning_rate": 0.000344361256762569,
      "loss": 2.6606,
      "step": 1336
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5244329880484088,
      "learning_rate": 0.00034365529941449456,
      "loss": 2.7595,
      "step": 1337
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.49156570465653665,
      "learning_rate": 0.0003429496873475664,
      "loss": 2.6735,
      "step": 1338
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.4999738343605983,
      "learning_rate": 0.0003422444221201009,
      "loss": 2.6461,
      "step": 1339
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5233411077451025,
      "learning_rate": 0.0003415395052896487,
      "loss": 2.6833,
      "step": 1340
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5219460272096302,
      "learning_rate": 0.0003408349384129912,
      "loss": 2.6299,
      "step": 1341
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.6077928523510566,
      "learning_rate": 0.00034013072304613643,
      "loss": 2.7924,
      "step": 1342
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5146843252289197,
      "learning_rate": 0.00033942686074431674,
      "loss": 2.7318,
      "step": 1343
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.522018439418992,
      "learning_rate": 0.0003387233530619843,
      "loss": 2.6913,
      "step": 1344
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5101831099857319,
      "learning_rate": 0.0003380202015528084,
      "loss": 2.5946,
      "step": 1345
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.46660384265215094,
      "learning_rate": 0.0003373174077696715,
      "loss": 2.6945,
      "step": 1346
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.516312427764236,
      "learning_rate": 0.0003366149732646661,
      "loss": 2.6382,
      "step": 1347
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5095199457046187,
      "learning_rate": 0.00033591289958909143,
      "loss": 2.6688,
      "step": 1348
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.4885245746572849,
      "learning_rate": 0.00033521118829344954,
      "loss": 2.5742,
      "step": 1349
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.521895943365929,
      "learning_rate": 0.0003345098409274423,
      "loss": 2.7726,
      "step": 1350
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5113189581344126,
      "learning_rate": 0.00033380885903996796,
      "loss": 2.6056,
      "step": 1351
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5521376184530942,
      "learning_rate": 0.00033310824417911766,
      "loss": 2.6662,
      "step": 1352
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.526853057352678,
      "learning_rate": 0.00033240799789217184,
      "loss": 2.7882,
      "step": 1353
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5413493777020469,
      "learning_rate": 0.00033170812172559694,
      "loss": 2.7082,
      "step": 1354
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5165695197844109,
      "learning_rate": 0.000331008617225042,
      "loss": 2.6602,
      "step": 1355
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5193692697977721,
      "learning_rate": 0.0003303094859353355,
      "loss": 2.6117,
      "step": 1356
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5255952757460676,
      "learning_rate": 0.0003296107294004812,
      "loss": 2.6354,
      "step": 1357
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5433918967408917,
      "learning_rate": 0.0003289123491636559,
      "loss": 2.5422,
      "step": 1358
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5341109023350894,
      "learning_rate": 0.00032821434676720443,
      "loss": 2.5804,
      "step": 1359
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.529402305940971,
      "learning_rate": 0.00032751672375263836,
      "loss": 2.6891,
      "step": 1360
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5016619793424085,
      "learning_rate": 0.0003268194816606305,
      "loss": 2.6179,
      "step": 1361
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.5657375052996718,
      "learning_rate": 0.00032612262203101267,
      "loss": 2.6414,
      "step": 1362
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5422066378923954,
      "learning_rate": 0.00032542614640277225,
      "loss": 2.6076,
      "step": 1363
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5287488249105601,
      "learning_rate": 0.0003247300563140481,
      "loss": 2.6615,
      "step": 1364
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5819683575763082,
      "learning_rate": 0.00032403435330212807,
      "loss": 2.7259,
      "step": 1365
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5475986305843664,
      "learning_rate": 0.00032333903890344515,
      "loss": 2.686,
      "step": 1366
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4935288695787895,
      "learning_rate": 0.00032264411465357333,
      "loss": 2.5674,
      "step": 1367
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5262052352606102,
      "learning_rate": 0.00032194958208722654,
      "loss": 2.6771,
      "step": 1368
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5445183733396111,
      "learning_rate": 0.00032125544273825204,
      "loss": 2.7843,
      "step": 1369
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5238394369182566,
      "learning_rate": 0.0003205616981396297,
      "loss": 2.6502,
      "step": 1370
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4977683537022086,
      "learning_rate": 0.00031986834982346713,
      "loss": 2.7124,
      "step": 1371
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4874506234474769,
      "learning_rate": 0.00031917539932099694,
      "loss": 2.6795,
      "step": 1372
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5023626411399686,
      "learning_rate": 0.00031848284816257336,
      "loss": 2.7011,
      "step": 1373
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5426472166234935,
      "learning_rate": 0.0003177906978776682,
      "loss": 2.6814,
      "step": 1374
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5642880016533615,
      "learning_rate": 0.0003170989499948683,
      "loss": 2.7464,
      "step": 1375
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5157116294003709,
      "learning_rate": 0.0003164076060418719,
      "loss": 2.6379,
      "step": 1376
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5186909042468566,
      "learning_rate": 0.000315716667545485,
      "loss": 2.6998,
      "step": 1377
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4952085663608333,
      "learning_rate": 0.00031502613603161836,
      "loss": 2.6487,
      "step": 1378
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5850830319130214,
      "learning_rate": 0.00031433601302528335,
      "loss": 2.6798,
      "step": 1379
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5492048047397665,
      "learning_rate": 0.00031364630005058995,
      "loss": 2.627,
      "step": 1380
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5335190534299099,
      "learning_rate": 0.0003129569986307422,
      "loss": 2.6604,
      "step": 1381
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5066734226947467,
      "learning_rate": 0.00031226811028803515,
      "loss": 2.6743,
      "step": 1382
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5138738632470816,
      "learning_rate": 0.00031157963654385173,
      "loss": 2.638,
      "step": 1383
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4956355925211338,
      "learning_rate": 0.0003108915789186592,
      "loss": 2.7389,
      "step": 1384
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5208069789695808,
      "learning_rate": 0.00031020393893200604,
      "loss": 2.7045,
      "step": 1385
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5316353965257978,
      "learning_rate": 0.00030951671810251823,
      "loss": 2.7065,
      "step": 1386
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5260349868498675,
      "learning_rate": 0.0003088299179478959,
      "loss": 2.6821,
      "step": 1387
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5042055701627023,
      "learning_rate": 0.0003081435399849104,
      "loss": 2.6329,
      "step": 1388
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.564684890477677,
      "learning_rate": 0.0003074575857294004,
      "loss": 2.7557,
      "step": 1389
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5103677357724797,
      "learning_rate": 0.0003067720566962691,
      "loss": 2.6885,
      "step": 1390
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5740174553771336,
      "learning_rate": 0.0003060869543994806,
      "loss": 2.6182,
      "step": 1391
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5738561979214507,
      "learning_rate": 0.0003054022803520562,
      "loss": 2.6612,
      "step": 1392
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5372126198016173,
      "learning_rate": 0.0003047180360660721,
      "loss": 2.6364,
      "step": 1393
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.501160716823567,
      "learning_rate": 0.00030403422305265475,
      "loss": 2.6712,
      "step": 1394
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5114866334109534,
      "learning_rate": 0.0003033508428219785,
      "loss": 2.6873,
      "step": 1395
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5232681840541096,
      "learning_rate": 0.00030266789688326184,
      "loss": 2.5691,
      "step": 1396
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5393814668125796,
      "learning_rate": 0.00030198538674476393,
      "loss": 2.6997,
      "step": 1397
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5967989026102972,
      "learning_rate": 0.00030130331391378185,
      "loss": 2.7028,
      "step": 1398
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5600477146014537,
      "learning_rate": 0.0003006216798966468,
      "loss": 2.763,
      "step": 1399
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5654385397756314,
      "learning_rate": 0.00029994048619872034,
      "loss": 2.693,
      "step": 1400
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5299226380316983,
      "learning_rate": 0.0002992597343243927,
      "loss": 2.6608,
      "step": 1401
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5332431168796034,
      "learning_rate": 0.0002985794257770773,
      "loss": 2.6744,
      "step": 1402
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5599907413920047,
      "learning_rate": 0.0002978995620592092,
      "loss": 2.7762,
      "step": 1403
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5259219784993198,
      "learning_rate": 0.0002972201446722405,
      "loss": 2.6704,
      "step": 1404
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5289110190907368,
      "learning_rate": 0.00029654117511663803,
      "loss": 2.6691,
      "step": 1405
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5666789751206317,
      "learning_rate": 0.0002958626548918795,
      "loss": 2.6319,
      "step": 1406
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.537924097368643,
      "learning_rate": 0.00029518458549645014,
      "loss": 2.6829,
      "step": 1407
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5250651072830079,
      "learning_rate": 0.00029450696842783954,
      "loss": 2.5617,
      "step": 1408
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5413987082913225,
      "learning_rate": 0.00029382980518253865,
      "loss": 2.6939,
      "step": 1409
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5198850495320728,
      "learning_rate": 0.00029315309725603595,
      "loss": 2.7114,
      "step": 1410
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5205726163557534,
      "learning_rate": 0.00029247684614281446,
      "loss": 2.6288,
      "step": 1411
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5066031300184763,
      "learning_rate": 0.0002918010533363481,
      "loss": 2.6293,
      "step": 1412
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5052016588684003,
      "learning_rate": 0.0002911257203290987,
      "loss": 2.6302,
      "step": 1413
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.4830621053248146,
      "learning_rate": 0.00029045084861251314,
      "loss": 2.6059,
      "step": 1414
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5105135117228916,
      "learning_rate": 0.00028977643967701897,
      "loss": 2.7091,
      "step": 1415
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5181454402780247,
      "learning_rate": 0.00028910249501202156,
      "loss": 2.6829,
      "step": 1416
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.49973317100856485,
      "learning_rate": 0.00028842901610590165,
      "loss": 2.5842,
      "step": 1417
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5073894080741461,
      "learning_rate": 0.00028775600444601123,
      "loss": 2.6679,
      "step": 1418
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5437132557995807,
      "learning_rate": 0.00028708346151866973,
      "loss": 2.6164,
      "step": 1419
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.4985011262072265,
      "learning_rate": 0.0002864113888091622,
      "loss": 2.6482,
      "step": 1420
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.526944767280008,
      "learning_rate": 0.0002857397878017348,
      "loss": 2.6875,
      "step": 1421
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5203838861769262,
      "learning_rate": 0.00028506865997959173,
      "loss": 2.6779,
      "step": 1422
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5102656749515754,
      "learning_rate": 0.000284398006824893,
      "loss": 2.5647,
      "step": 1423
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.520395819567321,
      "learning_rate": 0.00028372782981874963,
      "loss": 2.6876,
      "step": 1424
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.49829621726395185,
      "learning_rate": 0.00028305813044122096,
      "loss": 2.6435,
      "step": 1425
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5440936122894126,
      "learning_rate": 0.0002823889101713122,
      "loss": 2.6147,
      "step": 1426
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5764556440051147,
      "learning_rate": 0.0002817201704869701,
      "loss": 2.7322,
      "step": 1427
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5362159401223764,
      "learning_rate": 0.00028105191286508,
      "loss": 2.6486,
      "step": 1428
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5466499533987904,
      "learning_rate": 0.00028038413878146245,
      "loss": 2.7046,
      "step": 1429
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5132498804391473,
      "learning_rate": 0.00027971684971087073,
      "loss": 2.6387,
      "step": 1430
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5150065927763824,
      "learning_rate": 0.00027905004712698643,
      "loss": 2.5806,
      "step": 1431
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5060318307730233,
      "learning_rate": 0.0002783837325024167,
      "loss": 2.6609,
      "step": 1432
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5175130911221689,
      "learning_rate": 0.00027771790730869153,
      "loss": 2.6201,
      "step": 1433
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5127763109143173,
      "learning_rate": 0.0002770525730162599,
      "loss": 2.6918,
      "step": 1434
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5101713640524365,
      "learning_rate": 0.00027638773109448645,
      "loss": 2.6363,
      "step": 1435
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.4977353453449889,
      "learning_rate": 0.00027572338301164824,
      "loss": 2.6749,
      "step": 1436
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5152001930798664,
      "learning_rate": 0.0002750595302349324,
      "loss": 2.7359,
      "step": 1437
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5418196502338989,
      "learning_rate": 0.00027439617423043145,
      "loss": 2.6405,
      "step": 1438
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5202765834346041,
      "learning_rate": 0.00027373331646314114,
      "loss": 2.7226,
      "step": 1439
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5397980586715199,
      "learning_rate": 0.0002730709583969572,
      "loss": 2.6664,
      "step": 1440
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5043474054028523,
      "learning_rate": 0.0002724091014946711,
      "loss": 2.5438,
      "step": 1441
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.4946802793438189,
      "learning_rate": 0.00027174774721796824,
      "loss": 2.6366,
      "step": 1442
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5417840765906496,
      "learning_rate": 0.0002710868970274232,
      "loss": 2.6567,
      "step": 1443
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5489253642537957,
      "learning_rate": 0.0002704265523824982,
      "loss": 2.59,
      "step": 1444
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5803818513629694,
      "learning_rate": 0.00026976671474153826,
      "loss": 2.5948,
      "step": 1445
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5517370509149084,
      "learning_rate": 0.00026910738556176886,
      "loss": 2.7432,
      "step": 1446
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5135361314804905,
      "learning_rate": 0.0002684485662992929,
      "loss": 2.6115,
      "step": 1447
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5337181086537189,
      "learning_rate": 0.0002677902584090869,
      "loss": 2.5381,
      "step": 1448
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5610967536723442,
      "learning_rate": 0.00026713246334499774,
      "loss": 2.696,
      "step": 1449
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.564343060737695,
      "learning_rate": 0.00026647518255974023,
      "loss": 2.5733,
      "step": 1450
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5527781937344152,
      "learning_rate": 0.0002658184175048934,
      "loss": 2.6563,
      "step": 1451
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5421097229014369,
      "learning_rate": 0.00026516216963089694,
      "loss": 2.6844,
      "step": 1452
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5471746344692654,
      "learning_rate": 0.0002645064403870488,
      "loss": 2.6965,
      "step": 1453
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5376147689259011,
      "learning_rate": 0.0002638512312215011,
      "loss": 2.6935,
      "step": 1454
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5189339406840692,
      "learning_rate": 0.0002631965435812575,
      "loss": 2.6773,
      "step": 1455
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5518657545246427,
      "learning_rate": 0.00026254237891217046,
      "loss": 2.7785,
      "step": 1456
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5298639828296662,
      "learning_rate": 0.0002618887386589367,
      "loss": 2.6014,
      "step": 1457
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.510199465396997,
      "learning_rate": 0.0002612356242650949,
      "loss": 2.6573,
      "step": 1458
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5036444576831944,
      "learning_rate": 0.0002605830371730229,
      "loss": 2.6739,
      "step": 1459
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5542983995400107,
      "learning_rate": 0.0002599309788239339,
      "loss": 2.6603,
      "step": 1460
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5092107088269644,
      "learning_rate": 0.00025927945065787306,
      "loss": 2.6108,
      "step": 1461
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5064272689216358,
      "learning_rate": 0.0002586284541137145,
      "loss": 2.6153,
      "step": 1462
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5180283883834162,
      "learning_rate": 0.00025797799062915905,
      "loss": 2.5744,
      "step": 1463
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5055843608166704,
      "learning_rate": 0.00025732806164072966,
      "loss": 2.6848,
      "step": 1464
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5185821371538296,
      "learning_rate": 0.00025667866858376874,
      "loss": 2.5959,
      "step": 1465
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5176633431516448,
      "learning_rate": 0.0002560298128924358,
      "loss": 2.6282,
      "step": 1466
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5273843104219313,
      "learning_rate": 0.0002553814959997032,
      "loss": 2.7327,
      "step": 1467
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5711054615423432,
      "learning_rate": 0.00025473371933735334,
      "loss": 2.6961,
      "step": 1468
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5383704906387032,
      "learning_rate": 0.00025408648433597534,
      "loss": 2.6674,
      "step": 1469
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5427733271080686,
      "learning_rate": 0.00025343979242496283,
      "loss": 2.6195,
      "step": 1470
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5365929833764801,
      "learning_rate": 0.00025279364503250925,
      "loss": 2.6737,
      "step": 1471
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5115049533560345,
      "learning_rate": 0.0002521480435856056,
      "loss": 2.6393,
      "step": 1472
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5193419047301462,
      "learning_rate": 0.0002515029895100378,
      "loss": 2.5748,
      "step": 1473
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5349163495852879,
      "learning_rate": 0.0002508584842303822,
      "loss": 2.664,
      "step": 1474
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5231080571453198,
      "learning_rate": 0.0002502145291700038,
      "loss": 2.5552,
      "step": 1475
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5680389782159602,
      "learning_rate": 0.0002495711257510517,
      "loss": 2.7241,
      "step": 1476
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5002819969270994,
      "learning_rate": 0.0002489282753944575,
      "loss": 2.5844,
      "step": 1477
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5412066763071609,
      "learning_rate": 0.00024828597951993093,
      "loss": 2.6433,
      "step": 1478
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5676842560351678,
      "learning_rate": 0.00024764423954595706,
      "loss": 2.6848,
      "step": 1479
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5906613736610669,
      "learning_rate": 0.0002470030568897938,
      "loss": 2.6737,
      "step": 1480
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5673379624383489,
      "learning_rate": 0.00024636243296746773,
      "loss": 2.6745,
      "step": 1481
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5206353450569106,
      "learning_rate": 0.0002457223691937716,
      "loss": 2.6633,
      "step": 1482
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5160386604266843,
      "learning_rate": 0.0002450828669822613,
      "loss": 2.6439,
      "step": 1483
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5533087453045891,
      "learning_rate": 0.00024444392774525253,
      "loss": 2.7156,
      "step": 1484
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5465306666046615,
      "learning_rate": 0.00024380555289381733,
      "loss": 2.5933,
      "step": 1485
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.542473365369578,
      "learning_rate": 0.00024316774383778184,
      "loss": 2.6647,
      "step": 1486
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5237270013241679,
      "learning_rate": 0.0002425305019857222,
      "loss": 2.6483,
      "step": 1487
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5193715169276047,
      "learning_rate": 0.00024189382874496184,
      "loss": 2.6638,
      "step": 1488
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5897079369245194,
      "learning_rate": 0.00024125772552156916,
      "loss": 2.7208,
      "step": 1489
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5139682715458571,
      "learning_rate": 0.00024062219372035292,
      "loss": 2.6786,
      "step": 1490
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5179543000300554,
      "learning_rate": 0.00023998723474486007,
      "loss": 2.6111,
      "step": 1491
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5347407856165681,
      "learning_rate": 0.00023935284999737272,
      "loss": 2.6616,
      "step": 1492
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5272279684299376,
      "learning_rate": 0.00023871904087890505,
      "loss": 2.7029,
      "step": 1493
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5234490480223335,
      "learning_rate": 0.00023808580878919945,
      "loss": 2.6136,
      "step": 1494
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5582945697087572,
      "learning_rate": 0.00023745315512672398,
      "loss": 2.7117,
      "step": 1495
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5231832324951284,
      "learning_rate": 0.0002368210812886698,
      "loss": 2.6618,
      "step": 1496
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5616947979265463,
      "learning_rate": 0.0002361895886709471,
      "loss": 2.6816,
      "step": 1497
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5185055761386351,
      "learning_rate": 0.0002355586786681823,
      "loss": 2.608,
      "step": 1498
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.539317601724139,
      "learning_rate": 0.00023492835267371575,
      "loss": 2.6858,
      "step": 1499
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5345901343877524,
      "learning_rate": 0.0002342986120795978,
      "loss": 2.6481,
      "step": 1500
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5122342904163094,
      "learning_rate": 0.0002336694582765857,
      "loss": 2.5651,
      "step": 1501
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5196213534016761,
      "learning_rate": 0.00023304089265414085,
      "loss": 2.6116,
      "step": 1502
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5155030891507187,
      "learning_rate": 0.00023241291660042613,
      "loss": 2.6156,
      "step": 1503
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5729630278348291,
      "learning_rate": 0.00023178553150230186,
      "loss": 2.7175,
      "step": 1504
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5594060944325331,
      "learning_rate": 0.00023115873874532324,
      "loss": 2.713,
      "step": 1505
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5277105104636358,
      "learning_rate": 0.00023053253971373796,
      "loss": 2.654,
      "step": 1506
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5162148271146483,
      "learning_rate": 0.00022990693579048166,
      "loss": 2.5876,
      "step": 1507
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5300984993351018,
      "learning_rate": 0.00022928192835717644,
      "loss": 2.602,
      "step": 1508
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.513353442934914,
      "learning_rate": 0.00022865751879412634,
      "loss": 2.676,
      "step": 1509
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5097239768653745,
      "learning_rate": 0.00022803370848031585,
      "loss": 2.6202,
      "step": 1510
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5456174738564598,
      "learning_rate": 0.00022741049879340542,
      "loss": 2.6358,
      "step": 1511
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5484806040785072,
      "learning_rate": 0.00022678789110972897,
      "loss": 2.5856,
      "step": 1512
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5546216675903627,
      "learning_rate": 0.00022616588680429155,
      "loss": 2.6514,
      "step": 1513
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5617218968324993,
      "learning_rate": 0.00022554448725076526,
      "loss": 2.7548,
      "step": 1514
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.569613187708911,
      "learning_rate": 0.0002249236938214863,
      "loss": 2.5341,
      "step": 1515
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5501106307012626,
      "learning_rate": 0.00022430350788745296,
      "loss": 2.5603,
      "step": 1516
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5390219955294513,
      "learning_rate": 0.00022368393081832166,
      "loss": 2.69,
      "step": 1517
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5897811537094955,
      "learning_rate": 0.00022306496398240383,
      "loss": 2.6767,
      "step": 1518
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5347395112834396,
      "learning_rate": 0.00022244660874666373,
      "loss": 2.5479,
      "step": 1519
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.532099024339146,
      "learning_rate": 0.00022182886647671452,
      "loss": 2.6148,
      "step": 1520
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5197008526818121,
      "learning_rate": 0.0002212117385368157,
      "loss": 2.5834,
      "step": 1521
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.525993286414334,
      "learning_rate": 0.00022059522628987038,
      "loss": 2.7812,
      "step": 1522
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5283498413141905,
      "learning_rate": 0.00021997933109742162,
      "loss": 2.6849,
      "step": 1523
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5050910637464837,
      "learning_rate": 0.00021936405431964969,
      "loss": 2.5724,
      "step": 1524
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5250402825442426,
      "learning_rate": 0.00021874939731536926,
      "loss": 2.6456,
      "step": 1525
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5262360172728192,
      "learning_rate": 0.00021813536144202656,
      "loss": 2.6649,
      "step": 1526
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5386503618407155,
      "learning_rate": 0.00021752194805569553,
      "loss": 2.6524,
      "step": 1527
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5217466893978039,
      "learning_rate": 0.0002169091585110754,
      "loss": 2.6671,
      "step": 1528
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5400829363920241,
      "learning_rate": 0.00021629699416148828,
      "loss": 2.6783,
      "step": 1529
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5294514013487223,
      "learning_rate": 0.000215685456358875,
      "loss": 2.6293,
      "step": 1530
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5694555506648116,
      "learning_rate": 0.00021507454645379258,
      "loss": 2.5874,
      "step": 1531
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5265427964792164,
      "learning_rate": 0.00021446426579541184,
      "loss": 2.6416,
      "step": 1532
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5149494502222681,
      "learning_rate": 0.00021385461573151387,
      "loss": 2.6493,
      "step": 1533
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5832936251131656,
      "learning_rate": 0.00021324559760848677,
      "loss": 2.7218,
      "step": 1534
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5657731206716266,
      "learning_rate": 0.00021263721277132303,
      "loss": 2.5902,
      "step": 1535
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5242313701067509,
      "learning_rate": 0.0002120294625636171,
      "loss": 2.5691,
      "step": 1536
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5433520498868684,
      "learning_rate": 0.0002114223483275613,
      "loss": 2.6662,
      "step": 1537
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5541520795803548,
      "learning_rate": 0.0002108158714039435,
      "loss": 2.5679,
      "step": 1538
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5920854297181153,
      "learning_rate": 0.00021021003313214455,
      "loss": 2.6702,
      "step": 1539
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5346900369551251,
      "learning_rate": 0.00020960483485013432,
      "loss": 2.6517,
      "step": 1540
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5643621486573619,
      "learning_rate": 0.0002090002778944694,
      "loss": 2.6152,
      "step": 1541
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.567044693530176,
      "learning_rate": 0.00020839636360029025,
      "loss": 2.6198,
      "step": 1542
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.577927786948691,
      "learning_rate": 0.00020779309330131818,
      "loss": 2.5774,
      "step": 1543
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5719137133497179,
      "learning_rate": 0.00020719046832985184,
      "loss": 2.8123,
      "step": 1544
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5180931132175836,
      "learning_rate": 0.0002065884900167646,
      "loss": 2.6041,
      "step": 1545
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5224195494034202,
      "learning_rate": 0.0002059871596915024,
      "loss": 2.6042,
      "step": 1546
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5288820415224451,
      "learning_rate": 0.0002053864786820795,
      "loss": 2.7324,
      "step": 1547
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.48909132475895206,
      "learning_rate": 0.00020478644831507627,
      "loss": 2.6326,
      "step": 1548
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5264829559344815,
      "learning_rate": 0.00020418706991563634,
      "loss": 2.6185,
      "step": 1549
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5184649457276234,
      "learning_rate": 0.00020358834480746363,
      "loss": 2.5696,
      "step": 1550
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.4890895726634323,
      "learning_rate": 0.0002029902743128188,
      "loss": 2.5935,
      "step": 1551
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5073768980879371,
      "learning_rate": 0.0002023928597525174,
      "loss": 2.6385,
      "step": 1552
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5276495283221572,
      "learning_rate": 0.00020179610244592595,
      "loss": 2.7057,
      "step": 1553
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5165094433644738,
      "learning_rate": 0.00020120000371095937,
      "loss": 2.6762,
      "step": 1554
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.518793887619729,
      "learning_rate": 0.0002006045648640787,
      "loss": 2.7091,
      "step": 1555
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5246544971119942,
      "learning_rate": 0.00020000978722028713,
      "loss": 2.6711,
      "step": 1556
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5215130139384636,
      "learning_rate": 0.00019941567209312767,
      "loss": 2.5976,
      "step": 1557
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5463917375285926,
      "learning_rate": 0.00019882222079468036,
      "loss": 2.6617,
      "step": 1558
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.543868774564057,
      "learning_rate": 0.0001982294346355595,
      "loss": 2.7018,
      "step": 1559
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5378778964625525,
      "learning_rate": 0.00019763731492490976,
      "loss": 2.7008,
      "step": 1560
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5423401893488466,
      "learning_rate": 0.00019704586297040422,
      "loss": 2.6772,
      "step": 1561
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5332443738305217,
      "learning_rate": 0.0001964550800782417,
      "loss": 2.666,
      "step": 1562
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5444301665790526,
      "learning_rate": 0.00019586496755314288,
      "loss": 2.6345,
      "step": 1563
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.513558461810192,
      "learning_rate": 0.00019527552669834798,
      "loss": 2.6318,
      "step": 1564
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.49982186312761545,
      "learning_rate": 0.0001946867588156142,
      "loss": 2.5932,
      "step": 1565
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5055236621814083,
      "learning_rate": 0.00019409866520521258,
      "loss": 2.5897,
      "step": 1566
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.516065270746231,
      "learning_rate": 0.00019351124716592455,
      "loss": 2.7165,
      "step": 1567
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5277570504491967,
      "learning_rate": 0.0001929245059950397,
      "loss": 2.6549,
      "step": 1568
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5171019803008386,
      "learning_rate": 0.0001923384429883533,
      "loss": 2.6035,
      "step": 1569
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5262507808011542,
      "learning_rate": 0.00019175305944016237,
      "loss": 2.6037,
      "step": 1570
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5153470443332753,
      "learning_rate": 0.00019116835664326326,
      "loss": 2.6733,
      "step": 1571
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.523730142526998,
      "learning_rate": 0.0001905843358889497,
      "loss": 2.5972,
      "step": 1572
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.4923266765926923,
      "learning_rate": 0.00019000099846700836,
      "loss": 2.5601,
      "step": 1573
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5457293917465317,
      "learning_rate": 0.00018941834566571692,
      "loss": 2.6832,
      "step": 1574
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5068475300401649,
      "learning_rate": 0.00018883637877184145,
      "loss": 2.6852,
      "step": 1575
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5483264256408378,
      "learning_rate": 0.00018825509907063325,
      "loss": 2.6164,
      "step": 1576
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5400015884979669,
      "learning_rate": 0.00018767450784582557,
      "loss": 2.679,
      "step": 1577
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5462849398792158,
      "learning_rate": 0.00018709460637963122,
      "loss": 2.6862,
      "step": 1578
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.535656024635467,
      "learning_rate": 0.00018651539595274013,
      "loss": 2.6039,
      "step": 1579
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.545470749810151,
      "learning_rate": 0.00018593687784431578,
      "loss": 2.5927,
      "step": 1580
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5654794153265718,
      "learning_rate": 0.00018535905333199248,
      "loss": 2.6115,
      "step": 1581
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5442121995362399,
      "learning_rate": 0.0001847819236918733,
      "loss": 2.654,
      "step": 1582
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5361455297166935,
      "learning_rate": 0.00018420549019852655,
      "loss": 2.6771,
      "step": 1583
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.49199618672352247,
      "learning_rate": 0.00018362975412498266,
      "loss": 2.5546,
      "step": 1584
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5379377343184183,
      "learning_rate": 0.00018305471674273261,
      "loss": 2.645,
      "step": 1585
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5514526754304316,
      "learning_rate": 0.0001824803793217237,
      "loss": 2.6903,
      "step": 1586
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5092193066921264,
      "learning_rate": 0.00018190674313035737,
      "loss": 2.5942,
      "step": 1587
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4963657698046657,
      "learning_rate": 0.00018133380943548716,
      "loss": 2.6365,
      "step": 1588
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5385601164738788,
      "learning_rate": 0.00018076157950241452,
      "loss": 2.6713,
      "step": 1589
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5116368126754929,
      "learning_rate": 0.00018019005459488652,
      "loss": 2.5941,
      "step": 1590
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5150258847864664,
      "learning_rate": 0.00017961923597509388,
      "loss": 2.5683,
      "step": 1591
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5031381095040823,
      "learning_rate": 0.00017904912490366722,
      "loss": 2.6515,
      "step": 1592
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5371636510106691,
      "learning_rate": 0.00017847972263967433,
      "loss": 2.6688,
      "step": 1593
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5366368195190238,
      "learning_rate": 0.0001779110304406177,
      "loss": 2.4204,
      "step": 1594
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5265619086921874,
      "learning_rate": 0.000177343049562432,
      "loss": 2.7382,
      "step": 1595
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5466934167400589,
      "learning_rate": 0.0001767757812594807,
      "loss": 2.6928,
      "step": 1596
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5968912123524217,
      "learning_rate": 0.0001762092267845534,
      "loss": 2.6484,
      "step": 1597
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5364204205129771,
      "learning_rate": 0.00017564338738886365,
      "loss": 2.6145,
      "step": 1598
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5332747017716918,
      "learning_rate": 0.0001750782643220457,
      "loss": 2.676,
      "step": 1599
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5646977329788221,
      "learning_rate": 0.00017451385883215166,
      "loss": 2.6644,
      "step": 1600
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5213871071379184,
      "learning_rate": 0.00017395017216564863,
      "loss": 2.6017,
      "step": 1601
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5007839433276626,
      "learning_rate": 0.00017338720556741687,
      "loss": 2.6291,
      "step": 1602
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5590048493859461,
      "learning_rate": 0.00017282496028074606,
      "loss": 2.6167,
      "step": 1603
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5066745325111831,
      "learning_rate": 0.00017226343754733254,
      "loss": 2.6141,
      "step": 1604
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5422014666870855,
      "learning_rate": 0.00017170263860727769,
      "loss": 2.7176,
      "step": 1605
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5425212472023727,
      "learning_rate": 0.0001711425646990838,
      "loss": 2.6119,
      "step": 1606
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5272280508735294,
      "learning_rate": 0.00017058321705965202,
      "loss": 2.6717,
      "step": 1607
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5336839049554646,
      "learning_rate": 0.0001700245969242798,
      "loss": 2.6205,
      "step": 1608
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5482261027882128,
      "learning_rate": 0.00016946670552665804,
      "loss": 2.6019,
      "step": 1609
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5472390313948575,
      "learning_rate": 0.00016890954409886795,
      "loss": 2.7677,
      "step": 1610
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5642024499297316,
      "learning_rate": 0.00016835311387137836,
      "loss": 2.5689,
      "step": 1611
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5748762267918953,
      "learning_rate": 0.0001677974160730441,
      "loss": 2.7319,
      "step": 1612
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5274177155566764,
      "learning_rate": 0.00016724245193110176,
      "loss": 2.5966,
      "step": 1613
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5503209534240237,
      "learning_rate": 0.00016668822267116784,
      "loss": 2.6217,
      "step": 1614
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5626151078416814,
      "learning_rate": 0.00016613472951723597,
      "loss": 2.6272,
      "step": 1615
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.55445741179899,
      "learning_rate": 0.00016558197369167434,
      "loss": 2.6675,
      "step": 1616
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5619264941914996,
      "learning_rate": 0.00016502995641522216,
      "loss": 2.6902,
      "step": 1617
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5110522581709123,
      "learning_rate": 0.00016447867890698843,
      "loss": 2.5981,
      "step": 1618
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5405115968834316,
      "learning_rate": 0.00016392814238444753,
      "loss": 2.6154,
      "step": 1619
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5362386395477635,
      "learning_rate": 0.00016337834806343782,
      "loss": 2.6145,
      "step": 1620
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5562726061697222,
      "learning_rate": 0.0001628292971581588,
      "loss": 2.6752,
      "step": 1621
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5138901371850961,
      "learning_rate": 0.00016228099088116772,
      "loss": 2.6545,
      "step": 1622
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5240736483676336,
      "learning_rate": 0.00016173343044337734,
      "loss": 2.6586,
      "step": 1623
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5322491046086423,
      "learning_rate": 0.00016118661705405356,
      "loss": 2.6537,
      "step": 1624
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5244359554006681,
      "learning_rate": 0.00016064055192081255,
      "loss": 2.5796,
      "step": 1625
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5680219325603981,
      "learning_rate": 0.00016009523624961757,
      "loss": 2.6471,
      "step": 1626
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5740940996864791,
      "learning_rate": 0.00015955067124477678,
      "loss": 2.6028,
      "step": 1627
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.6023190021050099,
      "learning_rate": 0.000159006858108941,
      "loss": 2.6125,
      "step": 1628
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5830735261622197,
      "learning_rate": 0.00015846379804310002,
      "loss": 2.6765,
      "step": 1629
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5913668041474787,
      "learning_rate": 0.00015792149224658054,
      "loss": 2.6482,
      "step": 1630
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5261665924200206,
      "learning_rate": 0.00015737994191704385,
      "loss": 2.6442,
      "step": 1631
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5298900785738464,
      "learning_rate": 0.0001568391482504829,
      "loss": 2.6429,
      "step": 1632
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5719526933494885,
      "learning_rate": 0.00015629911244121903,
      "loss": 2.5693,
      "step": 1633
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5222692889586663,
      "learning_rate": 0.0001557598356819,
      "loss": 2.6491,
      "step": 1634
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5867055605998461,
      "learning_rate": 0.00015522131916349786,
      "loss": 2.7019,
      "step": 1635
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5294334833962776,
      "learning_rate": 0.00015468356407530493,
      "loss": 2.7075,
      "step": 1636
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5359405824538117,
      "learning_rate": 0.00015414657160493217,
      "loss": 2.7096,
      "step": 1637
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5490111826525037,
      "learning_rate": 0.00015361034293830673,
      "loss": 2.5614,
      "step": 1638
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5137240563860516,
      "learning_rate": 0.00015307487925966844,
      "loss": 2.5864,
      "step": 1639
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5281239830520816,
      "learning_rate": 0.00015254018175156776,
      "loss": 2.5691,
      "step": 1640
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5407841898908815,
      "learning_rate": 0.0001520062515948632,
      "loss": 2.645,
      "step": 1641
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5712122303596,
      "learning_rate": 0.0001514730899687189,
      "loss": 2.6661,
      "step": 1642
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5419987346279507,
      "learning_rate": 0.00015094069805060122,
      "loss": 2.5803,
      "step": 1643
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5503105457045007,
      "learning_rate": 0.00015040907701627666,
      "loss": 2.7235,
      "step": 1644
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5680547403546333,
      "learning_rate": 0.00014987822803980976,
      "loss": 2.6256,
      "step": 1645
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5701688762099579,
      "learning_rate": 0.00014934815229355965,
      "loss": 2.6727,
      "step": 1646
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.563042871960405,
      "learning_rate": 0.00014881885094817748,
      "loss": 2.683,
      "step": 1647
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5815451375342733,
      "learning_rate": 0.00014829032517260488,
      "loss": 2.6041,
      "step": 1648
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5671998478554637,
      "learning_rate": 0.0001477625761340704,
      "loss": 2.6868,
      "step": 1649
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.545143664786361,
      "learning_rate": 0.0001472356049980868,
      "loss": 2.5912,
      "step": 1650
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5517594197395428,
      "learning_rate": 0.00014670941292844954,
      "loss": 2.6507,
      "step": 1651
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5344167945944867,
      "learning_rate": 0.00014618400108723295,
      "loss": 2.5732,
      "step": 1652
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5684446666618682,
      "learning_rate": 0.00014565937063478862,
      "loss": 2.6814,
      "step": 1653
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5359949038414973,
      "learning_rate": 0.00014513552272974207,
      "loss": 2.6375,
      "step": 1654
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5476790225417992,
      "learning_rate": 0.0001446124585289913,
      "loss": 2.6441,
      "step": 1655
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.538946989560625,
      "learning_rate": 0.00014409017918770266,
      "loss": 2.7112,
      "step": 1656
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5597086298018595,
      "learning_rate": 0.00014356868585930994,
      "loss": 2.6678,
      "step": 1657
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5441589345574198,
      "learning_rate": 0.00014304797969551077,
      "loss": 2.5752,
      "step": 1658
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5686067425585911,
      "learning_rate": 0.00014252806184626417,
      "loss": 2.6465,
      "step": 1659
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5466099142902209,
      "learning_rate": 0.00014200893345978817,
      "loss": 2.6135,
      "step": 1660
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5518992032124671,
      "learning_rate": 0.00014149059568255778,
      "loss": 2.647,
      "step": 1661
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5572770672745526,
      "learning_rate": 0.00014097304965930157,
      "loss": 2.5774,
      "step": 1662
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5276082705791915,
      "learning_rate": 0.00014045629653299953,
      "loss": 2.5275,
      "step": 1663
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5146435268229343,
      "learning_rate": 0.00013994033744488076,
      "loss": 2.6819,
      "step": 1664
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5213038004858518,
      "learning_rate": 0.00013942517353442092,
      "loss": 2.6667,
      "step": 1665
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5226528277689608,
      "learning_rate": 0.0001389108059393391,
      "loss": 2.6502,
      "step": 1666
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5412570142869617,
      "learning_rate": 0.00013839723579559581,
      "loss": 2.6226,
      "step": 1667
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5344368429361485,
      "learning_rate": 0.00013788446423739103,
      "loss": 2.5949,
      "step": 1668
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5451804398916766,
      "learning_rate": 0.00013737249239716042,
      "loss": 2.6875,
      "step": 1669
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5491917614718567,
      "learning_rate": 0.00013686132140557355,
      "loss": 2.6121,
      "step": 1670
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.570343125937439,
      "learning_rate": 0.00013635095239153188,
      "loss": 2.5425,
      "step": 1671
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5175849049558212,
      "learning_rate": 0.00013584138648216527,
      "loss": 2.5849,
      "step": 1672
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5425256919294746,
      "learning_rate": 0.0001353326248028298,
      "loss": 2.6148,
      "step": 1673
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5463945037322986,
      "learning_rate": 0.00013482466847710594,
      "loss": 2.5646,
      "step": 1674
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5876310619474768,
      "learning_rate": 0.00013431751862679554,
      "loss": 2.6538,
      "step": 1675
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5908732469594689,
      "learning_rate": 0.00013381117637191887,
      "loss": 2.6319,
      "step": 1676
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5535024471658077,
      "learning_rate": 0.00013330564283071293,
      "loss": 2.6389,
      "step": 1677
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5643002589392547,
      "learning_rate": 0.000132800919119629,
      "loss": 2.5598,
      "step": 1678
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5585889490701113,
      "learning_rate": 0.00013229700635332948,
      "loss": 2.6606,
      "step": 1679
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5707670251319692,
      "learning_rate": 0.00013179390564468585,
      "loss": 2.6027,
      "step": 1680
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5366146986324742,
      "learning_rate": 0.00013129161810477641,
      "loss": 2.6122,
      "step": 1681
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5239223265856179,
      "learning_rate": 0.0001307901448428837,
      "loss": 2.5756,
      "step": 1682
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5243083793090889,
      "learning_rate": 0.0001302894869664916,
      "loss": 2.5856,
      "step": 1683
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5397573897008999,
      "learning_rate": 0.00012978964558128336,
      "loss": 2.619,
      "step": 1684
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5284857990679738,
      "learning_rate": 0.00012929062179113925,
      "loss": 2.6086,
      "step": 1685
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5664310993763689,
      "learning_rate": 0.00012879241669813368,
      "loss": 2.6429,
      "step": 1686
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5426754049447348,
      "learning_rate": 0.00012829503140253295,
      "loss": 2.7112,
      "step": 1687
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5348538821526615,
      "learning_rate": 0.0001277984670027933,
      "loss": 2.6067,
      "step": 1688
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5569987614416546,
      "learning_rate": 0.00012730272459555737,
      "loss": 2.6706,
      "step": 1689
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5237205982167247,
      "learning_rate": 0.00012680780527565312,
      "loss": 2.5507,
      "step": 1690
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5409291958263605,
      "learning_rate": 0.0001263137101360905,
      "loss": 2.6481,
      "step": 1691
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5234761335877628,
      "learning_rate": 0.00012582044026805922,
      "loss": 2.6164,
      "step": 1692
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5373827268209918,
      "learning_rate": 0.00012532799676092627,
      "loss": 2.5859,
      "step": 1693
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.532880211848324,
      "learning_rate": 0.00012483638070223414,
      "loss": 2.6099,
      "step": 1694
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5557804437799018,
      "learning_rate": 0.00012434559317769752,
      "loss": 2.515,
      "step": 1695
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5577247470692999,
      "learning_rate": 0.0001238556352712012,
      "loss": 2.6751,
      "step": 1696
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5665113806272568,
      "learning_rate": 0.00012336650806479827,
      "loss": 2.6865,
      "step": 1697
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5628909625696256,
      "learning_rate": 0.00012287821263870708,
      "loss": 2.6627,
      "step": 1698
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5382150662864119,
      "learning_rate": 0.00012239075007130885,
      "loss": 2.6005,
      "step": 1699
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.52754821709818,
      "learning_rate": 0.00012190412143914536,
      "loss": 2.7571,
      "step": 1700
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5683479833833636,
      "learning_rate": 0.0001214183278169172,
      "loss": 2.6982,
      "step": 1701
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5125528282902689,
      "learning_rate": 0.00012093337027748042,
      "loss": 2.609,
      "step": 1702
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5451802392484422,
      "learning_rate": 0.00012044924989184459,
      "loss": 2.7034,
      "step": 1703
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5178563483276967,
      "learning_rate": 0.0001199659677291709,
      "loss": 2.5884,
      "step": 1704
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5642323984557994,
      "learning_rate": 0.00011948352485676895,
      "loss": 2.6054,
      "step": 1705
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5654319795204232,
      "learning_rate": 0.00011900192234009477,
      "loss": 2.7164,
      "step": 1706
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5563588698788585,
      "learning_rate": 0.00011852116124274875,
      "loss": 2.6106,
      "step": 1707
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5395462475560304,
      "learning_rate": 0.00011804124262647314,
      "loss": 2.619,
      "step": 1708
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5550487837993278,
      "learning_rate": 0.00011756216755114929,
      "loss": 2.546,
      "step": 1709
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.539296691819172,
      "learning_rate": 0.00011708393707479548,
      "loss": 2.643,
      "step": 1710
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5447409826922939,
      "learning_rate": 0.00011660655225356531,
      "loss": 2.6808,
      "step": 1711
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5482008436519805,
      "learning_rate": 0.0001161300141417444,
      "loss": 2.7427,
      "step": 1712
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5444583054887217,
      "learning_rate": 0.00011565432379174823,
      "loss": 2.5526,
      "step": 1713
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5598490028672822,
      "learning_rate": 0.00011517948225412056,
      "loss": 2.5821,
      "step": 1714
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5488518174144517,
      "learning_rate": 0.00011470549057753032,
      "loss": 2.6692,
      "step": 1715
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5448784312560685,
      "learning_rate": 0.00011423234980876957,
      "loss": 2.5754,
      "step": 1716
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5549707676888599,
      "learning_rate": 0.00011376006099275099,
      "loss": 2.5817,
      "step": 1717
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5465738827508374,
      "learning_rate": 0.00011328862517250609,
      "loss": 2.6125,
      "step": 1718
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5740267266945138,
      "learning_rate": 0.00011281804338918239,
      "loss": 2.6833,
      "step": 1719
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5446652038704665,
      "learning_rate": 0.00011234831668204115,
      "loss": 2.611,
      "step": 1720
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5635434129944377,
      "learning_rate": 0.00011187944608845569,
      "loss": 2.6382,
      "step": 1721
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5572752070930089,
      "learning_rate": 0.00011141143264390801,
      "loss": 2.675,
      "step": 1722
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.535967905002558,
      "learning_rate": 0.0001109442773819877,
      "loss": 2.6024,
      "step": 1723
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5399016275544049,
      "learning_rate": 0.0001104779813343889,
      "loss": 2.7266,
      "step": 1724
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5317105462501159,
      "learning_rate": 0.00011001254553090812,
      "loss": 2.5172,
      "step": 1725
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5350168909062053,
      "learning_rate": 0.00010954797099944186,
      "loss": 2.6889,
      "step": 1726
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5446504250382401,
      "learning_rate": 0.0001090842587659851,
      "loss": 2.6661,
      "step": 1727
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5279143446357414,
      "learning_rate": 0.00010862140985462804,
      "loss": 2.4785,
      "step": 1728
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5864000692972191,
      "learning_rate": 0.00010815942528755418,
      "loss": 2.6563,
      "step": 1729
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5470521318311045,
      "learning_rate": 0.00010769830608503844,
      "loss": 2.5164,
      "step": 1730
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5187775723646955,
      "learning_rate": 0.00010723805326544473,
      "loss": 2.5848,
      "step": 1731
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5552772845000137,
      "learning_rate": 0.00010677866784522316,
      "loss": 2.5786,
      "step": 1732
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5369081434778854,
      "learning_rate": 0.00010632015083890839,
      "loss": 2.6338,
      "step": 1733
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.62070806256124,
      "learning_rate": 0.00010586250325911745,
      "loss": 2.6428,
      "step": 1734
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5622142693494053,
      "learning_rate": 0.00010540572611654697,
      "loss": 2.6593,
      "step": 1735
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5583690218038259,
      "learning_rate": 0.00010494982041997126,
      "loss": 2.6499,
      "step": 1736
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5292073566770972,
      "learning_rate": 0.0001044947871762405,
      "loss": 2.6172,
      "step": 1737
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5191810033878912,
      "learning_rate": 0.00010404062739027753,
      "loss": 2.5655,
      "step": 1738
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5547799066811584,
      "learning_rate": 0.00010358734206507641,
      "loss": 2.603,
      "step": 1739
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5362438810091777,
      "learning_rate": 0.00010313493220170017,
      "loss": 2.5714,
      "step": 1740
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5899901340667328,
      "learning_rate": 0.00010268339879927836,
      "loss": 2.6195,
      "step": 1741
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5469847307481933,
      "learning_rate": 0.00010223274285500466,
      "loss": 2.5791,
      "step": 1742
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5587432277494334,
      "learning_rate": 0.00010178296536413495,
      "loss": 2.6166,
      "step": 1743
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5488966774348845,
      "learning_rate": 0.00010133406731998546,
      "loss": 2.6943,
      "step": 1744
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5883890367017691,
      "learning_rate": 0.00010088604971392979,
      "loss": 2.4534,
      "step": 1745
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5407202392821733,
      "learning_rate": 0.0001004389135353972,
      "loss": 2.6647,
      "step": 1746
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5506938558142956,
      "learning_rate": 9.999265977187049e-05,
      "loss": 2.6668,
      "step": 1747
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5479614136045822,
      "learning_rate": 9.95472894088838e-05,
      "loss": 2.6912,
      "step": 1748
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5487257574413881,
      "learning_rate": 9.910280343001993e-05,
      "loss": 2.5572,
      "step": 1749
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5564984764111545,
      "learning_rate": 9.865920281690866e-05,
      "loss": 2.6374,
      "step": 1750
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5381469587843284,
      "learning_rate": 9.821648854922482e-05,
      "loss": 2.5954,
      "step": 1751
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5170316744215978,
      "learning_rate": 9.77746616046854e-05,
      "loss": 2.5043,
      "step": 1752
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5518426972528372,
      "learning_rate": 9.733372295904774e-05,
      "loss": 2.5819,
      "step": 1753
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5907148581127032,
      "learning_rate": 9.68936735861079e-05,
      "loss": 2.6221,
      "step": 1754
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5306908205408962,
      "learning_rate": 9.645451445769737e-05,
      "loss": 2.5901,
      "step": 1755
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5490707915940605,
      "learning_rate": 9.601624654368196e-05,
      "loss": 2.6413,
      "step": 1756
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5714813514791723,
      "learning_rate": 9.557887081195938e-05,
      "loss": 2.6247,
      "step": 1757
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5408811783025709,
      "learning_rate": 9.514238822845667e-05,
      "loss": 2.6939,
      "step": 1758
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5544013801058092,
      "learning_rate": 9.470679975712837e-05,
      "loss": 2.544,
      "step": 1759
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5321985628370466,
      "learning_rate": 9.427210635995481e-05,
      "loss": 2.602,
      "step": 1760
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5451530908732214,
      "learning_rate": 9.383830899693923e-05,
      "loss": 2.6225,
      "step": 1761
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5565953836543832,
      "learning_rate": 9.340540862610591e-05,
      "loss": 2.6615,
      "step": 1762
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5441808450124133,
      "learning_rate": 9.297340620349854e-05,
      "loss": 2.7066,
      "step": 1763
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5344214290666994,
      "learning_rate": 9.25423026831777e-05,
      "loss": 2.6769,
      "step": 1764
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5491051615004859,
      "learning_rate": 9.211209901721846e-05,
      "loss": 2.6707,
      "step": 1765
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.6106753857847599,
      "learning_rate": 9.168279615570863e-05,
      "loss": 2.539,
      "step": 1766
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5344851828333826,
      "learning_rate": 9.125439504674699e-05,
      "loss": 2.6032,
      "step": 1767
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5324494724345998,
      "learning_rate": 9.082689663644057e-05,
      "loss": 2.6112,
      "step": 1768
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5575088929336367,
      "learning_rate": 9.040030186890264e-05,
      "loss": 2.7511,
      "step": 1769
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5239100832178339,
      "learning_rate": 8.997461168625138e-05,
      "loss": 2.6463,
      "step": 1770
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5801588569088706,
      "learning_rate": 8.954982702860664e-05,
      "loss": 2.6277,
      "step": 1771
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5133599350993528,
      "learning_rate": 8.912594883408865e-05,
      "loss": 2.7109,
      "step": 1772
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5682979468664974,
      "learning_rate": 8.870297803881589e-05,
      "loss": 2.5859,
      "step": 1773
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5129385586006339,
      "learning_rate": 8.828091557690287e-05,
      "loss": 2.5573,
      "step": 1774
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5290786428372344,
      "learning_rate": 8.785976238045801e-05,
      "loss": 2.6164,
      "step": 1775
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.545305501637547,
      "learning_rate": 8.743951937958144e-05,
      "loss": 2.6673,
      "step": 1776
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5657109058351861,
      "learning_rate": 8.702018750236357e-05,
      "loss": 2.6727,
      "step": 1777
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5582963981092707,
      "learning_rate": 8.660176767488237e-05,
      "loss": 2.7218,
      "step": 1778
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5735216135727991,
      "learning_rate": 8.618426082120146e-05,
      "loss": 2.5729,
      "step": 1779
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5631030619693372,
      "learning_rate": 8.576766786336854e-05,
      "loss": 2.5433,
      "step": 1780
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5550510943306851,
      "learning_rate": 8.535198972141294e-05,
      "loss": 2.5463,
      "step": 1781
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5435231618465673,
      "learning_rate": 8.493722731334347e-05,
      "loss": 2.6633,
      "step": 1782
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5585007276953949,
      "learning_rate": 8.452338155514644e-05,
      "loss": 2.6395,
      "step": 1783
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5485125839745638,
      "learning_rate": 8.411045336078426e-05,
      "loss": 2.693,
      "step": 1784
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5471434269174249,
      "learning_rate": 8.369844364219264e-05,
      "loss": 2.5727,
      "step": 1785
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5410174539665094,
      "learning_rate": 8.328735330927873e-05,
      "loss": 2.6142,
      "step": 1786
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5765358160346983,
      "learning_rate": 8.287718326991961e-05,
      "loss": 2.6867,
      "step": 1787
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5625347776401176,
      "learning_rate": 8.246793442995954e-05,
      "loss": 2.6872,
      "step": 1788
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5423815791449932,
      "learning_rate": 8.205960769320875e-05,
      "loss": 2.5215,
      "step": 1789
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5418964417672332,
      "learning_rate": 8.165220396144085e-05,
      "loss": 2.5728,
      "step": 1790
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5308861677059935,
      "learning_rate": 8.12457241343909e-05,
      "loss": 2.6112,
      "step": 1791
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.559872390520589,
      "learning_rate": 8.084016910975367e-05,
      "loss": 2.6295,
      "step": 1792
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5330425745491147,
      "learning_rate": 8.043553978318169e-05,
      "loss": 2.6133,
      "step": 1793
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5373662749025255,
      "learning_rate": 8.003183704828281e-05,
      "loss": 2.5563,
      "step": 1794
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5697522265009489,
      "learning_rate": 7.962906179661872e-05,
      "loss": 2.5703,
      "step": 1795
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5403239474908315,
      "learning_rate": 7.922721491770296e-05,
      "loss": 2.6816,
      "step": 1796
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5357331282649023,
      "learning_rate": 7.882629729899832e-05,
      "loss": 2.5324,
      "step": 1797
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5330425392505703,
      "learning_rate": 7.842630982591598e-05,
      "loss": 2.6271,
      "step": 1798
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5497993099013603,
      "learning_rate": 7.802725338181232e-05,
      "loss": 2.6284,
      "step": 1799
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5785227746536995,
      "learning_rate": 7.762912884798812e-05,
      "loss": 2.5593,
      "step": 1800
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5796617564855405,
      "learning_rate": 7.723193710368564e-05,
      "loss": 2.6891,
      "step": 1801
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5717313633312403,
      "learning_rate": 7.683567902608729e-05,
      "loss": 2.6167,
      "step": 1802
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5544617598187658,
      "learning_rate": 7.644035549031364e-05,
      "loss": 2.6502,
      "step": 1803
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5707450426280389,
      "learning_rate": 7.604596736942115e-05,
      "loss": 2.6446,
      "step": 1804
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5689919735836062,
      "learning_rate": 7.56525155344004e-05,
      "loss": 2.6216,
      "step": 1805
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5481009818940108,
      "learning_rate": 7.52600008541745e-05,
      "loss": 2.6349,
      "step": 1806
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5724154404305976,
      "learning_rate": 7.486842419559681e-05,
      "loss": 2.6432,
      "step": 1807
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5372138222283107,
      "learning_rate": 7.447778642344898e-05,
      "loss": 2.5991,
      "step": 1808
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5609282609839842,
      "learning_rate": 7.408808840043912e-05,
      "loss": 2.5021,
      "step": 1809
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5415748883648057,
      "learning_rate": 7.369933098720021e-05,
      "loss": 2.5238,
      "step": 1810
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.548574122861022,
      "learning_rate": 7.331151504228767e-05,
      "loss": 2.5734,
      "step": 1811
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5273357262431444,
      "learning_rate": 7.292464142217775e-05,
      "loss": 2.5457,
      "step": 1812
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5477377816470547,
      "learning_rate": 7.25387109812658e-05,
      "loss": 2.577,
      "step": 1813
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5149933044746808,
      "learning_rate": 7.215372457186415e-05,
      "loss": 2.6433,
      "step": 1814
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5349448460951758,
      "learning_rate": 7.176968304420007e-05,
      "loss": 2.6173,
      "step": 1815
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5476430810329321,
      "learning_rate": 7.138658724641417e-05,
      "loss": 2.6073,
      "step": 1816
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5496804331090777,
      "learning_rate": 7.10044380245587e-05,
      "loss": 2.5704,
      "step": 1817
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5673331480646342,
      "learning_rate": 7.062323622259515e-05,
      "loss": 2.6118,
      "step": 1818
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5358399014340838,
      "learning_rate": 7.024298268239265e-05,
      "loss": 2.6441,
      "step": 1819
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5600576303312883,
      "learning_rate": 6.986367824372647e-05,
      "loss": 2.6055,
      "step": 1820
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5494191886853074,
      "learning_rate": 6.948532374427541e-05,
      "loss": 2.4867,
      "step": 1821
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5658075152185014,
      "learning_rate": 6.910792001962063e-05,
      "loss": 2.6615,
      "step": 1822
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5362673718007406,
      "learning_rate": 6.873146790324358e-05,
      "loss": 2.569,
      "step": 1823
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5592104425968037,
      "learning_rate": 6.83559682265239e-05,
      "loss": 2.6465,
      "step": 1824
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5623609993071005,
      "learning_rate": 6.798142181873784e-05,
      "loss": 2.6165,
      "step": 1825
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5459918493923973,
      "learning_rate": 6.760782950705662e-05,
      "loss": 2.5953,
      "step": 1826
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5943883152641627,
      "learning_rate": 6.723519211654422e-05,
      "loss": 2.5188,
      "step": 1827
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5638920635113399,
      "learning_rate": 6.686351047015554e-05,
      "loss": 2.6788,
      "step": 1828
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5615639471648708,
      "learning_rate": 6.649278538873515e-05,
      "loss": 2.6668,
      "step": 1829
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5088103445168204,
      "learning_rate": 6.612301769101465e-05,
      "loss": 2.6292,
      "step": 1830
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5406730930468409,
      "learning_rate": 6.575420819361177e-05,
      "loss": 2.624,
      "step": 1831
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5388718228721487,
      "learning_rate": 6.538635771102757e-05,
      "loss": 2.6235,
      "step": 1832
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5328894666272541,
      "learning_rate": 6.501946705564566e-05,
      "loss": 2.5642,
      "step": 1833
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5567356052375771,
      "learning_rate": 6.465353703772959e-05,
      "loss": 2.6026,
      "step": 1834
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5582170076448739,
      "learning_rate": 6.428856846542136e-05,
      "loss": 2.6433,
      "step": 1835
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5741568161669568,
      "learning_rate": 6.392456214473996e-05,
      "loss": 2.6161,
      "step": 1836
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5213392174036122,
      "learning_rate": 6.3561518879579e-05,
      "loss": 2.6734,
      "step": 1837
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5406757269685757,
      "learning_rate": 6.31994394717052e-05,
      "loss": 2.5142,
      "step": 1838
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5722243661896202,
      "learning_rate": 6.283832472075685e-05,
      "loss": 2.5703,
      "step": 1839
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5355174723845649,
      "learning_rate": 6.247817542424178e-05,
      "loss": 2.6415,
      "step": 1840
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5626031451509184,
      "learning_rate": 6.211899237753559e-05,
      "loss": 2.5507,
      "step": 1841
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5553912574581619,
      "learning_rate": 6.176077637387984e-05,
      "loss": 2.5559,
      "step": 1842
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5343724008924052,
      "learning_rate": 6.140352820438066e-05,
      "loss": 2.5922,
      "step": 1843
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5717148800912805,
      "learning_rate": 6.104724865800665e-05,
      "loss": 2.6343,
      "step": 1844
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5627965263510422,
      "learning_rate": 6.069193852158711e-05,
      "loss": 2.6531,
      "step": 1845
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.532720445355231,
      "learning_rate": 6.0337598579810584e-05,
      "loss": 2.6855,
      "step": 1846
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5594427258109053,
      "learning_rate": 5.9984229615223096e-05,
      "loss": 2.562,
      "step": 1847
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.557453981521758,
      "learning_rate": 5.963183240822606e-05,
      "loss": 2.4471,
      "step": 1848
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5784412960381519,
      "learning_rate": 5.9280407737074825e-05,
      "loss": 2.662,
      "step": 1849
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.529999920497906,
      "learning_rate": 5.8929956377877125e-05,
      "loss": 2.4706,
      "step": 1850
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5279777567586011,
      "learning_rate": 5.8580479104591075e-05,
      "loss": 2.6327,
      "step": 1851
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5303784890895924,
      "learning_rate": 5.823197668902341e-05,
      "loss": 2.5787,
      "step": 1852
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5615200627267017,
      "learning_rate": 5.78844499008282e-05,
      "loss": 2.5963,
      "step": 1853
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5556635131994543,
      "learning_rate": 5.753789950750454e-05,
      "loss": 2.6531,
      "step": 1854
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5365574692874224,
      "learning_rate": 5.719232627439558e-05,
      "loss": 2.5386,
      "step": 1855
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5528271396864795,
      "learning_rate": 5.6847730964686315e-05,
      "loss": 2.5724,
      "step": 1856
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5563348612936558,
      "learning_rate": 5.650411433940189e-05,
      "loss": 2.5846,
      "step": 1857
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5488295613427898,
      "learning_rate": 5.61614771574061e-05,
      "loss": 2.5516,
      "step": 1858
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5711293329632678,
      "learning_rate": 5.581982017539988e-05,
      "loss": 2.6656,
      "step": 1859
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5465930217077909,
      "learning_rate": 5.5479144147919216e-05,
      "loss": 2.6724,
      "step": 1860
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5376997482569861,
      "learning_rate": 5.51394498273336e-05,
      "loss": 2.6298,
      "step": 1861
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5385324458653116,
      "learning_rate": 5.480073796384494e-05,
      "loss": 2.5873,
      "step": 1862
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.6128694428166196,
      "learning_rate": 5.446300930548492e-05,
      "loss": 2.6758,
      "step": 1863
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.565728493901671,
      "learning_rate": 5.412626459811415e-05,
      "loss": 2.6221,
      "step": 1864
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.561917637579157,
      "learning_rate": 5.3790504585419954e-05,
      "loss": 2.6401,
      "step": 1865
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5540313083764529,
      "learning_rate": 5.345573000891541e-05,
      "loss": 2.5593,
      "step": 1866
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5744651038959714,
      "learning_rate": 5.312194160793693e-05,
      "loss": 2.6149,
      "step": 1867
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.555728674263838,
      "learning_rate": 5.278914011964303e-05,
      "loss": 2.692,
      "step": 1868
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5747536830564772,
      "learning_rate": 5.2457326279013006e-05,
      "loss": 2.7261,
      "step": 1869
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5680417469566157,
      "learning_rate": 5.2126500818844514e-05,
      "loss": 2.5819,
      "step": 1870
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5653359070758456,
      "learning_rate": 5.1796664469752566e-05,
      "loss": 2.6669,
      "step": 1871
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5576077609092146,
      "learning_rate": 5.1467817960167975e-05,
      "loss": 2.4854,
      "step": 1872
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5575213830188462,
      "learning_rate": 5.113996201633536e-05,
      "loss": 2.6286,
      "step": 1873
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5450600362465774,
      "learning_rate": 5.0813097362311765e-05,
      "loss": 2.6339,
      "step": 1874
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5293144625593531,
      "learning_rate": 5.048722471996475e-05,
      "loss": 2.5199,
      "step": 1875
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5506130282725965,
      "learning_rate": 5.016234480897158e-05,
      "loss": 2.6013,
      "step": 1876
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5515824496578585,
      "learning_rate": 4.9838458346816664e-05,
      "loss": 2.5965,
      "step": 1877
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5638673406490248,
      "learning_rate": 4.9515566048790485e-05,
      "loss": 2.7194,
      "step": 1878
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5597673939948471,
      "learning_rate": 4.9193668627988074e-05,
      "loss": 2.6027,
      "step": 1879
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5555095251873071,
      "learning_rate": 4.887276679530744e-05,
      "loss": 2.5756,
      "step": 1880
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5797999831068436,
      "learning_rate": 4.855286125944752e-05,
      "loss": 2.6448,
      "step": 1881
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.566653169360532,
      "learning_rate": 4.8233952726907224e-05,
      "loss": 2.5706,
      "step": 1882
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.546248178287042,
      "learning_rate": 4.7916041901983565e-05,
      "loss": 2.6407,
      "step": 1883
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.556134752042644,
      "learning_rate": 4.7599129486770145e-05,
      "loss": 2.6353,
      "step": 1884
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5395160196183679,
      "learning_rate": 4.728321618115555e-05,
      "loss": 2.5826,
      "step": 1885
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5441132004996001,
      "learning_rate": 4.696830268282204e-05,
      "loss": 2.6388,
      "step": 1886
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5457343814525684,
      "learning_rate": 4.665438968724361e-05,
      "loss": 2.5419,
      "step": 1887
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5559220140590277,
      "learning_rate": 4.634147788768489e-05,
      "loss": 2.6599,
      "step": 1888
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.56038111484417,
      "learning_rate": 4.6029567975199414e-05,
      "loss": 2.6585,
      "step": 1889
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5707031086434,
      "learning_rate": 4.571866063862795e-05,
      "loss": 2.5907,
      "step": 1890
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5778647507623041,
      "learning_rate": 4.540875656459703e-05,
      "loss": 2.615,
      "step": 1891
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5580906433187225,
      "learning_rate": 4.509985643751785e-05,
      "loss": 2.554,
      "step": 1892
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5340167900556103,
      "learning_rate": 4.479196093958421e-05,
      "loss": 2.5467,
      "step": 1893
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5415175921688076,
      "learning_rate": 4.4485070750771187e-05,
      "loss": 2.6266,
      "step": 1894
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5320737601625204,
      "learning_rate": 4.417918654883363e-05,
      "loss": 2.6269,
      "step": 1895
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5484890170694927,
      "learning_rate": 4.3874309009305e-05,
      "loss": 2.6587,
      "step": 1896
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5404961269753865,
      "learning_rate": 4.357043880549538e-05,
      "loss": 2.5962,
      "step": 1897
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5659249952145858,
      "learning_rate": 4.326757660849012e-05,
      "loss": 2.5435,
      "step": 1898
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5832755132000856,
      "learning_rate": 4.2965723087148635e-05,
      "loss": 2.6425,
      "step": 1899
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5612634462750079,
      "learning_rate": 4.266487890810256e-05,
      "loss": 2.5898,
      "step": 1900
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5163529465472395,
      "learning_rate": 4.2365044735754365e-05,
      "loss": 2.6262,
      "step": 1901
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5549367226778276,
      "learning_rate": 4.2066221232276266e-05,
      "loss": 2.6214,
      "step": 1902
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5701982005948943,
      "learning_rate": 4.176840905760815e-05,
      "loss": 2.5619,
      "step": 1903
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5605467505726942,
      "learning_rate": 4.1471608869456443e-05,
      "loss": 2.5536,
      "step": 1904
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5554422965625773,
      "learning_rate": 4.117582132329284e-05,
      "loss": 2.584,
      "step": 1905
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.548234285706048,
      "learning_rate": 4.088104707235263e-05,
      "loss": 2.5208,
      "step": 1906
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5736323684411657,
      "learning_rate": 4.058728676763313e-05,
      "loss": 2.6048,
      "step": 1907
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5336417815764188,
      "learning_rate": 4.0294541057892375e-05,
      "loss": 2.5608,
      "step": 1908
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5658236931945755,
      "learning_rate": 4.000281058964794e-05,
      "loss": 2.6635,
      "step": 1909
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5608256544030341,
      "learning_rate": 3.971209600717507e-05,
      "loss": 2.5535,
      "step": 1910
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5748952501024838,
      "learning_rate": 3.9422397952505465e-05,
      "loss": 2.7331,
      "step": 1911
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5749733120053022,
      "learning_rate": 3.913371706542596e-05,
      "loss": 2.5722,
      "step": 1912
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5278295237135808,
      "learning_rate": 3.884605398347707e-05,
      "loss": 2.6533,
      "step": 1913
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5543588962568842,
      "learning_rate": 3.8559409341951456e-05,
      "loss": 2.6164,
      "step": 1914
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5949226489812678,
      "learning_rate": 3.8273783773892404e-05,
      "loss": 2.7377,
      "step": 1915
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5323498199290453,
      "learning_rate": 3.798917791009293e-05,
      "loss": 2.6285,
      "step": 1916
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.563348606963326,
      "learning_rate": 3.770559237909393e-05,
      "loss": 2.6438,
      "step": 1917
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5187195495450032,
      "learning_rate": 3.742302780718288e-05,
      "loss": 2.579,
      "step": 1918
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5488845723888442,
      "learning_rate": 3.7141484818392635e-05,
      "loss": 2.6211,
      "step": 1919
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5841904034758841,
      "learning_rate": 3.686096403449973e-05,
      "loss": 2.7047,
      "step": 1920
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5474220941034659,
      "learning_rate": 3.658146607502344e-05,
      "loss": 2.5843,
      "step": 1921
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5892908011748751,
      "learning_rate": 3.630299155722411e-05,
      "loss": 2.7218,
      "step": 1922
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5284963043794547,
      "learning_rate": 3.6025541096101676e-05,
      "loss": 2.6381,
      "step": 1923
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5540251586144554,
      "learning_rate": 3.574911530439473e-05,
      "loss": 2.7206,
      "step": 1924
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.547543586283262,
      "learning_rate": 3.5473714792578606e-05,
      "loss": 2.5744,
      "step": 1925
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5258221256374013,
      "learning_rate": 3.519934016886478e-05,
      "loss": 2.5755,
      "step": 1926
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5923978559455237,
      "learning_rate": 3.4925992039198776e-05,
      "loss": 2.5743,
      "step": 1927
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5445422820473839,
      "learning_rate": 3.465367100725908e-05,
      "loss": 2.6345,
      "step": 1928
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5714715217198404,
      "learning_rate": 3.438237767445618e-05,
      "loss": 2.5756,
      "step": 1929
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5636338269073103,
      "learning_rate": 3.4112112639930804e-05,
      "loss": 2.652,
      "step": 1930
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5569823094512336,
      "learning_rate": 3.3842876500552564e-05,
      "loss": 2.6339,
      "step": 1931
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.549198770775433,
      "learning_rate": 3.357466985091906e-05,
      "loss": 2.675,
      "step": 1932
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5558372522382634,
      "learning_rate": 3.330749328335414e-05,
      "loss": 2.6178,
      "step": 1933
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5618887762122602,
      "learning_rate": 3.304134738790659e-05,
      "loss": 2.5464,
      "step": 1934
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5815484656885673,
      "learning_rate": 3.277623275234953e-05,
      "loss": 2.6934,
      "step": 1935
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5767150180503202,
      "learning_rate": 3.2512149962177994e-05,
      "loss": 2.5918,
      "step": 1936
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5708263055908799,
      "learning_rate": 3.224909960060851e-05,
      "loss": 2.658,
      "step": 1937
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5713859977920954,
      "learning_rate": 3.198708224857755e-05,
      "loss": 2.5925,
      "step": 1938
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5786494780437252,
      "learning_rate": 3.172609848474023e-05,
      "loss": 2.6289,
      "step": 1939
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5412803674694997,
      "learning_rate": 3.1466148885468895e-05,
      "loss": 2.601,
      "step": 1940
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5677518875545632,
      "learning_rate": 3.120723402485198e-05,
      "loss": 2.5157,
      "step": 1941
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5369470045006737,
      "learning_rate": 3.094935447469294e-05,
      "loss": 2.48,
      "step": 1942
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5414999957089642,
      "learning_rate": 3.069251080450863e-05,
      "loss": 2.5812,
      "step": 1943
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5703195241957268,
      "learning_rate": 3.0436703581528113e-05,
      "loss": 2.56,
      "step": 1944
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5308083448069597,
      "learning_rate": 3.0181933370691694e-05,
      "loss": 2.5313,
      "step": 1945
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.6305228273904662,
      "learning_rate": 2.9928200734649523e-05,
      "loss": 2.6101,
      "step": 1946
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5703336386914977,
      "learning_rate": 2.9675506233760142e-05,
      "loss": 2.6373,
      "step": 1947
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5765640918305002,
      "learning_rate": 2.942385042608925e-05,
      "loss": 2.6079,
      "step": 1948
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5598241949294356,
      "learning_rate": 2.9173233867409054e-05,
      "loss": 2.545,
      "step": 1949
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5678715812727344,
      "learning_rate": 2.892365711119638e-05,
      "loss": 2.6138,
      "step": 1950
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5786734644760139,
      "learning_rate": 2.8675120708631596e-05,
      "loss": 2.7019,
      "step": 1951
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5540140426315893,
      "learning_rate": 2.8427625208597764e-05,
      "loss": 2.6644,
      "step": 1952
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.553583199893664,
      "learning_rate": 2.8181171157678874e-05,
      "loss": 2.5402,
      "step": 1953
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5575865710103217,
      "learning_rate": 2.7935759100159053e-05,
      "loss": 2.664,
      "step": 1954
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5554717486693093,
      "learning_rate": 2.7691389578021365e-05,
      "loss": 2.657,
      "step": 1955
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5583210849179047,
      "learning_rate": 2.7448063130946223e-05,
      "loss": 2.4653,
      "step": 1956
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5412067199731216,
      "learning_rate": 2.7205780296310544e-05,
      "loss": 2.5567,
      "step": 1957
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.55625528253911,
      "learning_rate": 2.6964541609186378e-05,
      "loss": 2.5468,
      "step": 1958
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5589387151421397,
      "learning_rate": 2.6724347602340104e-05,
      "loss": 2.6495,
      "step": 1959
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5700018897605778,
      "learning_rate": 2.6485198806230682e-05,
      "loss": 2.5676,
      "step": 1960
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5550350043330647,
      "learning_rate": 2.6247095749008797e-05,
      "loss": 2.702,
      "step": 1961
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5553559995117753,
      "learning_rate": 2.6010038956515826e-05,
      "loss": 2.5779,
      "step": 1962
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5586032531347148,
      "learning_rate": 2.5774028952282423e-05,
      "loss": 2.6463,
      "step": 1963
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5819573899832859,
      "learning_rate": 2.5539066257527277e-05,
      "loss": 2.6219,
      "step": 1964
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.570469545623377,
      "learning_rate": 2.530515139115652e-05,
      "loss": 2.5564,
      "step": 1965
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5323936406036233,
      "learning_rate": 2.5072284869761874e-05,
      "loss": 2.6222,
      "step": 1966
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5649571504661642,
      "learning_rate": 2.4840467207619786e-05,
      "loss": 2.545,
      "step": 1967
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5383741778359044,
      "learning_rate": 2.460969891669068e-05,
      "loss": 2.7287,
      "step": 1968
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5776750777006242,
      "learning_rate": 2.4379980506617272e-05,
      "loss": 2.6357,
      "step": 1969
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5383209832443907,
      "learning_rate": 2.4151312484723464e-05,
      "loss": 2.6186,
      "step": 1970
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5515820712853415,
      "learning_rate": 2.3923695356013798e-05,
      "loss": 2.5774,
      "step": 1971
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5515145921446176,
      "learning_rate": 2.3697129623171833e-05,
      "loss": 2.654,
      "step": 1972
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5546555397613108,
      "learning_rate": 2.3471615786559042e-05,
      "loss": 2.6185,
      "step": 1973
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5432639152693998,
      "learning_rate": 2.3247154344213818e-05,
      "loss": 2.5488,
      "step": 1974
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5676140348441109,
      "learning_rate": 2.3023745791850625e-05,
      "loss": 2.5906,
      "step": 1975
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5359552127588788,
      "learning_rate": 2.2801390622858354e-05,
      "loss": 2.53,
      "step": 1976
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5563462736046625,
      "learning_rate": 2.2580089328299746e-05,
      "loss": 2.6612,
      "step": 1977
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5738741312851797,
      "learning_rate": 2.235984239690997e-05,
      "loss": 2.5911,
      "step": 1978
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.6041119881334491,
      "learning_rate": 2.2140650315095934e-05,
      "loss": 2.5549,
      "step": 1979
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5643772450688785,
      "learning_rate": 2.192251356693459e-05,
      "loss": 2.6162,
      "step": 1980
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5753423188167993,
      "learning_rate": 2.170543263417246e-05,
      "loss": 2.6977,
      "step": 1981
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5659412879568343,
      "learning_rate": 2.1489407996224286e-05,
      "loss": 2.5832,
      "step": 1982
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.595598053777555,
      "learning_rate": 2.127444013017199e-05,
      "loss": 2.6555,
      "step": 1983
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5798754753275857,
      "learning_rate": 2.1060529510763648e-05,
      "loss": 2.6103,
      "step": 1984
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.584147449452292,
      "learning_rate": 2.084767661041259e-05,
      "loss": 2.5332,
      "step": 1985
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.60196740428947,
      "learning_rate": 2.063588189919596e-05,
      "loss": 2.6075,
      "step": 1986
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5482427737178445,
      "learning_rate": 2.0425145844854275e-05,
      "loss": 2.6379,
      "step": 1987
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5607081909696284,
      "learning_rate": 2.0215468912789693e-05,
      "loss": 2.624,
      "step": 1988
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5593411189123502,
      "learning_rate": 2.0006851566065575e-05,
      "loss": 2.541,
      "step": 1989
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5654861259582422,
      "learning_rate": 1.9799294265405166e-05,
      "loss": 2.5475,
      "step": 1990
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5788071402697547,
      "learning_rate": 1.9592797469190572e-05,
      "loss": 2.6639,
      "step": 1991
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5408683455456988,
      "learning_rate": 1.938736163346194e-05,
      "loss": 2.5134,
      "step": 1992
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5430106968101099,
      "learning_rate": 1.9182987211916246e-05,
      "loss": 2.6233,
      "step": 1993
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.5422432728401679,
      "learning_rate": 1.8979674655906332e-05,
      "loss": 2.6307,
      "step": 1994
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.583073805234189,
      "learning_rate": 1.8777424414440024e-05,
      "loss": 2.6182,
      "step": 1995
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5791752278489167,
      "learning_rate": 1.8576236934179202e-05,
      "loss": 2.5133,
      "step": 1996
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5664425614757755,
      "learning_rate": 1.8376112659438393e-05,
      "loss": 2.5911,
      "step": 1997
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.6037602975112603,
      "learning_rate": 1.8177052032184282e-05,
      "loss": 2.6039,
      "step": 1998
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5282021780354484,
      "learning_rate": 1.7979055492034435e-05,
      "loss": 2.5463,
      "step": 1999
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5594585572464797,
      "learning_rate": 1.7782123476256407e-05,
      "loss": 2.6044,
      "step": 2000
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5468791615059586,
      "learning_rate": 1.7586256419766965e-05,
      "loss": 2.4832,
      "step": 2001
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5699449854950449,
      "learning_rate": 1.7391454755130766e-05,
      "loss": 2.6199,
      "step": 2002
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.6055361068323695,
      "learning_rate": 1.7197718912559557e-05,
      "loss": 2.654,
      "step": 2003
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.550062263239398,
      "learning_rate": 1.700504931991148e-05,
      "loss": 2.5736,
      "step": 2004
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5585744144969212,
      "learning_rate": 1.681344640268978e-05,
      "loss": 2.5365,
      "step": 2005
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5835383094196378,
      "learning_rate": 1.6622910584041974e-05,
      "loss": 2.6064,
      "step": 2006
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5495878980535746,
      "learning_rate": 1.6433442284758903e-05,
      "loss": 2.7435,
      "step": 2007
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5728677402158682,
      "learning_rate": 1.624504192327392e-05,
      "loss": 2.6706,
      "step": 2008
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5827190995800148,
      "learning_rate": 1.6057709915661856e-05,
      "loss": 2.6244,
      "step": 2009
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5576657271073403,
      "learning_rate": 1.5871446675638057e-05,
      "loss": 2.6192,
      "step": 2010
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5668046744903577,
      "learning_rate": 1.5686252614557638e-05,
      "loss": 2.6168,
      "step": 2011
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5578617827113114,
      "learning_rate": 1.5502128141414497e-05,
      "loss": 2.629,
      "step": 2012
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5629790073525897,
      "learning_rate": 1.5319073662840188e-05,
      "loss": 2.611,
      "step": 2013
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5742780040458331,
      "learning_rate": 1.5137089583103391e-05,
      "loss": 2.6562,
      "step": 2014
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.5840306678585309,
      "learning_rate": 1.4956176304108893e-05,
      "loss": 2.5728,
      "step": 2015
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.565662016557998,
      "learning_rate": 1.4776334225396481e-05,
      "loss": 2.5833,
      "step": 2016
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5473874428764963,
      "learning_rate": 1.4597563744140397e-05,
      "loss": 2.6196,
      "step": 2017
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5742914832325806,
      "learning_rate": 1.4419865255148269e-05,
      "loss": 2.6388,
      "step": 2018
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5832428790768223,
      "learning_rate": 1.4243239150860122e-05,
      "loss": 2.6109,
      "step": 2019
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5474813761774111,
      "learning_rate": 1.4067685821347932e-05,
      "loss": 2.6495,
      "step": 2020
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5743090296722626,
      "learning_rate": 1.389320565431429e-05,
      "loss": 2.4992,
      "step": 2021
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5611969145396326,
      "learning_rate": 1.3719799035091851e-05,
      "loss": 2.6273,
      "step": 2022
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5642264550228595,
      "learning_rate": 1.3547466346642278e-05,
      "loss": 2.6551,
      "step": 2023
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5578223016295393,
      "learning_rate": 1.3376207969555577e-05,
      "loss": 2.5434,
      "step": 2024
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5591708618447149,
      "learning_rate": 1.32060242820492e-05,
      "loss": 2.5776,
      "step": 2025
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5281557235890622,
      "learning_rate": 1.3036915659967118e-05,
      "loss": 2.6204,
      "step": 2026
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5409089565451778,
      "learning_rate": 1.2868882476779087e-05,
      "loss": 2.6111,
      "step": 2027
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.556044649290045,
      "learning_rate": 1.2701925103579815e-05,
      "loss": 2.5592,
      "step": 2028
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5713805225054195,
      "learning_rate": 1.2536043909088191e-05,
      "loss": 2.6052,
      "step": 2029
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5790642069738291,
      "learning_rate": 1.2371239259646228e-05,
      "loss": 2.5494,
      "step": 2030
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.585255255102651,
      "learning_rate": 1.2207511519218672e-05,
      "loss": 2.6427,
      "step": 2031
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5649573174523728,
      "learning_rate": 1.2044861049391676e-05,
      "loss": 2.6349,
      "step": 2032
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5786562186417628,
      "learning_rate": 1.1883288209372512e-05,
      "loss": 2.5655,
      "step": 2033
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5579688162203225,
      "learning_rate": 1.1722793355988471e-05,
      "loss": 2.5627,
      "step": 2034
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5370557926554806,
      "learning_rate": 1.1563376843686135e-05,
      "loss": 2.6402,
      "step": 2035
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5529135492195391,
      "learning_rate": 1.140503902453055e-05,
      "loss": 2.6197,
      "step": 2036
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5971178712109217,
      "learning_rate": 1.1247780248204665e-05,
      "loss": 2.5404,
      "step": 2037
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.5662732256894579,
      "learning_rate": 1.1091600862008333e-05,
      "loss": 2.6307,
      "step": 2038
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5753445115564004,
      "learning_rate": 1.0936501210857652e-05,
      "loss": 2.5623,
      "step": 2039
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5460918727883507,
      "learning_rate": 1.0782481637284013e-05,
      "loss": 2.5622,
      "step": 2040
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5366200522676484,
      "learning_rate": 1.0629542481433663e-05,
      "loss": 2.5414,
      "step": 2041
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5443654657351544,
      "learning_rate": 1.0477684081066751e-05,
      "loss": 2.5711,
      "step": 2042
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5731169615476871,
      "learning_rate": 1.0326906771556566e-05,
      "loss": 2.587,
      "step": 2043
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5727770499153597,
      "learning_rate": 1.017721088588891e-05,
      "loss": 2.6988,
      "step": 2044
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5696538920467711,
      "learning_rate": 1.0028596754661334e-05,
      "loss": 2.4932,
      "step": 2045
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5371149279838393,
      "learning_rate": 9.881064706082298e-06,
      "loss": 2.5983,
      "step": 2046
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.563129387514548,
      "learning_rate": 9.734615065970454e-06,
      "loss": 2.5931,
      "step": 2047
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5668196886222572,
      "learning_rate": 9.58924815775425e-06,
      "loss": 2.5436,
      "step": 2048
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5696311629164159,
      "learning_rate": 9.444964302470715e-06,
      "loss": 2.6363,
      "step": 2049
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5623287843579446,
      "learning_rate": 9.301763818765018e-06,
      "loss": 2.6385,
      "step": 2050
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5929867465817565,
      "learning_rate": 9.15964702288996e-06,
      "loss": 2.7061,
      "step": 2051
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5708810810824109,
      "learning_rate": 9.018614228704925e-06,
      "loss": 2.677,
      "step": 2052
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.535918052656494,
      "learning_rate": 8.878665747675152e-06,
      "loss": 2.516,
      "step": 2053
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5546241578958482,
      "learning_rate": 8.739801888871469e-06,
      "loss": 2.6661,
      "step": 2054
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5567785062752941,
      "learning_rate": 8.602022958969336e-06,
      "loss": 2.6273,
      "step": 2055
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5448922601135263,
      "learning_rate": 8.465329262248078e-06,
      "loss": 2.6086,
      "step": 2056
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.552292675792888,
      "learning_rate": 8.32972110059027e-06,
      "loss": 2.6343,
      "step": 2057
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.561376888209488,
      "learning_rate": 8.195198773481406e-06,
      "loss": 2.567,
      "step": 2058
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5819295374372363,
      "learning_rate": 8.061762578008613e-06,
      "loss": 2.6298,
      "step": 2059
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5326175076295221,
      "learning_rate": 7.929412808860559e-06,
      "loss": 2.5034,
      "step": 2060
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5797491706004261,
      "learning_rate": 7.79814975832649e-06,
      "loss": 2.5665,
      "step": 2061
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.57193157490852,
      "learning_rate": 7.667973716295851e-06,
      "loss": 2.567,
      "step": 2062
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.538564404834179,
      "learning_rate": 7.5388849702571205e-06,
      "loss": 2.5525,
      "step": 2063
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5883414851403509,
      "learning_rate": 7.4108838052979185e-06,
      "loss": 2.5515,
      "step": 2064
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5478563367474956,
      "learning_rate": 7.283970504103732e-06,
      "loss": 2.5973,
      "step": 2065
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5495135966394802,
      "learning_rate": 7.1581453469575785e-06,
      "loss": 2.6312,
      "step": 2066
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5576200541002543,
      "learning_rate": 7.033408611739456e-06,
      "loss": 2.6026,
      "step": 2067
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5724250187046992,
      "learning_rate": 6.909760573925561e-06,
      "loss": 2.7012,
      "step": 2068
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5823113108747624,
      "learning_rate": 6.787201506587626e-06,
      "loss": 2.6885,
      "step": 2069
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.6150148014329102,
      "learning_rate": 6.66573168039264e-06,
      "loss": 2.6405,
      "step": 2070
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5703291555552336,
      "learning_rate": 6.545351363601959e-06,
      "loss": 2.6393,
      "step": 2071
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.558210757437288,
      "learning_rate": 6.426060822070812e-06,
      "loss": 2.6657,
      "step": 2072
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5769210607972206,
      "learning_rate": 6.3078603192475716e-06,
      "loss": 2.5515,
      "step": 2073
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5570014480852326,
      "learning_rate": 6.1907501161735934e-06,
      "loss": 2.594,
      "step": 2074
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5545379674171393,
      "learning_rate": 6.074730471482049e-06,
      "loss": 2.5655,
      "step": 2075
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.54907370488049,
      "learning_rate": 5.959801641397755e-06,
      "loss": 2.5946,
      "step": 2076
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5811989138676003,
      "learning_rate": 5.845963879736627e-06,
      "loss": 2.5009,
      "step": 2077
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5544534333870564,
      "learning_rate": 5.733217437904892e-06,
      "loss": 2.5362,
      "step": 2078
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.575375017553867,
      "learning_rate": 5.621562564898597e-06,
      "loss": 2.5593,
      "step": 2079
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.6016595488047402,
      "learning_rate": 5.51099950730316e-06,
      "loss": 2.6339,
      "step": 2080
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5474253167053842,
      "learning_rate": 5.401528509292763e-06,
      "loss": 2.6224,
      "step": 2081
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5530691588081266,
      "learning_rate": 5.2931498126298495e-06,
      "loss": 2.5955,
      "step": 2082
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5562050791935046,
      "learning_rate": 5.1858636566645135e-06,
      "loss": 2.6917,
      "step": 2083
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5217604552340692,
      "learning_rate": 5.0796702783340035e-06,
      "loss": 2.584,
      "step": 2084
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5666016806480639,
      "learning_rate": 4.97456991216233e-06,
      "loss": 2.5472,
      "step": 2085
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5652803803620395,
      "learning_rate": 4.870562790259325e-06,
      "loss": 2.5567,
      "step": 2086
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5842044033424149,
      "learning_rate": 4.7676491423208625e-06,
      "loss": 2.6401,
      "step": 2087
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.562770883080947,
      "learning_rate": 4.66582919562758e-06,
      "loss": 2.4939,
      "step": 2088
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5484103346356,
      "learning_rate": 4.5651031750448825e-06,
      "loss": 2.5868,
      "step": 2089
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.590502732538457,
      "learning_rate": 4.465471303022217e-06,
      "loss": 2.6658,
      "step": 2090
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5392488344319273,
      "learning_rate": 4.366933799592743e-06,
      "loss": 2.568,
      "step": 2091
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5601854038391295,
      "learning_rate": 4.269490882372551e-06,
      "loss": 2.6078,
      "step": 2092
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5369166153756106,
      "learning_rate": 4.1731427665606115e-06,
      "loss": 2.4947,
      "step": 2093
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5737511820047038,
      "learning_rate": 4.077889664937884e-06,
      "loss": 2.5969,
      "step": 2094
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.558281963674582,
      "learning_rate": 3.983731787867207e-06,
      "loss": 2.6377,
      "step": 2095
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5425270059103064,
      "learning_rate": 3.890669343292464e-06,
      "loss": 2.5631,
      "step": 2096
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5755389713792203,
      "learning_rate": 3.7987025367384743e-06,
      "loss": 2.6226,
      "step": 2097
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5271638126428261,
      "learning_rate": 3.707831571310327e-06,
      "loss": 2.5256,
      "step": 2098
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5530518242945796,
      "learning_rate": 3.6180566476929912e-06,
      "loss": 2.6348,
      "step": 2099
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5749689291026111,
      "learning_rate": 3.529377964150815e-06,
      "loss": 2.5372,
      "step": 2100
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5715509056340315,
      "learning_rate": 3.441795716527307e-06,
      "loss": 2.597,
      "step": 2101
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5557179996764093,
      "learning_rate": 3.355310098244302e-06,
      "loss": 2.5676,
      "step": 2102
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5876664933903689,
      "learning_rate": 3.269921300301959e-06,
      "loss": 2.6379,
      "step": 2103
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5727898864825025,
      "learning_rate": 3.1856295112780988e-06,
      "loss": 2.6334,
      "step": 2104
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.561437565959899,
      "learning_rate": 3.102434917327812e-06,
      "loss": 2.6564,
      "step": 2105
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5699702891480478,
      "learning_rate": 3.0203377021831292e-06,
      "loss": 2.6226,
      "step": 2106
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5501392716239842,
      "learning_rate": 2.939338047152573e-06,
      "loss": 2.5599,
      "step": 2107
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5578602440758822,
      "learning_rate": 2.8594361311206073e-06,
      "loss": 2.5811,
      "step": 2108
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.6037927835743853,
      "learning_rate": 2.7806321305475225e-06,
      "loss": 2.6589,
      "step": 2109
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5495922254467186,
      "learning_rate": 2.7029262194688818e-06,
      "loss": 2.5824,
      "step": 2110
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5994577535759268,
      "learning_rate": 2.626318569495134e-06,
      "loss": 2.5628,
      "step": 2111
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5491923830903185,
      "learning_rate": 2.550809349811334e-06,
      "loss": 2.477,
      "step": 2112
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5548717311043736,
      "learning_rate": 2.476398727176532e-06,
      "loss": 2.6395,
      "step": 2113
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5145742075401096,
      "learning_rate": 2.4030868659237204e-06,
      "loss": 2.6146,
      "step": 2114
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5690303563584321,
      "learning_rate": 2.3308739279593317e-06,
      "loss": 2.5855,
      "step": 2115
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5680869325547985,
      "learning_rate": 2.2597600727626845e-06,
      "loss": 2.657,
      "step": 2116
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.569408621593729,
      "learning_rate": 2.1897454573860387e-06,
      "loss": 2.6001,
      "step": 2117
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5975424714823333,
      "learning_rate": 2.1208302364538746e-06,
      "loss": 2.5924,
      "step": 2118
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5678170914754146,
      "learning_rate": 2.0530145621627804e-06,
      "loss": 2.5508,
      "step": 2119
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5969850906819759,
      "learning_rate": 1.9862985842810653e-06,
      "loss": 2.5829,
      "step": 2120
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5464213897357608,
      "learning_rate": 1.920682450148259e-06,
      "loss": 2.5638,
      "step": 2121
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.533204945886039,
      "learning_rate": 1.856166304675111e-06,
      "loss": 2.5597,
      "step": 2122
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5387409086860829,
      "learning_rate": 1.792750290342926e-06,
      "loss": 2.598,
      "step": 2123
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5321870162745318,
      "learning_rate": 1.7304345472035632e-06,
      "loss": 2.4648,
      "step": 2124
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.615407469644478,
      "learning_rate": 1.6692192128788253e-06,
      "loss": 2.7057,
      "step": 2125
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5640937389787656,
      "learning_rate": 1.6091044225604035e-06,
      "loss": 2.6419,
      "step": 2126
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5526559917564546,
      "learning_rate": 1.5500903090094888e-06,
      "loss": 2.5395,
      "step": 2127
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5787960082357315,
      "learning_rate": 1.492177002556383e-06,
      "loss": 2.7413,
      "step": 2128
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5334815177188844,
      "learning_rate": 1.4353646311004443e-06,
      "loss": 2.5712,
      "step": 2129
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5770218013300737,
      "learning_rate": 1.3796533201094752e-06,
      "loss": 2.6552,
      "step": 2130
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5544288424487838,
      "learning_rate": 1.3250431926197793e-06,
      "loss": 2.5616,
      "step": 2131
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5760819198964946,
      "learning_rate": 1.2715343692356607e-06,
      "loss": 2.623,
      "step": 2132
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5338414500489966,
      "learning_rate": 1.2191269681292582e-06,
      "loss": 2.5759,
      "step": 2133
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5828763685421412,
      "learning_rate": 1.1678211050402676e-06,
      "loss": 2.6175,
      "step": 2134
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5652881569850006,
      "learning_rate": 1.117616893275719e-06,
      "loss": 2.5748,
      "step": 2135
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5469540710208443,
      "learning_rate": 1.068514443709534e-06,
      "loss": 2.5444,
      "step": 2136
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5368199842213015,
      "learning_rate": 1.0205138647826905e-06,
      "loss": 2.5753,
      "step": 2137
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5865570276536973,
      "learning_rate": 9.73615262502503e-07,
      "loss": 2.642,
      "step": 2138
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5605668168900698,
      "learning_rate": 9.278187404426763e-07,
      "loss": 2.6098,
      "step": 2139
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5466697240552202,
      "learning_rate": 8.831243997431404e-07,
      "loss": 2.6533,
      "step": 2140
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5398369965045885,
      "learning_rate": 8.395323391094944e-07,
      "loss": 2.6,
      "step": 2141
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5639106271845373,
      "learning_rate": 7.970426548131183e-07,
      "loss": 2.6474,
      "step": 2142
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5692805816724555,
      "learning_rate": 7.556554406908389e-07,
      "loss": 2.6883,
      "step": 2143
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.582118094538976,
      "learning_rate": 7.153707881446536e-07,
      "loss": 2.6251,
      "step": 2144
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5917192021982479,
      "learning_rate": 6.761887861417293e-07,
      "loss": 2.5689,
      "step": 2145
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5584357193276607,
      "learning_rate": 6.381095212139032e-07,
      "loss": 2.5713,
      "step": 2146
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.5576038585260222,
      "learning_rate": 6.011330774577384e-07,
      "loss": 2.5544,
      "step": 2147
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5572460048731604,
      "learning_rate": 5.652595365343016e-07,
      "loss": 2.4765,
      "step": 2148
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5607024720979649,
      "learning_rate": 5.304889776688859e-07,
      "loss": 2.6253,
      "step": 2149
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5447052188232957,
      "learning_rate": 4.968214776508994e-07,
      "loss": 2.5313,
      "step": 2150
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5611454336654501,
      "learning_rate": 4.6425711083375454e-07,
      "loss": 2.6776,
      "step": 2151
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5466767361485623,
      "learning_rate": 4.3279594913447906e-07,
      "loss": 2.6717,
      "step": 2152
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5577187528376797,
      "learning_rate": 4.02438062033883e-07,
      "loss": 2.6044,
      "step": 2153
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5607446849435871,
      "learning_rate": 3.7318351657616987e-07,
      "loss": 2.5623,
      "step": 2154
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5517146351312021,
      "learning_rate": 3.4503237736882573e-07,
      "loss": 2.58,
      "step": 2155
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5603092292178588,
      "learning_rate": 3.179847065825081e-07,
      "loss": 2.5625,
      "step": 2156
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5693783620812112,
      "learning_rate": 2.9204056395104594e-07,
      "loss": 2.4685,
      "step": 2157
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5637952579264054,
      "learning_rate": 2.672000067709956e-07,
      "loss": 2.5847,
      "step": 2158
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5325411946842509,
      "learning_rate": 2.4346308990175204e-07,
      "loss": 2.5384,
      "step": 2159
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5609540323280334,
      "learning_rate": 2.208298657653818e-07,
      "loss": 2.6109,
      "step": 2160
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5588085941871965,
      "learning_rate": 1.9930038434645692e-07,
      "loss": 2.5886,
      "step": 2161
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5574497203529588,
      "learning_rate": 1.7887469319205484e-07,
      "loss": 2.6863,
      "step": 2162
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.550018806988555,
      "learning_rate": 1.5955283741142523e-07,
      "loss": 2.673,
      "step": 2163
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5132920005683055,
      "learning_rate": 1.4133485967615655e-07,
      "loss": 2.4689,
      "step": 2164
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5446178222938799,
      "learning_rate": 1.2422080021995407e-07,
      "loss": 2.5621,
      "step": 2165
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5697766399990973,
      "learning_rate": 1.082106968385288e-07,
      "loss": 2.6783,
      "step": 2166
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5854048181430894,
      "learning_rate": 9.330458488959748e-08,
      "loss": 2.5511,
      "step": 2167
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5579299065791804,
      "learning_rate": 7.950249729271608e-08,
      "loss": 2.5943,
      "step": 2168
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.568465917328927,
      "learning_rate": 6.680446452922429e-08,
      "loss": 2.6458,
      "step": 2169
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5568779990325727,
      "learning_rate": 5.521051464230098e-08,
      "loss": 2.7003,
      "step": 2170
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.576973229349878,
      "learning_rate": 4.4720673236631206e-08,
      "loss": 2.5554,
      "step": 2171
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.6097422706224399,
      "learning_rate": 3.53349634786837e-08,
      "loss": 2.5406,
      "step": 2172
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.560346128797687,
      "learning_rate": 2.7053406096433365e-08,
      "loss": 2.6486,
      "step": 2173
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.554247394722822,
      "learning_rate": 1.987601937930572e-08,
      "loss": 2.5732,
      "step": 2174
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5801201749039946,
      "learning_rate": 1.3802819178398984e-08,
      "loss": 2.6181,
      "step": 2175
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5638533512464483,
      "learning_rate": 8.833818906039959e-09,
      "loss": 2.5228,
      "step": 2176
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5449136104275335,
      "learning_rate": 4.969029536061598e-09,
      "loss": 2.535,
      "step": 2177
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5326423483420331,
      "learning_rate": 2.2084596038030037e-09,
      "loss": 2.6149,
      "step": 2178
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5863816211561724,
      "learning_rate": 5.521152057763601e-10,
      "loss": 2.7574,
      "step": 2179
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5474329241291628,
      "learning_rate": 0.0,
      "loss": 2.603,
      "step": 2180
    },
    {
      "epoch": 1.0,
      "step": 2180,
      "total_flos": 5.89735773670998e+17,
      "train_loss": 2.8288836673859064,
      "train_runtime": 92131.8092,
      "train_samples_per_second": 6.058,
      "train_steps_per_second": 0.024
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 2180,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "total_flos": 5.89735773670998e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}