{
  "best_metric": 0.4351051665913261,
  "best_model_checkpoint": "/m/triton/scratch/elec/puhe/p/palp3/MUCS/indicwav2vec_outputs/pd_warmup_500/s300_shuff100/checkpoint-1000",
  "epoch": 1.6,
  "eval_steps": 1000,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0016,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 65.7309,
      "step": 1
    },
    {
      "epoch": 0.0032,
      "grad_norm": 19.431121826171875,
      "learning_rate": 1.2e-06,
      "loss": 40.497,
      "step": 2
    },
    {
      "epoch": 0.0048,
      "grad_norm": 20.21518898010254,
      "learning_rate": 2.4e-06,
      "loss": 36.2325,
      "step": 3
    },
    {
      "epoch": 0.0064,
      "grad_norm": 11.604636192321777,
      "learning_rate": 3.6e-06,
      "loss": 27.9292,
      "step": 4
    },
    {
      "epoch": 0.008,
      "grad_norm": NaN,
      "learning_rate": 3.6e-06,
      "loss": 27.842,
      "step": 5
    },
    {
      "epoch": 0.0096,
      "grad_norm": 13.326777458190918,
      "learning_rate": 4.8e-06,
      "loss": 28.3298,
      "step": 6
    },
    {
      "epoch": 0.0112,
      "grad_norm": 12.479119300842285,
      "learning_rate": 5.999999999999999e-06,
      "loss": 28.7461,
      "step": 7
    },
    {
      "epoch": 0.0128,
      "grad_norm": 11.469672203063965,
      "learning_rate": 7.2e-06,
      "loss": 26.4075,
      "step": 8
    },
    {
      "epoch": 0.0144,
      "grad_norm": 10.416142463684082,
      "learning_rate": 8.4e-06,
      "loss": 23.0338,
      "step": 9
    },
    {
      "epoch": 0.016,
      "grad_norm": 10.866898536682129,
      "learning_rate": 9.6e-06,
      "loss": 25.844,
      "step": 10
    },
    {
      "epoch": 0.0176,
      "grad_norm": 11.97667121887207,
      "learning_rate": 1.0799999999999998e-05,
      "loss": 24.9471,
      "step": 11
    },
    {
      "epoch": 0.0192,
      "grad_norm": 10.51934814453125,
      "learning_rate": 1.1999999999999999e-05,
      "loss": 24.6028,
      "step": 12
    },
    {
      "epoch": 0.0208,
      "grad_norm": 10.643861770629883,
      "learning_rate": 1.3199999999999997e-05,
      "loss": 23.1144,
      "step": 13
    },
    {
      "epoch": 0.0224,
      "grad_norm": 11.642038345336914,
      "learning_rate": 1.44e-05,
      "loss": 23.0041,
      "step": 14
    },
    {
      "epoch": 0.024,
      "grad_norm": 10.239847183227539,
      "learning_rate": 1.5599999999999996e-05,
      "loss": 22.1151,
      "step": 15
    },
    {
      "epoch": 0.0256,
      "grad_norm": 10.31641960144043,
      "learning_rate": 1.68e-05,
      "loss": 20.955,
      "step": 16
    },
    {
      "epoch": 0.0272,
      "grad_norm": 10.14737606048584,
      "learning_rate": 1.7999999999999997e-05,
      "loss": 20.0026,
      "step": 17
    },
    {
      "epoch": 0.0288,
      "grad_norm": 10.14389419555664,
      "learning_rate": 1.92e-05,
      "loss": 20.053,
      "step": 18
    },
    {
      "epoch": 0.0304,
      "grad_norm": 13.049298286437988,
      "learning_rate": 2.04e-05,
      "loss": 23.2289,
      "step": 19
    },
    {
      "epoch": 0.032,
      "grad_norm": 10.918747901916504,
      "learning_rate": 2.1599999999999996e-05,
      "loss": 20.6362,
      "step": 20
    },
    {
      "epoch": 0.0336,
      "grad_norm": 10.782907485961914,
      "learning_rate": 2.28e-05,
      "loss": 19.6928,
      "step": 21
    },
    {
      "epoch": 0.0352,
      "grad_norm": 13.15986156463623,
      "learning_rate": 2.3999999999999997e-05,
      "loss": 22.1249,
      "step": 22
    },
    {
      "epoch": 0.0368,
      "grad_norm": 11.684220314025879,
      "learning_rate": 2.52e-05,
      "loss": 20.4745,
      "step": 23
    },
    {
      "epoch": 0.0384,
      "grad_norm": 11.818272590637207,
      "learning_rate": 2.6399999999999995e-05,
      "loss": 20.3504,
      "step": 24
    },
    {
      "epoch": 0.04,
      "grad_norm": 12.81021499633789,
      "learning_rate": 2.7599999999999997e-05,
      "loss": 21.0093,
      "step": 25
    },
    {
      "epoch": 0.0416,
      "grad_norm": 14.015459060668945,
      "learning_rate": 2.88e-05,
      "loss": 22.1517,
      "step": 26
    },
    {
      "epoch": 0.0432,
      "grad_norm": 11.829425811767578,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 18.2946,
      "step": 27
    },
    {
      "epoch": 0.0448,
      "grad_norm": 15.79493236541748,
      "learning_rate": 3.119999999999999e-05,
      "loss": 22.9019,
      "step": 28
    },
    {
      "epoch": 0.0464,
      "grad_norm": 14.349289894104004,
      "learning_rate": 3.2399999999999995e-05,
      "loss": 20.221,
      "step": 29
    },
    {
      "epoch": 0.048,
      "grad_norm": 14.111164093017578,
      "learning_rate": 3.36e-05,
      "loss": 18.9742,
      "step": 30
    },
    {
      "epoch": 0.0496,
      "grad_norm": 15.603620529174805,
      "learning_rate": 3.48e-05,
      "loss": 20.2483,
      "step": 31
    },
    {
      "epoch": 0.0512,
      "grad_norm": 16.44049072265625,
      "learning_rate": 3.5999999999999994e-05,
      "loss": 19.9449,
      "step": 32
    },
    {
      "epoch": 0.0528,
      "grad_norm": 18.68276596069336,
      "learning_rate": 3.7199999999999996e-05,
      "loss": 21.1125,
      "step": 33
    },
    {
      "epoch": 0.0544,
      "grad_norm": 14.831990242004395,
      "learning_rate": 3.84e-05,
      "loss": 17.0528,
      "step": 34
    },
    {
      "epoch": 0.056,
      "grad_norm": 16.635009765625,
      "learning_rate": 3.96e-05,
      "loss": 18.3324,
      "step": 35
    },
    {
      "epoch": 0.0576,
      "grad_norm": 18.090103149414062,
      "learning_rate": 4.08e-05,
      "loss": 18.495,
      "step": 36
    },
    {
      "epoch": 0.0592,
      "grad_norm": 19.209562301635742,
      "learning_rate": 4.2e-05,
      "loss": 18.7353,
      "step": 37
    },
    {
      "epoch": 0.0608,
      "grad_norm": 19.63134765625,
      "learning_rate": 4.319999999999999e-05,
      "loss": 17.8893,
      "step": 38
    },
    {
      "epoch": 0.0624,
      "grad_norm": 20.713550567626953,
      "learning_rate": 4.4399999999999995e-05,
      "loss": 17.9679,
      "step": 39
    },
    {
      "epoch": 0.064,
      "grad_norm": 25.246625900268555,
      "learning_rate": 4.56e-05,
      "loss": 19.6261,
      "step": 40
    },
    {
      "epoch": 0.0656,
      "grad_norm": 28.108665466308594,
      "learning_rate": 4.68e-05,
      "loss": 20.795,
      "step": 41
    },
    {
      "epoch": 0.0672,
      "grad_norm": 26.678646087646484,
      "learning_rate": 4.7999999999999994e-05,
      "loss": 19.3122,
      "step": 42
    },
    {
      "epoch": 0.0688,
      "grad_norm": 28.03876304626465,
      "learning_rate": 4.9199999999999997e-05,
      "loss": 19.0949,
      "step": 43
    },
    {
      "epoch": 0.0704,
      "grad_norm": 28.40298080444336,
      "learning_rate": 5.04e-05,
      "loss": 18.7253,
      "step": 44
    },
    {
      "epoch": 0.072,
      "grad_norm": 30.833574295043945,
      "learning_rate": 5.1599999999999994e-05,
      "loss": 18.9767,
      "step": 45
    },
    {
      "epoch": 0.0736,
      "grad_norm": 25.621904373168945,
      "learning_rate": 5.279999999999999e-05,
      "loss": 16.0377,
      "step": 46
    },
    {
      "epoch": 0.0752,
      "grad_norm": 26.582237243652344,
      "learning_rate": 5.399999999999999e-05,
      "loss": 15.6131,
      "step": 47
    },
    {
      "epoch": 0.0768,
      "grad_norm": 34.89401626586914,
      "learning_rate": 5.519999999999999e-05,
      "loss": 17.8755,
      "step": 48
    },
    {
      "epoch": 0.0784,
      "grad_norm": 32.71629333496094,
      "learning_rate": 5.6399999999999995e-05,
      "loss": 16.0825,
      "step": 49
    },
    {
      "epoch": 0.08,
      "grad_norm": 27.743824005126953,
      "learning_rate": 5.76e-05,
      "loss": 13.6992,
      "step": 50
    },
    {
      "epoch": 0.0816,
      "grad_norm": Infinity,
      "learning_rate": 5.76e-05,
      "loss": 30.6838,
      "step": 51
    },
    {
      "epoch": 0.0832,
      "grad_norm": 85.50408935546875,
      "learning_rate": 5.88e-05,
      "loss": 26.4379,
      "step": 52
    },
    {
      "epoch": 0.0848,
      "grad_norm": Infinity,
      "learning_rate": 5.88e-05,
      "loss": 27.0726,
      "step": 53
    },
    {
      "epoch": 0.0864,
      "grad_norm": 50.16768264770508,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 18.3707,
      "step": 54
    },
    {
      "epoch": 0.088,
      "grad_norm": 131.34494018554688,
      "learning_rate": 6.12e-05,
      "loss": 23.1862,
      "step": 55
    },
    {
      "epoch": 0.0896,
      "grad_norm": 44.524417877197266,
      "learning_rate": 6.239999999999999e-05,
      "loss": 14.7586,
      "step": 56
    },
    {
      "epoch": 0.0912,
      "grad_norm": 71.3255615234375,
      "learning_rate": 6.359999999999999e-05,
      "loss": 19.4668,
      "step": 57
    },
    {
      "epoch": 0.0928,
      "grad_norm": 43.701072692871094,
      "learning_rate": 6.479999999999999e-05,
      "loss": 13.1064,
      "step": 58
    },
    {
      "epoch": 0.0944,
      "grad_norm": 45.97714614868164,
      "learning_rate": 6.599999999999999e-05,
      "loss": 12.3225,
      "step": 59
    },
    {
      "epoch": 0.096,
      "grad_norm": 49.00947952270508,
      "learning_rate": 6.72e-05,
      "loss": 12.3515,
      "step": 60
    },
    {
      "epoch": 0.0976,
      "grad_norm": 54.81338119506836,
      "learning_rate": 6.84e-05,
      "loss": 12.8657,
      "step": 61
    },
    {
      "epoch": 0.0992,
      "grad_norm": 44.2459831237793,
      "learning_rate": 6.96e-05,
      "loss": 10.2856,
      "step": 62
    },
    {
      "epoch": 0.1008,
      "grad_norm": 73.28941345214844,
      "learning_rate": 7.079999999999999e-05,
      "loss": 9.7418,
      "step": 63
    },
    {
      "epoch": 0.1024,
      "grad_norm": 40.017478942871094,
      "learning_rate": 7.199999999999999e-05,
      "loss": 8.9116,
      "step": 64
    },
    {
      "epoch": 0.104,
      "grad_norm": 35.54833221435547,
      "learning_rate": 7.319999999999999e-05,
      "loss": 7.9677,
      "step": 65
    },
    {
      "epoch": 0.1056,
      "grad_norm": 42.94904708862305,
      "learning_rate": 7.439999999999999e-05,
      "loss": 8.5134,
      "step": 66
    },
    {
      "epoch": 0.1072,
      "grad_norm": 38.947715759277344,
      "learning_rate": 7.56e-05,
      "loss": 7.7674,
      "step": 67
    },
    {
      "epoch": 0.1088,
      "grad_norm": 35.40559768676758,
      "learning_rate": 7.68e-05,
      "loss": 7.2294,
      "step": 68
    },
    {
      "epoch": 0.1104,
      "grad_norm": 31.678964614868164,
      "learning_rate": 7.8e-05,
      "loss": 6.7146,
      "step": 69
    },
    {
      "epoch": 0.112,
      "grad_norm": 31.80811309814453,
      "learning_rate": 7.92e-05,
      "loss": 6.5397,
      "step": 70
    },
    {
      "epoch": 0.1136,
      "grad_norm": 25.740388870239258,
      "learning_rate": 8.04e-05,
      "loss": 6.0079,
      "step": 71
    },
    {
      "epoch": 0.1152,
      "grad_norm": 23.663726806640625,
      "learning_rate": 8.16e-05,
      "loss": 5.7781,
      "step": 72
    },
    {
      "epoch": 0.1168,
      "grad_norm": 20.862640380859375,
      "learning_rate": 8.28e-05,
      "loss": 5.5184,
      "step": 73
    },
    {
      "epoch": 0.1184,
      "grad_norm": 17.784168243408203,
      "learning_rate": 8.4e-05,
      "loss": 5.3262,
      "step": 74
    },
    {
      "epoch": 0.12,
      "grad_norm": 16.173917770385742,
      "learning_rate": 8.519999999999998e-05,
      "loss": 5.2462,
      "step": 75
    },
    {
      "epoch": 0.1216,
      "grad_norm": 11.109197616577148,
      "learning_rate": 8.639999999999999e-05,
      "loss": 4.989,
      "step": 76
    },
    {
      "epoch": 0.1232,
      "grad_norm": 9.369626998901367,
      "learning_rate": 8.759999999999999e-05,
      "loss": 4.9122,
      "step": 77
    },
    {
      "epoch": 0.1248,
      "grad_norm": 9.535905838012695,
      "learning_rate": 8.879999999999999e-05,
      "loss": 4.9956,
      "step": 78
    },
    {
      "epoch": 0.1264,
      "grad_norm": 3.752645969390869,
      "learning_rate": 8.999999999999999e-05,
      "loss": 4.7227,
      "step": 79
    },
    {
      "epoch": 0.128,
      "grad_norm": 4.03261137008667,
      "learning_rate": 9.12e-05,
      "loss": 4.7404,
      "step": 80
    },
    {
      "epoch": 0.1296,
      "grad_norm": 3.3785626888275146,
      "learning_rate": 9.24e-05,
      "loss": 4.6154,
      "step": 81
    },
    {
      "epoch": 0.1312,
      "grad_norm": 3.4677908420562744,
      "learning_rate": 9.36e-05,
      "loss": 4.6707,
      "step": 82
    },
    {
      "epoch": 0.1328,
      "grad_norm": 4.760728359222412,
      "learning_rate": 9.479999999999999e-05,
      "loss": 4.5489,
      "step": 83
    },
    {
      "epoch": 0.1344,
      "grad_norm": 2.995011329650879,
      "learning_rate": 9.599999999999999e-05,
      "loss": 4.5604,
      "step": 84
    },
    {
      "epoch": 0.136,
      "grad_norm": 3.4023890495300293,
      "learning_rate": 9.719999999999999e-05,
      "loss": 4.4197,
      "step": 85
    },
    {
      "epoch": 0.1376,
      "grad_norm": 3.294135570526123,
      "learning_rate": 9.839999999999999e-05,
      "loss": 4.3887,
      "step": 86
    },
    {
      "epoch": 0.1392,
      "grad_norm": 2.753955841064453,
      "learning_rate": 9.96e-05,
      "loss": 4.3751,
      "step": 87
    },
    {
      "epoch": 0.1408,
      "grad_norm": 3.3620405197143555,
      "learning_rate": 0.0001008,
      "loss": 4.2838,
      "step": 88
    },
    {
      "epoch": 0.1424,
      "grad_norm": 2.385225534439087,
      "learning_rate": 0.000102,
      "loss": 4.183,
      "step": 89
    },
    {
      "epoch": 0.144,
      "grad_norm": 2.7207183837890625,
      "learning_rate": 0.00010319999999999999,
      "loss": 4.1382,
      "step": 90
    },
    {
      "epoch": 0.1456,
      "grad_norm": 2.0092227458953857,
      "learning_rate": 0.00010439999999999999,
      "loss": 4.1519,
      "step": 91
    },
    {
      "epoch": 0.1472,
      "grad_norm": 1.8385318517684937,
      "learning_rate": 0.00010559999999999998,
      "loss": 4.136,
      "step": 92
    },
    {
      "epoch": 0.1488,
      "grad_norm": 1.3982303142547607,
      "learning_rate": 0.00010679999999999998,
      "loss": 4.084,
      "step": 93
    },
    {
      "epoch": 0.1504,
      "grad_norm": 1.4304462671279907,
      "learning_rate": 0.00010799999999999998,
      "loss": 4.0993,
      "step": 94
    },
    {
      "epoch": 0.152,
      "grad_norm": 5.04845666885376,
      "learning_rate": 0.00010919999999999998,
      "loss": 4.1147,
      "step": 95
    },
    {
      "epoch": 0.1536,
      "grad_norm": 1.1188616752624512,
      "learning_rate": 0.00011039999999999999,
      "loss": 4.0138,
      "step": 96
    },
    {
      "epoch": 0.1552,
      "grad_norm": 2.9660868644714355,
      "learning_rate": 0.00011159999999999999,
      "loss": 4.0886,
      "step": 97
    },
    {
      "epoch": 0.1568,
      "grad_norm": 2.909925699234009,
      "learning_rate": 0.00011279999999999999,
      "loss": 4.0743,
      "step": 98
    },
    {
      "epoch": 0.1584,
      "grad_norm": 2.233374834060669,
      "learning_rate": 0.00011399999999999999,
      "loss": 4.0643,
      "step": 99
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.9833197593688965,
      "learning_rate": 0.0001152,
      "loss": 4.1569,
      "step": 100
    },
    {
      "epoch": 0.1616,
      "grad_norm": 31.450782775878906,
      "learning_rate": 0.0001164,
      "loss": 4.756,
      "step": 101
    },
    {
      "epoch": 0.1632,
      "grad_norm": 11.571601867675781,
      "learning_rate": 0.0001176,
      "loss": 4.0323,
      "step": 102
    },
    {
      "epoch": 0.1648,
      "grad_norm": 9.144487380981445,
      "learning_rate": 0.0001188,
      "loss": 3.9735,
      "step": 103
    },
    {
      "epoch": 0.1664,
      "grad_norm": 16.850412368774414,
      "learning_rate": 0.00011999999999999999,
      "loss": 4.218,
      "step": 104
    },
    {
      "epoch": 0.168,
      "grad_norm": 2.83750319480896,
      "learning_rate": 0.00012119999999999999,
      "loss": 3.9086,
      "step": 105
    },
    {
      "epoch": 0.1696,
      "grad_norm": 1.1910499334335327,
      "learning_rate": 0.0001224,
      "loss": 3.8875,
      "step": 106
    },
    {
      "epoch": 0.1712,
      "grad_norm": 1.7703046798706055,
      "learning_rate": 0.0001236,
      "loss": 3.9357,
      "step": 107
    },
    {
      "epoch": 0.1728,
      "grad_norm": 4.045347690582275,
      "learning_rate": 0.00012479999999999997,
      "loss": 3.8306,
      "step": 108
    },
    {
      "epoch": 0.1744,
      "grad_norm": 3.414834499359131,
      "learning_rate": 0.00012599999999999997,
      "loss": 3.8273,
      "step": 109
    },
    {
      "epoch": 0.176,
      "grad_norm": 5.279017925262451,
      "learning_rate": 0.00012719999999999997,
      "loss": 3.8331,
      "step": 110
    },
    {
      "epoch": 0.1776,
      "grad_norm": 2.1371779441833496,
      "learning_rate": 0.00012839999999999998,
      "loss": 3.8236,
      "step": 111
    },
    {
      "epoch": 0.1792,
      "grad_norm": 0.6024636030197144,
      "learning_rate": 0.00012959999999999998,
      "loss": 3.7802,
      "step": 112
    },
    {
      "epoch": 0.1808,
      "grad_norm": 3.284954786300659,
      "learning_rate": 0.00013079999999999998,
      "loss": 3.837,
      "step": 113
    },
    {
      "epoch": 0.1824,
      "grad_norm": 3.2697064876556396,
      "learning_rate": 0.00013199999999999998,
      "loss": 3.8099,
      "step": 114
    },
    {
      "epoch": 0.184,
      "grad_norm": 2.267110824584961,
      "learning_rate": 0.00013319999999999999,
      "loss": 3.7916,
      "step": 115
    },
    {
      "epoch": 0.1856,
      "grad_norm": 0.9686329364776611,
      "learning_rate": 0.0001344,
      "loss": 3.7612,
      "step": 116
    },
    {
      "epoch": 0.1872,
      "grad_norm": 1.1181795597076416,
      "learning_rate": 0.0001356,
      "loss": 3.817,
      "step": 117
    },
    {
      "epoch": 0.1888,
      "grad_norm": 4.291112899780273,
      "learning_rate": 0.0001368,
      "loss": 3.75,
      "step": 118
    },
    {
      "epoch": 0.1904,
      "grad_norm": 1.1991240978240967,
      "learning_rate": 0.000138,
      "loss": 3.8755,
      "step": 119
    },
    {
      "epoch": 0.192,
      "grad_norm": 1.800611138343811,
      "learning_rate": 0.0001392,
      "loss": 3.7636,
      "step": 120
    },
    {
      "epoch": 0.1936,
      "grad_norm": 0.6300956606864929,
      "learning_rate": 0.0001404,
      "loss": 3.8781,
      "step": 121
    },
    {
      "epoch": 0.1952,
      "grad_norm": 1.9492263793945312,
      "learning_rate": 0.00014159999999999997,
      "loss": 3.758,
      "step": 122
    },
    {
      "epoch": 0.1968,
      "grad_norm": 1.5942705869674683,
      "learning_rate": 0.00014279999999999997,
      "loss": 3.7344,
      "step": 123
    },
    {
      "epoch": 0.1984,
      "grad_norm": 1.0828086137771606,
      "learning_rate": 0.00014399999999999998,
      "loss": 3.7535,
      "step": 124
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.54654860496521,
      "learning_rate": 0.00014519999999999998,
      "loss": 3.7764,
      "step": 125
    },
    {
      "epoch": 0.2016,
      "grad_norm": 3.488430976867676,
      "learning_rate": 0.00014639999999999998,
      "loss": 3.8125,
      "step": 126
    },
    {
      "epoch": 0.2032,
      "grad_norm": 1.4746885299682617,
      "learning_rate": 0.00014759999999999998,
      "loss": 3.8042,
      "step": 127
    },
    {
      "epoch": 0.2048,
      "grad_norm": 2.114961862564087,
      "learning_rate": 0.00014879999999999998,
      "loss": 3.7654,
      "step": 128
    },
    {
      "epoch": 0.2064,
      "grad_norm": 1.086930751800537,
      "learning_rate": 0.00015,
      "loss": 3.797,
      "step": 129
    },
    {
      "epoch": 0.208,
      "grad_norm": 0.5767809152603149,
      "learning_rate": 0.0001512,
      "loss": 3.7507,
      "step": 130
    },
    {
      "epoch": 0.2096,
      "grad_norm": 2.0122199058532715,
      "learning_rate": 0.0001524,
      "loss": 3.7508,
      "step": 131
    },
    {
      "epoch": 0.2112,
      "grad_norm": 3.8814480304718018,
      "learning_rate": 0.0001536,
      "loss": 3.8003,
      "step": 132
    },
    {
      "epoch": 0.2128,
      "grad_norm": 1.92637038230896,
      "learning_rate": 0.0001548,
      "loss": 3.8009,
      "step": 133
    },
    {
      "epoch": 0.2144,
      "grad_norm": 0.7897951006889343,
      "learning_rate": 0.000156,
      "loss": 3.7989,
      "step": 134
    },
    {
      "epoch": 0.216,
      "grad_norm": 4.2987751960754395,
      "learning_rate": 0.0001572,
      "loss": 3.7605,
      "step": 135
    },
    {
      "epoch": 0.2176,
      "grad_norm": 1.7444305419921875,
      "learning_rate": 0.0001584,
      "loss": 3.8734,
      "step": 136
    },
    {
      "epoch": 0.2192,
      "grad_norm": 3.6664788722991943,
      "learning_rate": 0.0001596,
      "loss": 3.7828,
      "step": 137
    },
    {
      "epoch": 0.2208,
      "grad_norm": 1.8154056072235107,
      "learning_rate": 0.0001608,
      "loss": 3.9124,
      "step": 138
    },
    {
      "epoch": 0.2224,
      "grad_norm": 0.8211868405342102,
      "learning_rate": 0.000162,
      "loss": 3.7861,
      "step": 139
    },
    {
      "epoch": 0.224,
      "grad_norm": 1.5533865690231323,
      "learning_rate": 0.0001632,
      "loss": 3.787,
      "step": 140
    },
    {
      "epoch": 0.2256,
      "grad_norm": 2.957888603210449,
      "learning_rate": 0.0001644,
      "loss": 3.7569,
      "step": 141
    },
    {
      "epoch": 0.2272,
      "grad_norm": 3.3956074714660645,
      "learning_rate": 0.0001656,
      "loss": 3.854,
      "step": 142
    },
    {
      "epoch": 0.2288,
      "grad_norm": 4.3899993896484375,
      "learning_rate": 0.0001668,
      "loss": 3.8498,
      "step": 143
    },
    {
      "epoch": 0.2304,
      "grad_norm": 2.408184051513672,
      "learning_rate": 0.000168,
      "loss": 3.7483,
      "step": 144
    },
    {
      "epoch": 0.232,
      "grad_norm": 2.7876155376434326,
      "learning_rate": 0.00016919999999999997,
      "loss": 3.7817,
      "step": 145
    },
    {
      "epoch": 0.2336,
      "grad_norm": 2.2329955101013184,
      "learning_rate": 0.00017039999999999997,
      "loss": 3.7676,
      "step": 146
    },
    {
      "epoch": 0.2352,
      "grad_norm": 1.5696635246276855,
      "learning_rate": 0.00017159999999999997,
      "loss": 3.811,
      "step": 147
    },
    {
      "epoch": 0.2368,
      "grad_norm": 2.209275007247925,
      "learning_rate": 0.00017279999999999997,
      "loss": 3.8107,
      "step": 148
    },
    {
      "epoch": 0.2384,
      "grad_norm": 1.1010030508041382,
      "learning_rate": 0.00017399999999999997,
      "loss": 3.8716,
      "step": 149
    },
    {
      "epoch": 0.24,
      "grad_norm": 3.2085022926330566,
      "learning_rate": 0.00017519999999999998,
      "loss": 3.9153,
      "step": 150
    },
    {
      "epoch": 0.2416,
      "grad_norm": 27.279512405395508,
      "learning_rate": 0.00017639999999999998,
      "loss": 4.6372,
      "step": 151
    },
    {
      "epoch": 0.2432,
      "grad_norm": 6.734384536743164,
      "learning_rate": 0.00017759999999999998,
      "loss": 3.8794,
      "step": 152
    },
    {
      "epoch": 0.2448,
      "grad_norm": 8.395252227783203,
      "learning_rate": 0.00017879999999999998,
      "loss": 3.9047,
      "step": 153
    },
    {
      "epoch": 0.2464,
      "grad_norm": 2.386127471923828,
      "learning_rate": 0.00017999999999999998,
      "loss": 3.8455,
      "step": 154
    },
    {
      "epoch": 0.248,
      "grad_norm": 5.339539051055908,
      "learning_rate": 0.00018119999999999999,
      "loss": 4.004,
      "step": 155
    },
    {
      "epoch": 0.2496,
      "grad_norm": 6.078152179718018,
      "learning_rate": 0.0001824,
      "loss": 3.8299,
      "step": 156
    },
    {
      "epoch": 0.2512,
      "grad_norm": 6.789322853088379,
      "learning_rate": 0.0001836,
      "loss": 3.863,
      "step": 157
    },
    {
      "epoch": 0.2528,
      "grad_norm": 6.238964557647705,
      "learning_rate": 0.0001848,
      "loss": 3.8967,
      "step": 158
    },
    {
      "epoch": 0.2544,
      "grad_norm": 4.713714122772217,
      "learning_rate": 0.000186,
      "loss": 3.8696,
      "step": 159
    },
    {
      "epoch": 0.256,
      "grad_norm": 2.388154983520508,
      "learning_rate": 0.0001872,
      "loss": 3.752,
      "step": 160
    },
    {
      "epoch": 0.2576,
      "grad_norm": 3.3313629627227783,
      "learning_rate": 0.00018839999999999997,
      "loss": 3.7579,
      "step": 161
    },
    {
      "epoch": 0.2592,
      "grad_norm": 5.692492485046387,
      "learning_rate": 0.00018959999999999997,
      "loss": 3.7794,
      "step": 162
    },
    {
      "epoch": 0.2608,
      "grad_norm": 5.304704189300537,
      "learning_rate": 0.00019079999999999998,
      "loss": 3.8021,
      "step": 163
    },
    {
      "epoch": 0.2624,
      "grad_norm": 5.1804118156433105,
      "learning_rate": 0.00019199999999999998,
      "loss": 3.7633,
      "step": 164
    },
    {
      "epoch": 0.264,
      "grad_norm": 3.4341394901275635,
      "learning_rate": 0.00019319999999999998,
      "loss": 3.802,
      "step": 165
    },
    {
      "epoch": 0.2656,
      "grad_norm": 0.7816358804702759,
      "learning_rate": 0.00019439999999999998,
      "loss": 3.7327,
      "step": 166
    },
    {
      "epoch": 0.2672,
      "grad_norm": 3.8968913555145264,
      "learning_rate": 0.00019559999999999998,
      "loss": 3.7701,
      "step": 167
    },
    {
      "epoch": 0.2688,
      "grad_norm": 6.041684150695801,
      "learning_rate": 0.00019679999999999999,
      "loss": 3.7651,
      "step": 168
    },
    {
      "epoch": 0.2704,
      "grad_norm": 1.5194474458694458,
      "learning_rate": 0.000198,
      "loss": 3.9153,
      "step": 169
    },
    {
      "epoch": 0.272,
      "grad_norm": 5.066249847412109,
      "learning_rate": 0.0001992,
      "loss": 3.7799,
      "step": 170
    },
    {
      "epoch": 0.2736,
      "grad_norm": 2.518998146057129,
      "learning_rate": 0.0002004,
      "loss": 3.7326,
      "step": 171
    },
    {
      "epoch": 0.2752,
      "grad_norm": 0.759394645690918,
      "learning_rate": 0.0002016,
      "loss": 3.7889,
      "step": 172
    },
    {
      "epoch": 0.2768,
      "grad_norm": 3.221085786819458,
      "learning_rate": 0.0002028,
      "loss": 3.709,
      "step": 173
    },
    {
      "epoch": 0.2784,
      "grad_norm": 5.799932956695557,
      "learning_rate": 0.000204,
      "loss": 3.8104,
      "step": 174
    },
    {
      "epoch": 0.28,
      "grad_norm": 4.494676113128662,
      "learning_rate": 0.0002052,
      "loss": 3.7602,
      "step": 175
    },
    {
      "epoch": 0.2816,
      "grad_norm": 1.4825067520141602,
      "learning_rate": 0.00020639999999999998,
      "loss": 3.6961,
      "step": 176
    },
    {
      "epoch": 0.2832,
      "grad_norm": 1.2989227771759033,
      "learning_rate": 0.00020759999999999998,
      "loss": 3.7053,
      "step": 177
    },
    {
      "epoch": 0.2848,
      "grad_norm": 2.4774816036224365,
      "learning_rate": 0.00020879999999999998,
      "loss": 3.7072,
      "step": 178
    },
    {
      "epoch": 0.2864,
      "grad_norm": 0.7821967005729675,
      "learning_rate": 0.00020999999999999998,
      "loss": 3.7205,
      "step": 179
    },
    {
      "epoch": 0.288,
      "grad_norm": 2.0641937255859375,
      "learning_rate": 0.00021119999999999996,
      "loss": 3.7218,
      "step": 180
    },
    {
      "epoch": 0.2896,
      "grad_norm": 0.7465824484825134,
      "learning_rate": 0.00021239999999999996,
      "loss": 3.7159,
      "step": 181
    },
    {
      "epoch": 0.2912,
      "grad_norm": 1.9227252006530762,
      "learning_rate": 0.00021359999999999996,
      "loss": 3.7154,
      "step": 182
    },
    {
      "epoch": 0.2928,
      "grad_norm": 2.588310480117798,
      "learning_rate": 0.00021479999999999996,
      "loss": 3.7532,
      "step": 183
    },
    {
      "epoch": 0.2944,
      "grad_norm": 0.8672159314155579,
      "learning_rate": 0.00021599999999999996,
      "loss": 3.7199,
      "step": 184
    },
    {
      "epoch": 0.296,
      "grad_norm": 2.015648365020752,
      "learning_rate": 0.00021719999999999997,
      "loss": 3.7661,
      "step": 185
    },
    {
      "epoch": 0.2976,
      "grad_norm": 1.280931830406189,
      "learning_rate": 0.00021839999999999997,
      "loss": 3.7427,
      "step": 186
    },
    {
      "epoch": 0.2992,
      "grad_norm": 1.2730371952056885,
      "learning_rate": 0.00021959999999999997,
      "loss": 3.8019,
      "step": 187
    },
    {
      "epoch": 0.3008,
      "grad_norm": 0.9145507216453552,
      "learning_rate": 0.00022079999999999997,
      "loss": 3.7265,
      "step": 188
    },
    {
      "epoch": 0.3024,
      "grad_norm": 1.7408198118209839,
      "learning_rate": 0.00022199999999999998,
      "loss": 3.6761,
      "step": 189
    },
    {
      "epoch": 0.304,
      "grad_norm": 4.375244140625,
      "learning_rate": 0.00022319999999999998,
      "loss": 3.7797,
      "step": 190
    },
    {
      "epoch": 0.3056,
      "grad_norm": 2.859325647354126,
      "learning_rate": 0.00022439999999999998,
      "loss": 3.7509,
      "step": 191
    },
    {
      "epoch": 0.3072,
      "grad_norm": 2.0287702083587646,
      "learning_rate": 0.00022559999999999998,
      "loss": 3.7396,
      "step": 192
    },
    {
      "epoch": 0.3088,
      "grad_norm": 0.7529137134552002,
      "learning_rate": 0.00022679999999999998,
      "loss": 3.7297,
      "step": 193
    },
    {
      "epoch": 0.3104,
      "grad_norm": 0.8018739223480225,
      "learning_rate": 0.00022799999999999999,
      "loss": 3.758,
      "step": 194
    },
    {
      "epoch": 0.312,
      "grad_norm": 1.8517733812332153,
      "learning_rate": 0.0002292,
      "loss": 3.7313,
      "step": 195
    },
    {
      "epoch": 0.3136,
      "grad_norm": 1.8108242750167847,
      "learning_rate": 0.0002304,
      "loss": 3.726,
      "step": 196
    },
    {
      "epoch": 0.3152,
      "grad_norm": 1.7063705921173096,
      "learning_rate": 0.0002316,
      "loss": 3.6929,
      "step": 197
    },
    {
      "epoch": 0.3168,
      "grad_norm": 1.4768927097320557,
      "learning_rate": 0.0002328,
      "loss": 3.7741,
      "step": 198
    },
    {
      "epoch": 0.3184,
      "grad_norm": 2.0207772254943848,
      "learning_rate": 0.000234,
      "loss": 3.8361,
      "step": 199
    },
    {
      "epoch": 0.32,
      "grad_norm": NaN,
      "learning_rate": 0.000234,
      "loss": 3.9766,
      "step": 200
    },
    {
      "epoch": 0.3216,
      "grad_norm": 28.53923797607422,
      "learning_rate": 0.0002352,
      "loss": 4.9473,
      "step": 201
    },
    {
      "epoch": 0.3232,
      "grad_norm": 19.729700088500977,
      "learning_rate": 0.0002364,
      "loss": 3.8604,
      "step": 202
    },
    {
      "epoch": 0.3248,
      "grad_norm": 15.63243293762207,
      "learning_rate": 0.0002376,
      "loss": 4.1119,
      "step": 203
    },
    {
      "epoch": 0.3264,
      "grad_norm": 4.515649795532227,
      "learning_rate": 0.0002388,
      "loss": 3.7548,
      "step": 204
    },
    {
      "epoch": 0.328,
      "grad_norm": 3.835597038269043,
      "learning_rate": 0.00023999999999999998,
      "loss": 3.7632,
      "step": 205
    },
    {
      "epoch": 0.3296,
      "grad_norm": 3.379566192626953,
      "learning_rate": 0.00024119999999999998,
      "loss": 3.9366,
      "step": 206
    },
    {
      "epoch": 0.3312,
      "grad_norm": 1.9478070735931396,
      "learning_rate": 0.00024239999999999998,
      "loss": 3.7263,
      "step": 207
    },
    {
      "epoch": 0.3328,
      "grad_norm": 2.06083083152771,
      "learning_rate": 0.00024359999999999999,
      "loss": 3.6968,
      "step": 208
    },
    {
      "epoch": 0.3344,
      "grad_norm": 3.923818826675415,
      "learning_rate": 0.0002448,
      "loss": 3.7776,
      "step": 209
    },
    {
      "epoch": 0.336,
      "grad_norm": 3.4094479084014893,
      "learning_rate": 0.00024599999999999996,
      "loss": 3.7088,
      "step": 210
    },
    {
      "epoch": 0.3376,
      "grad_norm": 4.099003314971924,
      "learning_rate": 0.0002472,
      "loss": 3.7462,
      "step": 211
    },
    {
      "epoch": 0.3392,
      "grad_norm": 0.9599695205688477,
      "learning_rate": 0.00024839999999999997,
      "loss": 3.7051,
      "step": 212
    },
    {
      "epoch": 0.3408,
      "grad_norm": 3.5445761680603027,
      "learning_rate": 0.00024959999999999994,
      "loss": 3.6189,
      "step": 213
    },
    {
      "epoch": 0.3424,
      "grad_norm": 3.2550857067108154,
      "learning_rate": 0.00025079999999999997,
      "loss": 3.5834,
      "step": 214
    },
    {
      "epoch": 0.344,
      "grad_norm": 1.9503517150878906,
      "learning_rate": 0.00025199999999999995,
      "loss": 3.5801,
      "step": 215
    },
    {
      "epoch": 0.3456,
      "grad_norm": 4.3856520652771,
      "learning_rate": 0.0002532,
      "loss": 3.7169,
      "step": 216
    },
    {
      "epoch": 0.3472,
      "grad_norm": 1.5111596584320068,
      "learning_rate": 0.00025439999999999995,
      "loss": 3.5968,
      "step": 217
    },
    {
      "epoch": 0.3488,
      "grad_norm": 1.61884605884552,
      "learning_rate": 0.0002556,
      "loss": 3.5628,
      "step": 218
    },
    {
      "epoch": 0.3504,
      "grad_norm": 0.7278621792793274,
      "learning_rate": 0.00025679999999999995,
      "loss": 3.6218,
      "step": 219
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.9155241847038269,
      "learning_rate": 0.000258,
      "loss": 3.4842,
      "step": 220
    },
    {
      "epoch": 0.3536,
      "grad_norm": 1.681706428527832,
      "learning_rate": 0.00025919999999999996,
      "loss": 3.4532,
      "step": 221
    },
    {
      "epoch": 0.3552,
      "grad_norm": 2.0195343494415283,
      "learning_rate": 0.0002604,
      "loss": 3.529,
      "step": 222
    },
    {
      "epoch": 0.3568,
      "grad_norm": 2.215996026992798,
      "learning_rate": 0.00026159999999999996,
      "loss": 3.5834,
      "step": 223
    },
    {
      "epoch": 0.3584,
      "grad_norm": 3.6210215091705322,
      "learning_rate": 0.0002628,
      "loss": 3.5076,
      "step": 224
    },
    {
      "epoch": 0.36,
      "grad_norm": 3.3881304264068604,
      "learning_rate": 0.00026399999999999997,
      "loss": 3.5052,
      "step": 225
    },
    {
      "epoch": 0.3616,
      "grad_norm": 1.2712512016296387,
      "learning_rate": 0.0002652,
      "loss": 3.4167,
      "step": 226
    },
    {
      "epoch": 0.3632,
      "grad_norm": 2.784555435180664,
      "learning_rate": 0.00026639999999999997,
      "loss": 3.4773,
      "step": 227
    },
    {
      "epoch": 0.3648,
      "grad_norm": 3.9173593521118164,
      "learning_rate": 0.0002676,
      "loss": 3.4792,
      "step": 228
    },
    {
      "epoch": 0.3664,
      "grad_norm": 5.97532844543457,
      "learning_rate": 0.0002688,
      "loss": 3.581,
      "step": 229
    },
    {
      "epoch": 0.368,
      "grad_norm": 1.1036829948425293,
      "learning_rate": 0.00027,
      "loss": 3.4026,
      "step": 230
    },
    {
      "epoch": 0.3696,
      "grad_norm": 2.0570623874664307,
      "learning_rate": 0.0002712,
      "loss": 3.3855,
      "step": 231
    },
    {
      "epoch": 0.3712,
      "grad_norm": 2.33469295501709,
      "learning_rate": 0.0002724,
      "loss": 3.3493,
      "step": 232
    },
    {
      "epoch": 0.3728,
      "grad_norm": 1.7420787811279297,
      "learning_rate": 0.0002736,
      "loss": 3.4412,
      "step": 233
    },
    {
      "epoch": 0.3744,
      "grad_norm": 1.597578525543213,
      "learning_rate": 0.0002748,
      "loss": 3.4772,
      "step": 234
    },
    {
      "epoch": 0.376,
      "grad_norm": 1.469741702079773,
      "learning_rate": 0.000276,
      "loss": 3.2345,
      "step": 235
    },
    {
      "epoch": 0.3776,
      "grad_norm": 1.8439031839370728,
      "learning_rate": 0.0002772,
      "loss": 3.2449,
      "step": 236
    },
    {
      "epoch": 0.3792,
      "grad_norm": 1.8763359785079956,
      "learning_rate": 0.0002784,
      "loss": 3.0865,
      "step": 237
    },
    {
      "epoch": 0.3808,
      "grad_norm": 1.3372673988342285,
      "learning_rate": 0.00027959999999999997,
      "loss": 3.2648,
      "step": 238
    },
    {
      "epoch": 0.3824,
      "grad_norm": 1.0363004207611084,
      "learning_rate": 0.0002808,
      "loss": 3.1095,
      "step": 239
    },
    {
      "epoch": 0.384,
      "grad_norm": 1.2411675453186035,
      "learning_rate": 0.00028199999999999997,
      "loss": 3.1516,
      "step": 240
    },
    {
      "epoch": 0.3856,
      "grad_norm": 1.1692326068878174,
      "learning_rate": 0.00028319999999999994,
      "loss": 3.1162,
      "step": 241
    },
    {
      "epoch": 0.3872,
      "grad_norm": 1.514488697052002,
      "learning_rate": 0.0002844,
      "loss": 3.1885,
      "step": 242
    },
    {
      "epoch": 0.3888,
      "grad_norm": 1.0548737049102783,
      "learning_rate": 0.00028559999999999995,
      "loss": 2.9995,
      "step": 243
    },
    {
      "epoch": 0.3904,
      "grad_norm": 1.435293436050415,
      "learning_rate": 0.0002868,
      "loss": 3.1228,
      "step": 244
    },
    {
      "epoch": 0.392,
      "grad_norm": 4.880733966827393,
      "learning_rate": 0.00028799999999999995,
      "loss": 3.523,
      "step": 245
    },
    {
      "epoch": 0.3936,
      "grad_norm": 2.134737730026245,
      "learning_rate": 0.0002892,
      "loss": 2.9433,
      "step": 246
    },
    {
      "epoch": 0.3952,
      "grad_norm": 1.4167667627334595,
      "learning_rate": 0.00029039999999999996,
      "loss": 3.0941,
      "step": 247
    },
    {
      "epoch": 0.3968,
      "grad_norm": 1.7885382175445557,
      "learning_rate": 0.0002916,
      "loss": 3.076,
      "step": 248
    },
    {
      "epoch": 0.3984,
      "grad_norm": 2.0048017501831055,
      "learning_rate": 0.00029279999999999996,
      "loss": 3.1318,
      "step": 249
    },
    {
      "epoch": 0.4,
      "grad_norm": 2.751702070236206,
      "learning_rate": 0.000294,
      "loss": 3.437,
      "step": 250
    },
    {
      "epoch": 0.4016,
      "grad_norm": 60.15131759643555,
      "learning_rate": 0.00029519999999999997,
      "loss": 6.7094,
      "step": 251
    },
    {
      "epoch": 0.4032,
      "grad_norm": 11.904341697692871,
      "learning_rate": 0.0002964,
      "loss": 3.3368,
      "step": 252
    },
    {
      "epoch": 0.4048,
      "grad_norm": 5.927310943603516,
      "learning_rate": 0.00029759999999999997,
      "loss": 3.084,
      "step": 253
    },
    {
      "epoch": 0.4064,
      "grad_norm": 1.259347915649414,
      "learning_rate": 0.0002988,
      "loss": 2.8328,
      "step": 254
    },
    {
      "epoch": 0.408,
      "grad_norm": 2.5598220825195312,
      "learning_rate": 0.0003,
      "loss": 2.9085,
      "step": 255
    },
    {
      "epoch": 0.4096,
      "grad_norm": 1.866328239440918,
      "learning_rate": 0.00030119999999999995,
      "loss": 2.8637,
      "step": 256
    },
    {
      "epoch": 0.4112,
      "grad_norm": 1.2088593244552612,
      "learning_rate": 0.0003024,
      "loss": 2.6899,
      "step": 257
    },
    {
      "epoch": 0.4128,
      "grad_norm": 4.605785369873047,
      "learning_rate": 0.00030359999999999995,
      "loss": 2.9239,
      "step": 258
    },
    {
      "epoch": 0.4144,
      "grad_norm": 1.7429507970809937,
      "learning_rate": 0.0003048,
      "loss": 2.5301,
      "step": 259
    },
    {
      "epoch": 0.416,
      "grad_norm": 2.9041364192962646,
      "learning_rate": 0.00030599999999999996,
      "loss": 2.4356,
      "step": 260
    },
    {
      "epoch": 0.4176,
      "grad_norm": 2.5931711196899414,
      "learning_rate": 0.0003072,
      "loss": 2.523,
      "step": 261
    },
    {
      "epoch": 0.4192,
      "grad_norm": 1.1678705215454102,
      "learning_rate": 0.00030839999999999996,
      "loss": 2.2468,
      "step": 262
    },
    {
      "epoch": 0.4208,
      "grad_norm": 1.711000680923462,
      "learning_rate": 0.0003096,
      "loss": 2.5393,
      "step": 263
    },
    {
      "epoch": 0.4224,
      "grad_norm": 1.8182960748672485,
      "learning_rate": 0.00031079999999999997,
      "loss": 2.4138,
      "step": 264
    },
    {
      "epoch": 0.424,
      "grad_norm": 1.3307945728302002,
      "learning_rate": 0.000312,
      "loss": 2.2623,
      "step": 265
    },
    {
      "epoch": 0.4256,
      "grad_norm": 1.9444172382354736,
      "learning_rate": 0.00031319999999999997,
      "loss": 2.1209,
      "step": 266
    },
    {
      "epoch": 0.4272,
      "grad_norm": 2.1593079566955566,
      "learning_rate": 0.0003144,
      "loss": 1.9432,
      "step": 267
    },
    {
      "epoch": 0.4288,
      "grad_norm": 2.1220879554748535,
      "learning_rate": 0.0003156,
      "loss": 2.1439,
      "step": 268
    },
    {
      "epoch": 0.4304,
      "grad_norm": 0.8869176506996155,
      "learning_rate": 0.0003168,
      "loss": 2.1583,
      "step": 269
    },
    {
      "epoch": 0.432,
      "grad_norm": 1.2422492504119873,
      "learning_rate": 0.000318,
      "loss": 1.8846,
      "step": 270
    },
    {
      "epoch": 0.4336,
      "grad_norm": 0.8885542154312134,
      "learning_rate": 0.0003192,
      "loss": 1.9293,
      "step": 271
    },
    {
      "epoch": 0.4352,
      "grad_norm": 1.3430697917938232,
      "learning_rate": 0.0003204,
      "loss": 2.1379,
      "step": 272
    },
    {
      "epoch": 0.4368,
      "grad_norm": 2.2559776306152344,
      "learning_rate": 0.0003216,
      "loss": 2.311,
      "step": 273
    },
    {
      "epoch": 0.4384,
      "grad_norm": 1.0061554908752441,
      "learning_rate": 0.0003228,
      "loss": 1.8522,
      "step": 274
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.9519514441490173,
      "learning_rate": 0.000324,
      "loss": 1.9418,
      "step": 275
    },
    {
      "epoch": 0.4416,
      "grad_norm": 1.034746527671814,
      "learning_rate": 0.0003252,
      "loss": 2.1713,
      "step": 276
    },
    {
      "epoch": 0.4432,
      "grad_norm": 1.4670822620391846,
      "learning_rate": 0.0003264,
      "loss": 2.2635,
      "step": 277
    },
    {
      "epoch": 0.4448,
      "grad_norm": 1.1949667930603027,
      "learning_rate": 0.0003276,
      "loss": 2.447,
      "step": 278
    },
    {
      "epoch": 0.4464,
      "grad_norm": 2.0934438705444336,
      "learning_rate": 0.0003288,
      "loss": 1.9578,
      "step": 279
    },
    {
      "epoch": 0.448,
      "grad_norm": 1.7934505939483643,
      "learning_rate": 0.00033,
      "loss": 2.3822,
      "step": 280
    },
    {
      "epoch": 0.4496,
      "grad_norm": 1.0247673988342285,
      "learning_rate": 0.0003312,
      "loss": 1.8567,
      "step": 281
    },
    {
      "epoch": 0.4512,
      "grad_norm": 2.136781692504883,
      "learning_rate": 0.0003324,
      "loss": 1.964,
      "step": 282
    },
    {
      "epoch": 0.4528,
      "grad_norm": 0.8177749514579773,
      "learning_rate": 0.0003336,
      "loss": 1.8132,
      "step": 283
    },
    {
      "epoch": 0.4544,
      "grad_norm": 1.0221589803695679,
      "learning_rate": 0.0003348,
      "loss": 1.8668,
      "step": 284
    },
    {
      "epoch": 0.456,
      "grad_norm": 1.517808198928833,
      "learning_rate": 0.000336,
      "loss": 2.1496,
      "step": 285
    },
    {
      "epoch": 0.4576,
      "grad_norm": 1.5944926738739014,
      "learning_rate": 0.0003372,
      "loss": 2.3953,
      "step": 286
    },
    {
      "epoch": 0.4592,
      "grad_norm": 2.405046224594116,
      "learning_rate": 0.00033839999999999993,
      "loss": 1.9663,
      "step": 287
    },
    {
      "epoch": 0.4608,
      "grad_norm": 1.6332200765609741,
      "learning_rate": 0.00033959999999999996,
      "loss": 1.7194,
      "step": 288
    },
    {
      "epoch": 0.4624,
      "grad_norm": 1.463877558708191,
      "learning_rate": 0.00034079999999999994,
      "loss": 1.9922,
      "step": 289
    },
    {
      "epoch": 0.464,
      "grad_norm": 1.3162888288497925,
      "learning_rate": 0.00034199999999999996,
      "loss": 1.9758,
      "step": 290
    },
    {
      "epoch": 0.4656,
      "grad_norm": 1.7767980098724365,
      "learning_rate": 0.00034319999999999994,
      "loss": 2.2791,
      "step": 291
    },
    {
      "epoch": 0.4672,
      "grad_norm": 1.7777577638626099,
      "learning_rate": 0.00034439999999999997,
      "loss": 1.5986,
      "step": 292
    },
    {
      "epoch": 0.4688,
      "grad_norm": 1.4968204498291016,
      "learning_rate": 0.00034559999999999994,
      "loss": 1.7837,
      "step": 293
    },
    {
      "epoch": 0.4704,
      "grad_norm": 1.639837384223938,
      "learning_rate": 0.0003467999999999999,
      "loss": 2.1542,
      "step": 294
    },
    {
      "epoch": 0.472,
      "grad_norm": 2.0573577880859375,
      "learning_rate": 0.00034799999999999995,
      "loss": 2.1574,
      "step": 295
    },
    {
      "epoch": 0.4736,
      "grad_norm": 2.6885788440704346,
      "learning_rate": 0.0003491999999999999,
      "loss": 2.0754,
      "step": 296
    },
    {
      "epoch": 0.4752,
      "grad_norm": 2.3689215183258057,
      "learning_rate": 0.00035039999999999995,
      "loss": 2.3191,
      "step": 297
    },
    {
      "epoch": 0.4768,
      "grad_norm": 1.4669629335403442,
      "learning_rate": 0.0003515999999999999,
      "loss": 2.1903,
      "step": 298
    },
    {
      "epoch": 0.4784,
      "grad_norm": 6.631944179534912,
      "learning_rate": 0.00035279999999999996,
      "loss": 2.4098,
      "step": 299
    },
    {
      "epoch": 0.48,
      "grad_norm": 3.869600296020508,
      "learning_rate": 0.00035399999999999993,
      "loss": 2.8155,
      "step": 300
    },
    {
      "epoch": 0.4816,
      "grad_norm": 18.670400619506836,
      "learning_rate": 0.00035519999999999996,
      "loss": 4.0821,
      "step": 301
    },
    {
      "epoch": 0.4832,
      "grad_norm": 6.5373640060424805,
      "learning_rate": 0.00035639999999999994,
      "loss": 2.6903,
      "step": 302
    },
    {
      "epoch": 0.4848,
      "grad_norm": 1.8103080987930298,
      "learning_rate": 0.00035759999999999996,
      "loss": 1.715,
      "step": 303
    },
    {
      "epoch": 0.4864,
      "grad_norm": 2.0340678691864014,
      "learning_rate": 0.00035879999999999994,
      "loss": 2.2299,
      "step": 304
    },
    {
      "epoch": 0.488,
      "grad_norm": 2.8201074600219727,
      "learning_rate": 0.00035999999999999997,
      "loss": 2.0269,
      "step": 305
    },
    {
      "epoch": 0.4896,
      "grad_norm": 3.9219393730163574,
      "learning_rate": 0.00036119999999999994,
      "loss": 2.0192,
      "step": 306
    },
    {
      "epoch": 0.4912,
      "grad_norm": 2.1272470951080322,
      "learning_rate": 0.00036239999999999997,
      "loss": 2.2065,
      "step": 307
    },
    {
      "epoch": 0.4928,
      "grad_norm": 1.2031842470169067,
      "learning_rate": 0.00036359999999999995,
      "loss": 1.6669,
      "step": 308
    },
    {
      "epoch": 0.4944,
      "grad_norm": 2.6753766536712646,
      "learning_rate": 0.0003648,
      "loss": 1.9611,
      "step": 309
    },
    {
      "epoch": 0.496,
      "grad_norm": 3.980130672454834,
      "learning_rate": 0.00036599999999999995,
      "loss": 1.8961,
      "step": 310
    },
    {
      "epoch": 0.4976,
      "grad_norm": 3.4683375358581543,
      "learning_rate": 0.0003672,
      "loss": 1.6241,
      "step": 311
    },
    {
      "epoch": 0.4992,
      "grad_norm": 3.478597402572632,
      "learning_rate": 0.00036839999999999996,
      "loss": 1.8472,
      "step": 312
    },
    {
      "epoch": 0.5008,
      "grad_norm": 1.360845923423767,
      "learning_rate": 0.0003696,
      "loss": 1.7252,
      "step": 313
    },
    {
      "epoch": 0.5024,
      "grad_norm": 0.9226222634315491,
      "learning_rate": 0.00037079999999999996,
      "loss": 1.4792,
      "step": 314
    },
    {
      "epoch": 0.504,
      "grad_norm": 1.2864586114883423,
      "learning_rate": 0.000372,
      "loss": 1.5637,
      "step": 315
    },
    {
      "epoch": 0.5056,
      "grad_norm": 1.395561695098877,
      "learning_rate": 0.00037319999999999996,
      "loss": 1.41,
      "step": 316
    },
    {
      "epoch": 0.5072,
      "grad_norm": 1.9199680089950562,
      "learning_rate": 0.0003744,
      "loss": 1.8221,
      "step": 317
    },
    {
      "epoch": 0.5088,
      "grad_norm": 1.2985516786575317,
      "learning_rate": 0.00037559999999999997,
      "loss": 1.9545,
      "step": 318
    },
    {
      "epoch": 0.5104,
      "grad_norm": 0.861045777797699,
      "learning_rate": 0.00037679999999999994,
      "loss": 1.2,
      "step": 319
    },
    {
      "epoch": 0.512,
      "grad_norm": 1.4724054336547852,
      "learning_rate": 0.00037799999999999997,
      "loss": 1.9655,
      "step": 320
    },
    {
      "epoch": 0.5136,
      "grad_norm": 1.121520757675171,
      "learning_rate": 0.00037919999999999995,
      "loss": 1.6427,
      "step": 321
    },
    {
      "epoch": 0.5152,
      "grad_norm": 0.8945459127426147,
      "learning_rate": 0.0003804,
      "loss": 1.8193,
      "step": 322
    },
    {
      "epoch": 0.5168,
      "grad_norm": 1.510231614112854,
      "learning_rate": 0.00038159999999999995,
      "loss": 1.7606,
      "step": 323
    },
    {
      "epoch": 0.5184,
      "grad_norm": 1.2624521255493164,
      "learning_rate": 0.0003828,
      "loss": 1.8403,
      "step": 324
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.1738777160644531,
      "learning_rate": 0.00038399999999999996,
      "loss": 1.5676,
      "step": 325
    },
    {
      "epoch": 0.5216,
      "grad_norm": 1.4238841533660889,
      "learning_rate": 0.0003852,
      "loss": 1.3816,
      "step": 326
    },
    {
      "epoch": 0.5232,
      "grad_norm": 0.9597876667976379,
      "learning_rate": 0.00038639999999999996,
      "loss": 1.4629,
      "step": 327
    },
    {
      "epoch": 0.5248,
      "grad_norm": 1.1177942752838135,
      "learning_rate": 0.0003876,
      "loss": 1.4452,
      "step": 328
    },
    {
      "epoch": 0.5264,
      "grad_norm": 1.0831985473632812,
      "learning_rate": 0.00038879999999999996,
      "loss": 1.6668,
      "step": 329
    },
    {
      "epoch": 0.528,
      "grad_norm": 1.1985876560211182,
      "learning_rate": 0.00039,
      "loss": 1.5963,
      "step": 330
    },
    {
      "epoch": 0.5296,
      "grad_norm": 1.4813580513000488,
      "learning_rate": 0.00039119999999999997,
      "loss": 1.6768,
      "step": 331
    },
    {
      "epoch": 0.5312,
      "grad_norm": 1.4565857648849487,
      "learning_rate": 0.0003924,
      "loss": 1.6633,
      "step": 332
    },
    {
      "epoch": 0.5328,
      "grad_norm": 1.1487796306610107,
      "learning_rate": 0.00039359999999999997,
      "loss": 1.4567,
      "step": 333
    },
    {
      "epoch": 0.5344,
      "grad_norm": 1.2718472480773926,
      "learning_rate": 0.0003948,
      "loss": 1.8923,
      "step": 334
    },
    {
      "epoch": 0.536,
      "grad_norm": 1.393419623374939,
      "learning_rate": 0.000396,
      "loss": 1.1306,
      "step": 335
    },
    {
      "epoch": 0.5376,
      "grad_norm": 2.660311222076416,
      "learning_rate": 0.0003972,
      "loss": 1.6447,
      "step": 336
    },
    {
      "epoch": 0.5392,
      "grad_norm": 1.1137669086456299,
      "learning_rate": 0.0003984,
      "loss": 1.6307,
      "step": 337
    },
    {
      "epoch": 0.5408,
      "grad_norm": 1.1761451959609985,
      "learning_rate": 0.0003996,
      "loss": 1.6008,
      "step": 338
    },
    {
      "epoch": 0.5424,
      "grad_norm": 1.350234866142273,
      "learning_rate": 0.0004008,
      "loss": 1.523,
      "step": 339
    },
    {
      "epoch": 0.544,
      "grad_norm": 1.166507601737976,
      "learning_rate": 0.000402,
      "loss": 1.2526,
      "step": 340
    },
    {
      "epoch": 0.5456,
      "grad_norm": 1.3881250619888306,
      "learning_rate": 0.0004032,
      "loss": 1.7568,
      "step": 341
    },
    {
      "epoch": 0.5472,
      "grad_norm": 3.9268593788146973,
      "learning_rate": 0.0004044,
      "loss": 2.0863,
      "step": 342
    },
    {
      "epoch": 0.5488,
      "grad_norm": 1.7067468166351318,
      "learning_rate": 0.0004056,
      "loss": 1.5831,
      "step": 343
    },
    {
      "epoch": 0.5504,
      "grad_norm": 1.4713155031204224,
      "learning_rate": 0.00040679999999999997,
      "loss": 1.7429,
      "step": 344
    },
    {
      "epoch": 0.552,
      "grad_norm": 1.212177038192749,
      "learning_rate": 0.000408,
      "loss": 1.6922,
      "step": 345
    },
    {
      "epoch": 0.5536,
      "grad_norm": 1.6186903715133667,
      "learning_rate": 0.00040919999999999997,
      "loss": 1.818,
      "step": 346
    },
    {
      "epoch": 0.5552,
      "grad_norm": 2.359452247619629,
      "learning_rate": 0.0004104,
      "loss": 2.4012,
      "step": 347
    },
    {
      "epoch": 0.5568,
      "grad_norm": 2.3700032234191895,
      "learning_rate": 0.0004116,
      "loss": 2.296,
      "step": 348
    },
    {
      "epoch": 0.5584,
      "grad_norm": 1.8284653425216675,
      "learning_rate": 0.00041279999999999995,
      "loss": 2.0039,
      "step": 349
    },
    {
      "epoch": 0.56,
      "grad_norm": 2.491885185241699,
      "learning_rate": 0.0004139999999999999,
      "loss": 2.089,
      "step": 350
    },
    {
      "epoch": 0.5616,
      "grad_norm": 2.3013062477111816,
      "learning_rate": 0.00041519999999999995,
      "loss": 1.6711,
      "step": 351
    },
    {
      "epoch": 0.5632,
      "grad_norm": 1.9905917644500732,
      "learning_rate": 0.00041639999999999993,
      "loss": 1.6065,
      "step": 352
    },
    {
      "epoch": 0.5648,
      "grad_norm": 1.480592966079712,
      "learning_rate": 0.00041759999999999996,
      "loss": 2.2154,
      "step": 353
    },
    {
      "epoch": 0.5664,
      "grad_norm": 2.0091054439544678,
      "learning_rate": 0.00041879999999999993,
      "loss": 1.9469,
      "step": 354
    },
    {
      "epoch": 0.568,
      "grad_norm": 4.029331207275391,
      "learning_rate": 0.00041999999999999996,
      "loss": 2.159,
      "step": 355
    },
    {
      "epoch": 0.5696,
      "grad_norm": 3.835162401199341,
      "learning_rate": 0.00042119999999999994,
      "loss": 1.9669,
      "step": 356
    },
    {
      "epoch": 0.5712,
      "grad_norm": 3.283651828765869,
      "learning_rate": 0.0004223999999999999,
      "loss": 1.9169,
      "step": 357
    },
    {
      "epoch": 0.5728,
      "grad_norm": 2.7341041564941406,
      "learning_rate": 0.00042359999999999994,
      "loss": 1.9569,
      "step": 358
    },
    {
      "epoch": 0.5744,
      "grad_norm": 3.72660231590271,
      "learning_rate": 0.0004247999999999999,
      "loss": 1.9345,
      "step": 359
    },
    {
      "epoch": 0.576,
      "grad_norm": 1.2118239402770996,
      "learning_rate": 0.00042599999999999995,
      "loss": 1.6209,
      "step": 360
    },
    {
      "epoch": 0.5776,
      "grad_norm": 2.773261547088623,
      "learning_rate": 0.0004271999999999999,
      "loss": 1.7308,
      "step": 361
    },
    {
      "epoch": 0.5792,
      "grad_norm": 1.5158611536026,
      "learning_rate": 0.00042839999999999995,
      "loss": 1.5566,
      "step": 362
    },
    {
      "epoch": 0.5808,
      "grad_norm": 1.7444158792495728,
      "learning_rate": 0.0004295999999999999,
      "loss": 1.517,
      "step": 363
    },
    {
      "epoch": 0.5824,
      "grad_norm": 2.0275840759277344,
      "learning_rate": 0.00043079999999999995,
      "loss": 1.3961,
      "step": 364
    },
    {
      "epoch": 0.584,
      "grad_norm": 1.0864077806472778,
      "learning_rate": 0.00043199999999999993,
      "loss": 1.3833,
      "step": 365
    },
    {
      "epoch": 0.5856,
      "grad_norm": 1.5802364349365234,
      "learning_rate": 0.00043319999999999996,
      "loss": 1.2144,
      "step": 366
    },
    {
      "epoch": 0.5872,
      "grad_norm": 1.1365376710891724,
      "learning_rate": 0.00043439999999999993,
      "loss": 1.4435,
      "step": 367
    },
    {
      "epoch": 0.5888,
      "grad_norm": 2.001816511154175,
      "learning_rate": 0.00043559999999999996,
      "loss": 1.8125,
      "step": 368
    },
    {
      "epoch": 0.5904,
      "grad_norm": 3.9782650470733643,
      "learning_rate": 0.00043679999999999994,
      "loss": 2.3917,
      "step": 369
    },
    {
      "epoch": 0.592,
      "grad_norm": 2.559185743331909,
      "learning_rate": 0.00043799999999999997,
      "loss": 1.4065,
      "step": 370
    },
    {
      "epoch": 0.5936,
      "grad_norm": 1.531436562538147,
      "learning_rate": 0.00043919999999999994,
      "loss": 1.3286,
      "step": 371
    },
    {
      "epoch": 0.5952,
      "grad_norm": 2.1857967376708984,
      "learning_rate": 0.00044039999999999997,
      "loss": 1.3389,
      "step": 372
    },
    {
      "epoch": 0.5968,
      "grad_norm": 1.2437169551849365,
      "learning_rate": 0.00044159999999999995,
      "loss": 1.305,
      "step": 373
    },
    {
      "epoch": 0.5984,
      "grad_norm": 1.8238170146942139,
      "learning_rate": 0.0004428,
      "loss": 1.6173,
      "step": 374
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.9433510303497314,
      "learning_rate": 0.00044399999999999995,
      "loss": 1.2818,
      "step": 375
    },
    {
      "epoch": 0.6016,
      "grad_norm": 1.7307677268981934,
      "learning_rate": 0.0004452,
      "loss": 1.3214,
      "step": 376
    },
    {
      "epoch": 0.6032,
      "grad_norm": 1.2876639366149902,
      "learning_rate": 0.00044639999999999995,
      "loss": 1.6527,
      "step": 377
    },
    {
      "epoch": 0.6048,
      "grad_norm": 1.0600067377090454,
      "learning_rate": 0.0004476,
      "loss": 1.4502,
      "step": 378
    },
    {
      "epoch": 0.6064,
      "grad_norm": 1.1501784324645996,
      "learning_rate": 0.00044879999999999996,
      "loss": 1.2737,
      "step": 379
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.959336519241333,
      "learning_rate": 0.00045,
      "loss": 1.7977,
      "step": 380
    },
    {
      "epoch": 0.6096,
      "grad_norm": 1.1226460933685303,
      "learning_rate": 0.00045119999999999996,
      "loss": 1.4758,
      "step": 381
    },
    {
      "epoch": 0.6112,
      "grad_norm": 0.9224548935890198,
      "learning_rate": 0.00045239999999999994,
      "loss": 1.5619,
      "step": 382
    },
    {
      "epoch": 0.6128,
      "grad_norm": 0.814278244972229,
      "learning_rate": 0.00045359999999999997,
      "loss": 1.3943,
      "step": 383
    },
    {
      "epoch": 0.6144,
      "grad_norm": 0.9084352850914001,
      "learning_rate": 0.00045479999999999994,
      "loss": 1.5962,
      "step": 384
    },
    {
      "epoch": 0.616,
      "grad_norm": 1.4289610385894775,
      "learning_rate": 0.00045599999999999997,
      "loss": 1.3404,
      "step": 385
    },
    {
      "epoch": 0.6176,
      "grad_norm": 1.2523407936096191,
      "learning_rate": 0.00045719999999999995,
      "loss": 1.4539,
      "step": 386
    },
    {
      "epoch": 0.6192,
      "grad_norm": 1.1009465456008911,
      "learning_rate": 0.0004584,
      "loss": 1.2913,
      "step": 387
    },
    {
      "epoch": 0.6208,
      "grad_norm": 1.369174838066101,
      "learning_rate": 0.00045959999999999995,
      "loss": 1.7786,
      "step": 388
    },
    {
      "epoch": 0.6224,
      "grad_norm": 1.6469858884811401,
      "learning_rate": 0.0004608,
      "loss": 1.5913,
      "step": 389
    },
    {
      "epoch": 0.624,
      "grad_norm": 1.0242924690246582,
      "learning_rate": 0.00046199999999999995,
      "loss": 1.4396,
      "step": 390
    },
    {
      "epoch": 0.6256,
      "grad_norm": 1.0944545269012451,
      "learning_rate": 0.0004632,
      "loss": 1.3629,
      "step": 391
    },
    {
      "epoch": 0.6272,
      "grad_norm": 1.2820014953613281,
      "learning_rate": 0.00046439999999999996,
      "loss": 1.4899,
      "step": 392
    },
    {
      "epoch": 0.6288,
      "grad_norm": 1.1084744930267334,
      "learning_rate": 0.0004656,
      "loss": 1.3476,
      "step": 393
    },
    {
      "epoch": 0.6304,
      "grad_norm": 1.2067919969558716,
      "learning_rate": 0.00046679999999999996,
      "loss": 1.6783,
      "step": 394
    },
    {
      "epoch": 0.632,
      "grad_norm": 1.3350406885147095,
      "learning_rate": 0.000468,
      "loss": 1.8544,
      "step": 395
    },
    {
      "epoch": 0.6336,
      "grad_norm": 2.817788600921631,
      "learning_rate": 0.00046919999999999997,
      "loss": 1.6983,
      "step": 396
    },
    {
      "epoch": 0.6352,
      "grad_norm": 1.577996850013733,
      "learning_rate": 0.0004704,
      "loss": 1.9165,
      "step": 397
    },
    {
      "epoch": 0.6368,
      "grad_norm": 1.992092251777649,
      "learning_rate": 0.00047159999999999997,
      "loss": 2.2309,
      "step": 398
    },
    {
      "epoch": 0.6384,
      "grad_norm": 1.5959856510162354,
      "learning_rate": 0.0004728,
      "loss": 1.977,
      "step": 399
    },
    {
      "epoch": 0.64,
      "grad_norm": 2.959681510925293,
      "learning_rate": 0.000474,
      "loss": 1.9346,
      "step": 400
    },
    {
      "epoch": 0.6416,
      "grad_norm": 3.0177066326141357,
      "learning_rate": 0.0004752,
      "loss": 1.9946,
      "step": 401
    },
    {
      "epoch": 0.6432,
      "grad_norm": 3.2861762046813965,
      "learning_rate": 0.0004764,
      "loss": 1.861,
      "step": 402
    },
    {
      "epoch": 0.6448,
      "grad_norm": 33.552921295166016,
      "learning_rate": 0.0004776,
      "loss": 7.6541,
      "step": 403
    },
    {
      "epoch": 0.6464,
      "grad_norm": 1.975786566734314,
      "learning_rate": 0.0004788,
      "loss": 1.7211,
      "step": 404
    },
    {
      "epoch": 0.648,
      "grad_norm": 3.4860012531280518,
      "learning_rate": 0.00047999999999999996,
      "loss": 1.8209,
      "step": 405
    },
    {
      "epoch": 0.6496,
      "grad_norm": 3.5102968215942383,
      "learning_rate": 0.0004812,
      "loss": 1.7063,
      "step": 406
    },
    {
      "epoch": 0.6512,
      "grad_norm": 4.879241943359375,
      "learning_rate": 0.00048239999999999996,
      "loss": 1.4615,
      "step": 407
    },
    {
      "epoch": 0.6528,
      "grad_norm": 2.7851948738098145,
      "learning_rate": 0.0004836,
      "loss": 1.6937,
      "step": 408
    },
    {
      "epoch": 0.6544,
      "grad_norm": 0.9268562197685242,
      "learning_rate": 0.00048479999999999997,
      "loss": 1.6795,
      "step": 409
    },
    {
      "epoch": 0.656,
      "grad_norm": 1.127549409866333,
      "learning_rate": 0.000486,
      "loss": 1.433,
      "step": 410
    },
    {
      "epoch": 0.6576,
      "grad_norm": 2.1968774795532227,
      "learning_rate": 0.00048719999999999997,
      "loss": 1.501,
      "step": 411
    },
    {
      "epoch": 0.6592,
      "grad_norm": 3.553457260131836,
      "learning_rate": 0.0004883999999999999,
      "loss": 1.4432,
      "step": 412
    },
    {
      "epoch": 0.6608,
      "grad_norm": 1.8429263830184937,
      "learning_rate": 0.0004896,
      "loss": 1.3622,
      "step": 413
    },
    {
      "epoch": 0.6624,
      "grad_norm": 0.8631522059440613,
      "learning_rate": 0.0004907999999999999,
      "loss": 1.157,
      "step": 414
    },
    {
      "epoch": 0.664,
      "grad_norm": 0.8570797443389893,
      "learning_rate": 0.0004919999999999999,
      "loss": 1.03,
      "step": 415
    },
    {
      "epoch": 0.6656,
      "grad_norm": 1.2855173349380493,
      "learning_rate": 0.0004932,
      "loss": 1.5622,
      "step": 416
    },
    {
      "epoch": 0.6672,
      "grad_norm": 1.406083106994629,
      "learning_rate": 0.0004944,
      "loss": 1.0664,
      "step": 417
    },
    {
      "epoch": 0.6688,
      "grad_norm": 1.5346801280975342,
      "learning_rate": 0.0004955999999999999,
      "loss": 1.5327,
      "step": 418
    },
    {
      "epoch": 0.6704,
      "grad_norm": 1.3836517333984375,
      "learning_rate": 0.0004967999999999999,
      "loss": 1.6197,
      "step": 419
    },
    {
      "epoch": 0.672,
      "grad_norm": 1.7024117708206177,
      "learning_rate": 0.000498,
      "loss": 1.2376,
      "step": 420
    },
    {
      "epoch": 0.6736,
      "grad_norm": 1.5172358751296997,
      "learning_rate": 0.0004991999999999999,
      "loss": 1.5432,
      "step": 421
    },
    {
      "epoch": 0.6752,
      "grad_norm": 1.142734408378601,
      "learning_rate": 0.0005003999999999999,
      "loss": 1.2484,
      "step": 422
    },
    {
      "epoch": 0.6768,
      "grad_norm": 0.8977586030960083,
      "learning_rate": 0.0005015999999999999,
      "loss": 1.3032,
      "step": 423
    },
    {
      "epoch": 0.6784,
      "grad_norm": 1.1880444288253784,
      "learning_rate": 0.0005028,
      "loss": 1.4404,
      "step": 424
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.214245080947876,
      "learning_rate": 0.0005039999999999999,
      "loss": 1.2976,
      "step": 425
    },
    {
      "epoch": 0.6816,
      "grad_norm": 0.9443445801734924,
      "learning_rate": 0.0005051999999999999,
      "loss": 1.418,
      "step": 426
    },
    {
      "epoch": 0.6832,
      "grad_norm": 1.4814517498016357,
      "learning_rate": 0.0005064,
      "loss": 1.3793,
      "step": 427
    },
    {
      "epoch": 0.6848,
      "grad_norm": 1.3838948011398315,
      "learning_rate": 0.0005076,
      "loss": 1.1823,
      "step": 428
    },
    {
      "epoch": 0.6864,
      "grad_norm": 0.9880338311195374,
      "learning_rate": 0.0005087999999999999,
      "loss": 1.108,
      "step": 429
    },
    {
      "epoch": 0.688,
      "grad_norm": 1.0871669054031372,
      "learning_rate": 0.0005099999999999999,
      "loss": 1.3633,
      "step": 430
    },
    {
      "epoch": 0.6896,
      "grad_norm": 1.2696417570114136,
      "learning_rate": 0.0005112,
      "loss": 1.3324,
      "step": 431
    },
    {
      "epoch": 0.6912,
      "grad_norm": 1.4589694738388062,
      "learning_rate": 0.0005124,
      "loss": 0.9689,
      "step": 432
    },
    {
      "epoch": 0.6928,
      "grad_norm": 0.948417603969574,
      "learning_rate": 0.0005135999999999999,
      "loss": 1.2153,
      "step": 433
    },
    {
      "epoch": 0.6944,
      "grad_norm": 1.4363794326782227,
      "learning_rate": 0.0005147999999999999,
      "loss": 1.7676,
      "step": 434
    },
    {
      "epoch": 0.696,
      "grad_norm": 0.8274084329605103,
      "learning_rate": 0.000516,
      "loss": 1.247,
      "step": 435
    },
    {
      "epoch": 0.6976,
      "grad_norm": 1.2022773027420044,
      "learning_rate": 0.0005172,
      "loss": 1.9134,
      "step": 436
    },
    {
      "epoch": 0.6992,
      "grad_norm": 1.0256644487380981,
      "learning_rate": 0.0005183999999999999,
      "loss": 1.5249,
      "step": 437
    },
    {
      "epoch": 0.7008,
      "grad_norm": 1.2487961053848267,
      "learning_rate": 0.0005195999999999999,
      "loss": 1.4572,
      "step": 438
    },
    {
      "epoch": 0.7024,
      "grad_norm": 1.103989601135254,
      "learning_rate": 0.0005208,
      "loss": 1.2583,
      "step": 439
    },
    {
      "epoch": 0.704,
      "grad_norm": 1.1899611949920654,
      "learning_rate": 0.000522,
      "loss": 1.249,
      "step": 440
    },
    {
      "epoch": 0.7056,
      "grad_norm": 1.1985859870910645,
      "learning_rate": 0.0005231999999999999,
      "loss": 1.3384,
      "step": 441
    },
    {
      "epoch": 0.7072,
      "grad_norm": 1.278523325920105,
      "learning_rate": 0.0005244,
      "loss": 1.6037,
      "step": 442
    },
    {
      "epoch": 0.7088,
      "grad_norm": 2.5164453983306885,
      "learning_rate": 0.0005256,
      "loss": 1.7037,
      "step": 443
    },
    {
      "epoch": 0.7104,
      "grad_norm": 1.4125896692276,
      "learning_rate": 0.0005267999999999999,
      "loss": 1.3628,
      "step": 444
    },
    {
      "epoch": 0.712,
      "grad_norm": 1.4269789457321167,
      "learning_rate": 0.0005279999999999999,
      "loss": 1.8712,
      "step": 445
    },
    {
      "epoch": 0.7136,
      "grad_norm": 1.321560263633728,
      "learning_rate": 0.0005292,
      "loss": 1.4942,
      "step": 446
    },
    {
      "epoch": 0.7152,
      "grad_norm": 1.5753790140151978,
      "learning_rate": 0.0005304,
      "loss": 1.9359,
      "step": 447
    },
    {
      "epoch": 0.7168,
      "grad_norm": 2.2859463691711426,
      "learning_rate": 0.0005315999999999999,
      "loss": 1.745,
      "step": 448
    },
    {
      "epoch": 0.7184,
      "grad_norm": 2.0483174324035645,
      "learning_rate": 0.0005327999999999999,
      "loss": 1.8343,
      "step": 449
    },
    {
      "epoch": 0.72,
      "grad_norm": 3.6337218284606934,
      "learning_rate": 0.000534,
      "loss": 2.2459,
      "step": 450
    },
    {
      "epoch": 0.7216,
      "grad_norm": 13.555660247802734,
      "learning_rate": 0.0005352,
      "loss": 3.3534,
      "step": 451
    },
    {
      "epoch": 0.7232,
      "grad_norm": 6.6364850997924805,
      "learning_rate": 0.0005363999999999999,
      "loss": 2.7515,
      "step": 452
    },
    {
      "epoch": 0.7248,
      "grad_norm": 3.817852258682251,
      "learning_rate": 0.0005376,
      "loss": 2.012,
      "step": 453
    },
    {
      "epoch": 0.7264,
      "grad_norm": 1.707593560218811,
      "learning_rate": 0.0005388,
      "loss": 1.9942,
      "step": 454
    },
    {
      "epoch": 0.728,
      "grad_norm": 2.77917218208313,
      "learning_rate": 0.00054,
      "loss": 1.6863,
      "step": 455
    },
    {
      "epoch": 0.7296,
      "grad_norm": 2.7656164169311523,
      "learning_rate": 0.0005411999999999999,
      "loss": 1.4779,
      "step": 456
    },
    {
      "epoch": 0.7312,
      "grad_norm": 1.720285177230835,
      "learning_rate": 0.0005424,
      "loss": 1.8024,
      "step": 457
    },
    {
      "epoch": 0.7328,
      "grad_norm": 3.847505807876587,
      "learning_rate": 0.0005436,
      "loss": 1.6034,
      "step": 458
    },
    {
      "epoch": 0.7344,
      "grad_norm": 2.7850637435913086,
      "learning_rate": 0.0005448,
      "loss": 1.3834,
      "step": 459
    },
    {
      "epoch": 0.736,
      "grad_norm": 1.2482507228851318,
      "learning_rate": 0.0005459999999999999,
      "loss": 1.6495,
      "step": 460
    },
    {
      "epoch": 0.7376,
      "grad_norm": 1.2714293003082275,
      "learning_rate": 0.0005472,
      "loss": 1.3752,
      "step": 461
    },
    {
      "epoch": 0.7392,
      "grad_norm": 1.8939746618270874,
      "learning_rate": 0.0005484,
      "loss": 1.4186,
      "step": 462
    },
    {
      "epoch": 0.7408,
      "grad_norm": 2.5669922828674316,
      "learning_rate": 0.0005496,
      "loss": 1.2681,
      "step": 463
    },
    {
      "epoch": 0.7424,
      "grad_norm": 1.9449177980422974,
      "learning_rate": 0.0005507999999999999,
      "loss": 1.3308,
      "step": 464
    },
    {
      "epoch": 0.744,
      "grad_norm": 0.872009813785553,
      "learning_rate": 0.000552,
      "loss": 1.2645,
      "step": 465
    },
    {
      "epoch": 0.7456,
      "grad_norm": 0.8539568185806274,
      "learning_rate": 0.0005532,
      "loss": 1.0258,
      "step": 466
    },
    {
      "epoch": 0.7472,
      "grad_norm": 1.0018901824951172,
      "learning_rate": 0.0005544,
      "loss": 1.38,
      "step": 467
    },
    {
      "epoch": 0.7488,
      "grad_norm": 0.9926770329475403,
      "learning_rate": 0.0005556,
      "loss": 1.3214,
      "step": 468
    },
    {
      "epoch": 0.7504,
      "grad_norm": 1.80833101272583,
      "learning_rate": 0.0005568,
      "loss": 1.3013,
      "step": 469
    },
    {
      "epoch": 0.752,
      "grad_norm": 1.4460201263427734,
      "learning_rate": 0.000558,
      "loss": 1.3652,
      "step": 470
    },
    {
      "epoch": 0.7536,
      "grad_norm": 1.0537903308868408,
      "learning_rate": 0.0005591999999999999,
      "loss": 1.1968,
      "step": 471
    },
    {
      "epoch": 0.7552,
      "grad_norm": 1.0886517763137817,
      "learning_rate": 0.0005604,
      "loss": 1.2747,
      "step": 472
    },
    {
      "epoch": 0.7568,
      "grad_norm": 1.2867629528045654,
      "learning_rate": 0.0005616,
      "loss": 1.2078,
      "step": 473
    },
    {
      "epoch": 0.7584,
      "grad_norm": 1.7365913391113281,
      "learning_rate": 0.0005627999999999999,
      "loss": 1.0753,
      "step": 474
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.7022826671600342,
      "learning_rate": 0.0005639999999999999,
      "loss": 1.2664,
      "step": 475
    },
    {
      "epoch": 0.7616,
      "grad_norm": 1.3423399925231934,
      "learning_rate": 0.0005652,
      "loss": 1.3553,
      "step": 476
    },
    {
      "epoch": 0.7632,
      "grad_norm": 1.3367669582366943,
      "learning_rate": 0.0005663999999999999,
      "loss": 1.4162,
      "step": 477
    },
    {
      "epoch": 0.7648,
      "grad_norm": 0.9925369024276733,
      "learning_rate": 0.0005675999999999999,
      "loss": 1.2366,
      "step": 478
    },
    {
      "epoch": 0.7664,
      "grad_norm": 1.741668939590454,
      "learning_rate": 0.0005688,
      "loss": 1.2624,
      "step": 479
    },
    {
      "epoch": 0.768,
      "grad_norm": 1.771984577178955,
      "learning_rate": 0.00057,
      "loss": 1.0515,
      "step": 480
    },
    {
      "epoch": 0.7696,
      "grad_norm": 1.5952296257019043,
      "learning_rate": 0.0005711999999999999,
      "loss": 1.2872,
      "step": 481
    },
    {
      "epoch": 0.7712,
      "grad_norm": 1.2628203630447388,
      "learning_rate": 0.0005723999999999999,
      "loss": 1.0806,
      "step": 482
    },
    {
      "epoch": 0.7728,
      "grad_norm": 1.4792985916137695,
      "learning_rate": 0.0005736,
      "loss": 1.962,
      "step": 483
    },
    {
      "epoch": 0.7744,
      "grad_norm": 2.177412509918213,
      "learning_rate": 0.0005747999999999999,
      "loss": 1.4523,
      "step": 484
    },
    {
      "epoch": 0.776,
      "grad_norm": 1.865964412689209,
      "learning_rate": 0.0005759999999999999,
      "loss": 1.401,
      "step": 485
    },
    {
      "epoch": 0.7776,
      "grad_norm": 1.4428671598434448,
      "learning_rate": 0.0005771999999999999,
      "loss": 1.3322,
      "step": 486
    },
    {
      "epoch": 0.7792,
      "grad_norm": 0.9237609505653381,
      "learning_rate": 0.0005784,
      "loss": 1.4704,
      "step": 487
    },
    {
      "epoch": 0.7808,
      "grad_norm": 2.0012426376342773,
      "learning_rate": 0.0005795999999999999,
      "loss": 1.6074,
      "step": 488
    },
    {
      "epoch": 0.7824,
      "grad_norm": 1.1296131610870361,
      "learning_rate": 0.0005807999999999999,
      "loss": 1.4989,
      "step": 489
    },
    {
      "epoch": 0.784,
      "grad_norm": 1.091180682182312,
      "learning_rate": 0.0005819999999999999,
      "loss": 1.4049,
      "step": 490
    },
    {
      "epoch": 0.7856,
      "grad_norm": 1.2416259050369263,
      "learning_rate": 0.0005832,
      "loss": 1.8097,
      "step": 491
    },
    {
      "epoch": 0.7872,
      "grad_norm": 0.9992501139640808,
      "learning_rate": 0.0005843999999999999,
      "loss": 1.5731,
      "step": 492
    },
    {
      "epoch": 0.7888,
      "grad_norm": 1.1882905960083008,
      "learning_rate": 0.0005855999999999999,
      "loss": 1.2349,
      "step": 493
    },
    {
      "epoch": 0.7904,
      "grad_norm": 1.3324135541915894,
      "learning_rate": 0.0005868,
      "loss": 1.749,
      "step": 494
    },
    {
      "epoch": 0.792,
      "grad_norm": 1.3113425970077515,
      "learning_rate": 0.000588,
      "loss": 1.3155,
      "step": 495
    },
    {
      "epoch": 0.7936,
      "grad_norm": 1.333341121673584,
      "learning_rate": 0.0005891999999999999,
      "loss": 1.0152,
      "step": 496
    },
    {
      "epoch": 0.7952,
      "grad_norm": 1.886502742767334,
      "learning_rate": 0.0005903999999999999,
      "loss": 1.5642,
      "step": 497
    },
    {
      "epoch": 0.7968,
      "grad_norm": 2.012117385864258,
      "learning_rate": 0.0005916,
      "loss": 1.6069,
      "step": 498
    },
    {
      "epoch": 0.7984,
      "grad_norm": 2.344853401184082,
      "learning_rate": 0.0005928,
      "loss": 2.0689,
      "step": 499
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.7430222034454346,
      "learning_rate": 0.0005939999999999999,
      "loss": 2.81,
      "step": 500
    },
    {
      "epoch": 0.8016,
      "grad_norm": 2.373655319213867,
      "learning_rate": 0.0005951999999999999,
      "loss": 1.8218,
      "step": 501
    },
    {
      "epoch": 0.8032,
      "grad_norm": 1.2365477085113525,
      "learning_rate": 0.0005964,
      "loss": 1.343,
      "step": 502
    },
    {
      "epoch": 0.8048,
      "grad_norm": 2.100356101989746,
      "learning_rate": 0.0005976,
      "loss": 2.0204,
      "step": 503
    },
    {
      "epoch": 0.8064,
      "grad_norm": 1.6222838163375854,
      "learning_rate": 0.0005987999999999999,
      "loss": 1.8366,
      "step": 504
    },
    {
      "epoch": 0.808,
      "grad_norm": 3.295870542526245,
      "learning_rate": 0.0006,
      "loss": 1.6686,
      "step": 505
    },
    {
      "epoch": 0.8096,
      "grad_norm": 3.5636391639709473,
      "learning_rate": 0.0005987999999999999,
      "loss": 1.958,
      "step": 506
    },
    {
      "epoch": 0.8112,
      "grad_norm": 4.600498199462891,
      "learning_rate": 0.0005976,
      "loss": 1.4633,
      "step": 507
    },
    {
      "epoch": 0.8128,
      "grad_norm": 3.999089241027832,
      "learning_rate": 0.0005964,
      "loss": 1.591,
      "step": 508
    },
    {
      "epoch": 0.8144,
      "grad_norm": 2.074601173400879,
      "learning_rate": 0.0005951999999999999,
      "loss": 1.4589,
      "step": 509
    },
    {
      "epoch": 0.816,
      "grad_norm": 1.2597025632858276,
      "learning_rate": 0.0005939999999999999,
      "loss": 1.517,
      "step": 510
    },
    {
      "epoch": 0.8176,
      "grad_norm": 1.491461157798767,
      "learning_rate": 0.0005928,
      "loss": 1.1305,
      "step": 511
    },
    {
      "epoch": 0.8192,
      "grad_norm": 2.2012178897857666,
      "learning_rate": 0.0005916,
      "loss": 1.2055,
      "step": 512
    },
    {
      "epoch": 0.8208,
      "grad_norm": 2.303264617919922,
      "learning_rate": 0.0005903999999999999,
      "loss": 1.4843,
      "step": 513
    },
    {
      "epoch": 0.8224,
      "grad_norm": 1.3678765296936035,
      "learning_rate": 0.0005891999999999999,
      "loss": 1.1635,
      "step": 514
    },
    {
      "epoch": 0.824,
      "grad_norm": 1.7093764543533325,
      "learning_rate": 0.000588,
      "loss": 1.1574,
      "step": 515
    },
    {
      "epoch": 0.8256,
      "grad_norm": 1.2002806663513184,
      "learning_rate": 0.0005868,
      "loss": 1.1366,
      "step": 516
    },
    {
      "epoch": 0.8272,
      "grad_norm": 1.1055371761322021,
      "learning_rate": 0.0005855999999999999,
      "loss": 1.1778,
      "step": 517
    },
    {
      "epoch": 0.8288,
      "grad_norm": 0.9321176409721375,
      "learning_rate": 0.0005843999999999999,
      "loss": 0.8233,
      "step": 518
    },
    {
      "epoch": 0.8304,
      "grad_norm": 1.3442676067352295,
      "learning_rate": 0.0005832,
      "loss": 1.2463,
      "step": 519
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.9121391177177429,
      "learning_rate": 0.0005819999999999999,
      "loss": 1.3646,
      "step": 520
    },
    {
      "epoch": 0.8336,
      "grad_norm": 1.1537225246429443,
      "learning_rate": 0.0005807999999999999,
      "loss": 1.4225,
      "step": 521
    },
    {
      "epoch": 0.8352,
      "grad_norm": 1.0641944408416748,
      "learning_rate": 0.0005795999999999999,
      "loss": 1.0161,
      "step": 522
    },
    {
      "epoch": 0.8368,
      "grad_norm": 0.7178429961204529,
      "learning_rate": 0.0005784,
      "loss": 1.1851,
      "step": 523
    },
    {
      "epoch": 0.8384,
      "grad_norm": 1.4418388605117798,
      "learning_rate": 0.0005771999999999999,
      "loss": 1.8398,
      "step": 524
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.382843255996704,
      "learning_rate": 0.0005759999999999999,
      "loss": 1.4898,
      "step": 525
    },
    {
      "epoch": 0.8416,
      "grad_norm": 1.0729074478149414,
      "learning_rate": 0.0005747999999999999,
      "loss": 1.2898,
      "step": 526
    },
    {
      "epoch": 0.8432,
      "grad_norm": 0.9983257055282593,
      "learning_rate": 0.0005736,
      "loss": 1.1974,
      "step": 527
    },
    {
      "epoch": 0.8448,
      "grad_norm": 1.1875462532043457,
      "learning_rate": 0.0005723999999999999,
      "loss": 1.1169,
      "step": 528
    },
    {
      "epoch": 0.8464,
      "grad_norm": 1.318334698677063,
      "learning_rate": 0.0005711999999999999,
      "loss": 1.1376,
      "step": 529
    },
    {
      "epoch": 0.848,
      "grad_norm": 1.537840485572815,
      "learning_rate": 0.00057,
      "loss": 1.2072,
      "step": 530
    },
    {
      "epoch": 0.8496,
      "grad_norm": 1.4589056968688965,
      "learning_rate": 0.0005688,
      "loss": 1.373,
      "step": 531
    },
    {
      "epoch": 0.8512,
      "grad_norm": 1.019971251487732,
      "learning_rate": 0.0005675999999999999,
      "loss": 1.0613,
      "step": 532
    },
    {
      "epoch": 0.8528,
      "grad_norm": 1.0122156143188477,
      "learning_rate": 0.0005663999999999999,
      "loss": 0.9874,
      "step": 533
    },
    {
      "epoch": 0.8544,
      "grad_norm": 1.1434595584869385,
      "learning_rate": 0.0005652,
      "loss": 1.2304,
      "step": 534
    },
    {
      "epoch": 0.856,
      "grad_norm": 1.2336163520812988,
      "learning_rate": 0.0005639999999999999,
      "loss": 1.6633,
      "step": 535
    },
    {
      "epoch": 0.8576,
      "grad_norm": 1.306872844696045,
      "learning_rate": 0.0005627999999999999,
      "loss": 1.7359,
      "step": 536
    },
    {
      "epoch": 0.8592,
      "grad_norm": 1.2644526958465576,
      "learning_rate": 0.0005616,
      "loss": 1.2328,
      "step": 537
    },
    {
      "epoch": 0.8608,
      "grad_norm": 1.262831449508667,
      "learning_rate": 0.0005604,
      "loss": 1.0457,
      "step": 538
    },
    {
      "epoch": 0.8624,
      "grad_norm": 0.9390996098518372,
      "learning_rate": 0.0005591999999999999,
      "loss": 1.2251,
      "step": 539
    },
    {
      "epoch": 0.864,
      "grad_norm": 1.208268165588379,
      "learning_rate": 0.000558,
      "loss": 1.3572,
      "step": 540
    },
    {
      "epoch": 0.8656,
      "grad_norm": 1.0636659860610962,
      "learning_rate": 0.0005568,
      "loss": 1.1701,
      "step": 541
    },
    {
      "epoch": 0.8672,
      "grad_norm": 0.9668271541595459,
      "learning_rate": 0.0005556,
      "loss": 1.5845,
      "step": 542
    },
    {
      "epoch": 0.8688,
      "grad_norm": 0.9891708493232727,
      "learning_rate": 0.0005544,
      "loss": 1.3511,
      "step": 543
    },
    {
      "epoch": 0.8704,
      "grad_norm": 1.7175298929214478,
      "learning_rate": 0.0005532,
      "loss": 1.2839,
      "step": 544
    },
    {
      "epoch": 0.872,
      "grad_norm": 0.9767908453941345,
      "learning_rate": 0.000552,
      "loss": 1.5042,
      "step": 545
    },
    {
      "epoch": 0.8736,
      "grad_norm": 1.3343541622161865,
      "learning_rate": 0.0005507999999999999,
      "loss": 1.6681,
      "step": 546
    },
    {
      "epoch": 0.8752,
      "grad_norm": 1.3749518394470215,
      "learning_rate": 0.0005496,
      "loss": 1.9624,
      "step": 547
    },
    {
      "epoch": 0.8768,
      "grad_norm": 2.514359712600708,
      "learning_rate": 0.0005484,
      "loss": 1.895,
      "step": 548
    },
    {
      "epoch": 0.8784,
      "grad_norm": 1.514849066734314,
      "learning_rate": 0.0005472,
      "loss": 1.7603,
      "step": 549
    },
    {
      "epoch": 0.88,
      "grad_norm": 1.4065580368041992,
      "learning_rate": 0.0005459999999999999,
      "loss": 1.9368,
      "step": 550
    },
    {
      "epoch": 0.8816,
      "grad_norm": 4.109616756439209,
      "learning_rate": 0.0005448,
      "loss": 1.9927,
      "step": 551
    },
    {
      "epoch": 0.8832,
      "grad_norm": 5.660764694213867,
      "learning_rate": 0.0005436,
      "loss": 2.4962,
      "step": 552
    },
    {
      "epoch": 0.8848,
      "grad_norm": 2.027193307876587,
      "learning_rate": 0.0005424,
      "loss": 1.6865,
      "step": 553
    },
    {
      "epoch": 0.8864,
      "grad_norm": 4.918600559234619,
      "learning_rate": 0.0005411999999999999,
      "loss": 2.1457,
      "step": 554
    },
    {
      "epoch": 0.888,
      "grad_norm": 2.522416114807129,
      "learning_rate": 0.00054,
      "loss": 1.5093,
      "step": 555
    },
    {
      "epoch": 0.8896,
      "grad_norm": 1.574242353439331,
      "learning_rate": 0.0005388,
      "loss": 1.1618,
      "step": 556
    },
    {
      "epoch": 0.8912,
      "grad_norm": 2.081839084625244,
      "learning_rate": 0.0005376,
      "loss": 1.2351,
      "step": 557
    },
    {
      "epoch": 0.8928,
      "grad_norm": 2.582669734954834,
      "learning_rate": 0.0005363999999999999,
      "loss": 1.6881,
      "step": 558
    },
    {
      "epoch": 0.8944,
      "grad_norm": 2.058865785598755,
      "learning_rate": 0.0005352,
      "loss": 1.3369,
      "step": 559
    },
    {
      "epoch": 0.896,
      "grad_norm": 5.098209381103516,
      "learning_rate": 0.000534,
      "loss": 1.7725,
      "step": 560
    },
    {
      "epoch": 0.8976,
      "grad_norm": 4.526047229766846,
      "learning_rate": 0.0005327999999999999,
      "loss": 1.9443,
      "step": 561
    },
    {
      "epoch": 0.8992,
      "grad_norm": 3.9119491577148438,
      "learning_rate": 0.0005315999999999999,
      "loss": 1.5399,
      "step": 562
    },
    {
      "epoch": 0.9008,
      "grad_norm": 0.8436188101768494,
      "learning_rate": 0.0005304,
      "loss": 1.1211,
      "step": 563
    },
    {
      "epoch": 0.9024,
      "grad_norm": 0.8627029061317444,
      "learning_rate": 0.0005292,
      "loss": 1.2774,
      "step": 564
    },
    {
      "epoch": 0.904,
      "grad_norm": 1.1870328187942505,
      "learning_rate": 0.0005279999999999999,
      "loss": 1.2342,
      "step": 565
    },
    {
      "epoch": 0.9056,
      "grad_norm": 1.0261473655700684,
      "learning_rate": 0.0005267999999999999,
      "loss": 1.0405,
      "step": 566
    },
    {
      "epoch": 0.9072,
      "grad_norm": 0.6334408521652222,
      "learning_rate": 0.0005256,
      "loss": 1.0547,
      "step": 567
    },
    {
      "epoch": 0.9088,
      "grad_norm": 0.7928243279457092,
      "learning_rate": 0.0005244,
      "loss": 0.9138,
      "step": 568
    },
    {
      "epoch": 0.9104,
      "grad_norm": 0.9043545126914978,
      "learning_rate": 0.0005231999999999999,
      "loss": 1.0988,
      "step": 569
    },
    {
      "epoch": 0.912,
      "grad_norm": 0.9109718203544617,
      "learning_rate": 0.000522,
      "loss": 1.0209,
      "step": 570
    },
    {
      "epoch": 0.9136,
      "grad_norm": 0.8105588555335999,
      "learning_rate": 0.0005208,
      "loss": 1.2233,
      "step": 571
    },
    {
      "epoch": 0.9152,
      "grad_norm": 2.0293259620666504,
      "learning_rate": 0.0005195999999999999,
      "loss": 1.4025,
      "step": 572
    },
    {
      "epoch": 0.9168,
      "grad_norm": 1.1730594635009766,
      "learning_rate": 0.0005183999999999999,
      "loss": 1.481,
      "step": 573
    },
    {
      "epoch": 0.9184,
      "grad_norm": 0.8027293086051941,
      "learning_rate": 0.0005172,
      "loss": 1.1333,
      "step": 574
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.8858001828193665,
      "learning_rate": 0.000516,
      "loss": 1.4056,
      "step": 575
    },
    {
      "epoch": 0.9216,
      "grad_norm": 1.3030261993408203,
      "learning_rate": 0.0005147999999999999,
      "loss": 1.212,
      "step": 576
    },
    {
      "epoch": 0.9232,
      "grad_norm": 0.9441176652908325,
      "learning_rate": 0.0005135999999999999,
      "loss": 1.6681,
      "step": 577
    },
    {
      "epoch": 0.9248,
      "grad_norm": 0.9312158226966858,
      "learning_rate": 0.0005124,
      "loss": 1.2352,
      "step": 578
    },
    {
      "epoch": 0.9264,
      "grad_norm": 1.1646393537521362,
      "learning_rate": 0.0005112,
      "loss": 1.1658,
      "step": 579
    },
    {
      "epoch": 0.928,
      "grad_norm": 1.2840147018432617,
      "learning_rate": 0.0005099999999999999,
      "loss": 1.0198,
      "step": 580
    },
    {
      "epoch": 0.9296,
      "grad_norm": 1.2597519159317017,
      "learning_rate": 0.0005087999999999999,
      "loss": 1.0831,
      "step": 581
    },
    {
      "epoch": 0.9312,
      "grad_norm": 0.7873828411102295,
      "learning_rate": 0.0005076,
      "loss": 1.0068,
      "step": 582
    },
    {
      "epoch": 0.9328,
      "grad_norm": 1.5674644708633423,
      "learning_rate": 0.0005064,
      "loss": 1.3783,
      "step": 583
    },
    {
      "epoch": 0.9344,
      "grad_norm": 0.9917027354240417,
      "learning_rate": 0.0005051999999999999,
      "loss": 0.9963,
      "step": 584
    },
    {
      "epoch": 0.936,
      "grad_norm": 0.9538294672966003,
      "learning_rate": 0.0005039999999999999,
      "loss": 0.9023,
      "step": 585
    },
    {
      "epoch": 0.9376,
      "grad_norm": 1.1332108974456787,
      "learning_rate": 0.0005028,
      "loss": 1.0699,
      "step": 586
    },
    {
      "epoch": 0.9392,
      "grad_norm": 1.0121673345565796,
      "learning_rate": 0.0005015999999999999,
      "loss": 1.0387,
      "step": 587
    },
    {
      "epoch": 0.9408,
      "grad_norm": 1.13718581199646,
      "learning_rate": 0.0005003999999999999,
      "loss": 1.1858,
      "step": 588
    },
    {
      "epoch": 0.9424,
      "grad_norm": 1.2693606615066528,
      "learning_rate": 0.0004991999999999999,
      "loss": 1.0405,
      "step": 589
    },
    {
      "epoch": 0.944,
      "grad_norm": 1.4965566396713257,
      "learning_rate": 0.000498,
      "loss": 1.326,
      "step": 590
    },
    {
      "epoch": 0.9456,
      "grad_norm": 1.2708925008773804,
      "learning_rate": 0.0004967999999999999,
      "loss": 1.2337,
      "step": 591
    },
    {
      "epoch": 0.9472,
      "grad_norm": 2.260007858276367,
      "learning_rate": 0.0004955999999999999,
      "loss": 1.5223,
      "step": 592
    },
    {
      "epoch": 0.9488,
      "grad_norm": 1.3588523864746094,
      "learning_rate": 0.0004944,
      "loss": 1.4486,
      "step": 593
    },
    {
      "epoch": 0.9504,
      "grad_norm": 3.3943569660186768,
      "learning_rate": 0.0004932,
      "loss": 1.5663,
      "step": 594
    },
    {
      "epoch": 0.952,
      "grad_norm": 1.5709065198898315,
      "learning_rate": 0.0004919999999999999,
      "loss": 1.3337,
      "step": 595
    },
    {
      "epoch": 0.9536,
      "grad_norm": 1.6011497974395752,
      "learning_rate": 0.0004907999999999999,
      "loss": 1.4023,
      "step": 596
    },
    {
      "epoch": 0.9552,
      "grad_norm": 1.2253276109695435,
      "learning_rate": 0.0004896,
      "loss": 1.0934,
      "step": 597
    },
    {
      "epoch": 0.9568,
      "grad_norm": 9.454032897949219,
      "learning_rate": 0.0004883999999999999,
      "loss": 2.9201,
      "step": 598
    },
    {
      "epoch": 0.9584,
      "grad_norm": 1.7915419340133667,
      "learning_rate": 0.00048719999999999997,
      "loss": 1.5261,
      "step": 599
    },
    {
      "epoch": 0.96,
      "grad_norm": NaN,
      "learning_rate": 0.00048719999999999997,
      "loss": 1.3199,
      "step": 600
    },
    {
      "epoch": 0.9616,
      "grad_norm": 18.144990921020508,
      "learning_rate": 0.000486,
      "loss": 4.1466,
      "step": 601
    },
    {
      "epoch": 0.9632,
      "grad_norm": 1.2664769887924194,
      "learning_rate": 0.00048479999999999997,
      "loss": 1.6238,
      "step": 602
    },
    {
      "epoch": 0.9648,
      "grad_norm": 2.453564405441284,
      "learning_rate": 0.0004836,
      "loss": 1.5469,
      "step": 603
    },
    {
      "epoch": 0.9664,
      "grad_norm": 2.737936496734619,
      "learning_rate": 0.00048239999999999996,
      "loss": 1.3178,
      "step": 604
    },
    {
      "epoch": 0.968,
      "grad_norm": 2.828806161880493,
      "learning_rate": 0.0004812,
      "loss": 1.4034,
      "step": 605
    },
    {
      "epoch": 0.9696,
      "grad_norm": 1.4219012260437012,
      "learning_rate": 0.00047999999999999996,
      "loss": 1.0615,
      "step": 606
    },
    {
      "epoch": 0.9712,
      "grad_norm": 2.025907039642334,
      "learning_rate": 0.0004788,
      "loss": 1.19,
      "step": 607
    },
    {
      "epoch": 0.9728,
      "grad_norm": 0.9138876795768738,
      "learning_rate": 0.0004776,
      "loss": 1.1638,
      "step": 608
    },
    {
      "epoch": 0.9744,
      "grad_norm": 0.7301196455955505,
      "learning_rate": 0.0004764,
      "loss": 1.2678,
      "step": 609
    },
    {
      "epoch": 0.976,
      "grad_norm": 1.6543656587600708,
      "learning_rate": 0.0004752,
      "loss": 1.1815,
      "step": 610
    },
    {
      "epoch": 0.9776,
      "grad_norm": 1.8122645616531372,
      "learning_rate": 0.000474,
      "loss": 1.0917,
      "step": 611
    },
    {
      "epoch": 0.9792,
      "grad_norm": 1.609754204750061,
      "learning_rate": 0.0004728,
      "loss": 1.1377,
      "step": 612
    },
    {
      "epoch": 0.9808,
      "grad_norm": 1.2491132020950317,
      "learning_rate": 0.00047159999999999997,
      "loss": 0.9748,
      "step": 613
    },
    {
      "epoch": 0.9824,
      "grad_norm": 1.6153700351715088,
      "learning_rate": 0.0004704,
      "loss": 1.1825,
      "step": 614
    },
    {
      "epoch": 0.984,
      "grad_norm": 2.4447743892669678,
      "learning_rate": 0.00046919999999999997,
      "loss": 1.1472,
      "step": 615
    },
    {
      "epoch": 0.9856,
      "grad_norm": 0.93949294090271,
      "learning_rate": 0.000468,
      "loss": 0.9327,
      "step": 616
    },
    {
      "epoch": 0.9872,
      "grad_norm": 1.0353221893310547,
      "learning_rate": 0.00046679999999999996,
      "loss": 1.3792,
      "step": 617
    },
    {
      "epoch": 0.9888,
      "grad_norm": 1.5396970510482788,
      "learning_rate": 0.0004656,
      "loss": 1.4076,
      "step": 618
    },
    {
      "epoch": 0.9904,
      "grad_norm": 2.1144979000091553,
      "learning_rate": 0.00046439999999999996,
      "loss": 1.2042,
      "step": 619
    },
    {
      "epoch": 0.992,
      "grad_norm": 1.649453043937683,
      "learning_rate": 0.0004632,
      "loss": 1.033,
      "step": 620
    },
    {
      "epoch": 0.9936,
      "grad_norm": 1.555721640586853,
      "learning_rate": 0.00046199999999999995,
      "loss": 1.4478,
      "step": 621
    },
    {
      "epoch": 0.9952,
      "grad_norm": 2.0463335514068604,
      "learning_rate": 0.0004608,
      "loss": 1.427,
      "step": 622
    },
    {
      "epoch": 0.9968,
      "grad_norm": 1.6396963596343994,
      "learning_rate": 0.00045959999999999995,
      "loss": 1.4215,
      "step": 623
    },
    {
      "epoch": 0.9984,
      "grad_norm": 1.0603671073913574,
      "learning_rate": 0.0004584,
      "loss": 1.2777,
      "step": 624
    },
    {
      "epoch": 1.0,
      "grad_norm": 2.150242328643799,
      "learning_rate": 0.00045719999999999995,
      "loss": 1.5985,
      "step": 625
    },
    {
      "epoch": 1.0016,
      "grad_norm": 13.092059135437012,
      "learning_rate": 0.00045599999999999997,
      "loss": 3.3297,
      "step": 626
    },
    {
      "epoch": 1.0032,
      "grad_norm": 5.357777118682861,
      "learning_rate": 0.00045479999999999994,
      "loss": 1.8506,
      "step": 627
    },
    {
      "epoch": 1.0048,
      "grad_norm": 3.857879400253296,
      "learning_rate": 0.00045359999999999997,
      "loss": 1.4082,
      "step": 628
    },
    {
      "epoch": 1.0064,
      "grad_norm": 5.025846004486084,
      "learning_rate": 0.00045239999999999994,
      "loss": 2.445,
      "step": 629
    },
    {
      "epoch": 1.008,
      "grad_norm": 3.71512508392334,
      "learning_rate": 0.00045119999999999996,
      "loss": 2.0449,
      "step": 630
    },
    {
      "epoch": 1.0096,
      "grad_norm": 4.47951078414917,
      "learning_rate": 0.00045,
      "loss": 1.9927,
      "step": 631
    },
    {
      "epoch": 1.0112,
      "grad_norm": 2.713778257369995,
      "learning_rate": 0.00044879999999999996,
      "loss": 1.4958,
      "step": 632
    },
    {
      "epoch": 1.0128,
      "grad_norm": 2.6333401203155518,
      "learning_rate": 0.0004476,
      "loss": 1.3756,
      "step": 633
    },
    {
      "epoch": 1.0144,
      "grad_norm": 3.1156418323516846,
      "learning_rate": 0.00044639999999999995,
      "loss": 1.6268,
      "step": 634
    },
    {
      "epoch": 1.016,
      "grad_norm": 2.78902268409729,
      "learning_rate": 0.0004452,
      "loss": 1.8211,
      "step": 635
    },
    {
      "epoch": 1.0176,
      "grad_norm": 3.0045857429504395,
      "learning_rate": 0.00044399999999999995,
      "loss": 1.1757,
      "step": 636
    },
    {
      "epoch": 1.0192,
      "grad_norm": 1.4704291820526123,
      "learning_rate": 0.0004428,
      "loss": 1.2811,
      "step": 637
    },
    {
      "epoch": 1.0208,
      "grad_norm": 1.3432084321975708,
      "learning_rate": 0.00044159999999999995,
      "loss": 1.0086,
      "step": 638
    },
    {
      "epoch": 1.0224,
      "grad_norm": 0.842569887638092,
      "learning_rate": 0.00044039999999999997,
      "loss": 1.3255,
      "step": 639
    },
    {
      "epoch": 1.024,
      "grad_norm": 0.8691660761833191,
      "learning_rate": 0.00043919999999999994,
      "loss": 1.1076,
      "step": 640
    },
    {
      "epoch": 1.0256,
      "grad_norm": 1.083778738975525,
      "learning_rate": 0.00043799999999999997,
      "loss": 1.0929,
      "step": 641
    },
    {
      "epoch": 1.0272,
      "grad_norm": 1.1408025026321411,
      "learning_rate": 0.00043679999999999994,
      "loss": 1.2497,
      "step": 642
    },
    {
      "epoch": 1.0288,
      "grad_norm": 0.8224440217018127,
      "learning_rate": 0.00043559999999999996,
      "loss": 1.2853,
      "step": 643
    },
    {
      "epoch": 1.0304,
      "grad_norm": 0.7420323491096497,
      "learning_rate": 0.00043439999999999993,
      "loss": 1.187,
      "step": 644
    },
    {
      "epoch": 1.032,
      "grad_norm": 0.7818359732627869,
      "learning_rate": 0.00043319999999999996,
      "loss": 1.0613,
      "step": 645
    },
    {
      "epoch": 1.0336,
      "grad_norm": 1.2085120677947998,
      "learning_rate": 0.00043199999999999993,
      "loss": 1.0593,
      "step": 646
    },
    {
      "epoch": 1.0352,
      "grad_norm": 0.6779820322990417,
      "learning_rate": 0.00043079999999999995,
      "loss": 0.9878,
      "step": 647
    },
    {
      "epoch": 1.0368,
      "grad_norm": 1.1257340908050537,
      "learning_rate": 0.0004295999999999999,
      "loss": 1.0076,
      "step": 648
    },
    {
      "epoch": 1.0384,
      "grad_norm": 0.6911525726318359,
      "learning_rate": 0.00042839999999999995,
      "loss": 1.1204,
      "step": 649
    },
    {
      "epoch": 1.04,
      "grad_norm": 1.0192064046859741,
      "learning_rate": 0.0004271999999999999,
      "loss": 1.217,
      "step": 650
    },
    {
      "epoch": 1.0416,
      "grad_norm": 0.8528116345405579,
      "learning_rate": 0.00042599999999999995,
      "loss": 1.1124,
      "step": 651
    },
    {
      "epoch": 1.0432,
      "grad_norm": 0.8429757356643677,
      "learning_rate": 0.0004247999999999999,
      "loss": 1.0387,
      "step": 652
    },
    {
      "epoch": 1.0448,
      "grad_norm": 1.0031050443649292,
      "learning_rate": 0.00042359999999999994,
      "loss": 1.1567,
      "step": 653
    },
    {
      "epoch": 1.0464,
      "grad_norm": 1.1668227910995483,
      "learning_rate": 0.0004223999999999999,
      "loss": 1.3264,
      "step": 654
    },
    {
      "epoch": 1.048,
      "grad_norm": 0.8505134582519531,
      "learning_rate": 0.00042119999999999994,
      "loss": 0.8013,
      "step": 655
    },
    {
      "epoch": 1.0496,
      "grad_norm": 0.7736939191818237,
      "learning_rate": 0.00041999999999999996,
      "loss": 1.0249,
      "step": 656
    },
    {
      "epoch": 1.0512,
      "grad_norm": 1.082900047302246,
      "learning_rate": 0.00041879999999999993,
      "loss": 1.0062,
      "step": 657
    },
    {
      "epoch": 1.0528,
      "grad_norm": 1.9835671186447144,
      "learning_rate": 0.00041759999999999996,
      "loss": 1.4081,
      "step": 658
    },
    {
      "epoch": 1.0544,
      "grad_norm": 0.9174219965934753,
      "learning_rate": 0.00041639999999999993,
      "loss": 0.8602,
      "step": 659
    },
    {
      "epoch": 1.056,
      "grad_norm": 0.8988387584686279,
      "learning_rate": 0.00041519999999999995,
      "loss": 1.0813,
      "step": 660
    },
    {
      "epoch": 1.0576,
      "grad_norm": 0.9161027669906616,
      "learning_rate": 0.0004139999999999999,
      "loss": 0.9137,
      "step": 661
    },
    {
      "epoch": 1.0592,
      "grad_norm": 1.0347057580947876,
      "learning_rate": 0.00041279999999999995,
      "loss": 0.9526,
      "step": 662
    },
    {
      "epoch": 1.0608,
      "grad_norm": 0.9697722792625427,
      "learning_rate": 0.0004116,
      "loss": 1.2357,
      "step": 663
    },
    {
      "epoch": 1.0624,
      "grad_norm": 1.1263432502746582,
      "learning_rate": 0.0004104,
      "loss": 1.2324,
      "step": 664
    },
    {
      "epoch": 1.064,
      "grad_norm": 1.4703420400619507,
      "learning_rate": 0.00040919999999999997,
      "loss": 1.2418,
      "step": 665
    },
    {
      "epoch": 1.0656,
      "grad_norm": 1.2557834386825562,
      "learning_rate": 0.000408,
      "loss": 1.5181,
      "step": 666
    },
    {
      "epoch": 1.0672,
      "grad_norm": 1.1496392488479614,
      "learning_rate": 0.00040679999999999997,
      "loss": 1.5552,
      "step": 667
    },
    {
      "epoch": 1.0688,
      "grad_norm": 1.0062506198883057,
      "learning_rate": 0.0004056,
      "loss": 1.3242,
      "step": 668
    },
    {
      "epoch": 1.0704,
      "grad_norm": 0.8978244066238403,
      "learning_rate": 0.0004044,
      "loss": 1.3264,
      "step": 669
    },
    {
      "epoch": 1.072,
      "grad_norm": 1.2901413440704346,
      "learning_rate": 0.0004032,
      "loss": 1.3077,
      "step": 670
    },
    {
      "epoch": 1.0735999999999999,
      "grad_norm": 1.202578067779541,
      "learning_rate": 0.000402,
      "loss": 1.1621,
      "step": 671
    },
    {
      "epoch": 1.0752,
      "grad_norm": 1.4110441207885742,
      "learning_rate": 0.0004008,
      "loss": 1.5065,
      "step": 672
    },
    {
      "epoch": 1.0768,
      "grad_norm": 1.0681021213531494,
      "learning_rate": 0.0003996,
      "loss": 1.5376,
      "step": 673
    },
    {
      "epoch": 1.0784,
      "grad_norm": 1.355599045753479,
      "learning_rate": 0.0003984,
      "loss": 1.7659,
      "step": 674
    },
    {
      "epoch": 1.08,
      "grad_norm": 1.245712161064148,
      "learning_rate": 0.0003972,
      "loss": 1.6335,
      "step": 675
    },
    {
      "epoch": 1.0816,
      "grad_norm": 10.161030769348145,
      "learning_rate": 0.000396,
      "loss": 2.7562,
      "step": 676
    },
    {
      "epoch": 1.0832,
      "grad_norm": 5.946995735168457,
      "learning_rate": 0.0003948,
      "loss": 2.2905,
      "step": 677
    },
    {
      "epoch": 1.0848,
      "grad_norm": 3.535452365875244,
      "learning_rate": 0.00039359999999999997,
      "loss": 1.8719,
      "step": 678
    },
    {
      "epoch": 1.0864,
      "grad_norm": 1.3516403436660767,
      "learning_rate": 0.0003924,
      "loss": 1.3978,
      "step": 679
    },
    {
      "epoch": 1.088,
      "grad_norm": 1.6472234725952148,
      "learning_rate": 0.00039119999999999997,
      "loss": 1.2273,
      "step": 680
    },
    {
      "epoch": 1.0896,
      "grad_norm": 3.0599803924560547,
      "learning_rate": 0.00039,
      "loss": 1.4096,
      "step": 681
    },
    {
      "epoch": 1.0912,
      "grad_norm": 3.078174352645874,
      "learning_rate": 0.00038879999999999996,
      "loss": 1.2859,
      "step": 682
    },
    {
      "epoch": 1.0928,
      "grad_norm": 3.5751090049743652,
      "learning_rate": 0.0003876,
      "loss": 1.4634,
      "step": 683
    },
    {
      "epoch": 1.0944,
      "grad_norm": 2.058401584625244,
      "learning_rate": 0.00038639999999999996,
      "loss": 1.6292,
      "step": 684
    },
    {
      "epoch": 1.096,
      "grad_norm": 1.8631670475006104,
      "learning_rate": 0.0003852,
      "loss": 1.1733,
      "step": 685
    },
    {
      "epoch": 1.0976,
      "grad_norm": 1.9622036218643188,
      "learning_rate": 0.00038399999999999996,
      "loss": 1.0265,
      "step": 686
    },
    {
      "epoch": 1.0992,
      "grad_norm": 1.0497528314590454,
      "learning_rate": 0.0003828,
      "loss": 0.9263,
      "step": 687
    },
    {
      "epoch": 1.1008,
      "grad_norm": 1.0042531490325928,
      "learning_rate": 0.00038159999999999995,
      "loss": 1.0379,
      "step": 688
    },
    {
      "epoch": 1.1024,
      "grad_norm": 0.7379323244094849,
      "learning_rate": 0.0003804,
      "loss": 0.8909,
      "step": 689
    },
    {
      "epoch": 1.104,
      "grad_norm": 0.9978313446044922,
      "learning_rate": 0.00037919999999999995,
      "loss": 0.9649,
      "step": 690
    },
    {
      "epoch": 1.1056,
      "grad_norm": 1.4265302419662476,
      "learning_rate": 0.00037799999999999997,
      "loss": 0.8303,
      "step": 691
    },
    {
      "epoch": 1.1072,
      "grad_norm": 1.609402060508728,
      "learning_rate": 0.00037679999999999994,
      "loss": 1.0412,
      "step": 692
    },
    {
      "epoch": 1.1088,
      "grad_norm": 1.7801131010055542,
      "learning_rate": 0.00037559999999999997,
      "loss": 1.2652,
      "step": 693
    },
    {
      "epoch": 1.1104,
      "grad_norm": 0.9285919666290283,
      "learning_rate": 0.0003744,
      "loss": 1.2096,
      "step": 694
    },
    {
      "epoch": 1.112,
      "grad_norm": 0.9512993693351746,
      "learning_rate": 0.00037319999999999996,
      "loss": 1.1913,
      "step": 695
    },
    {
      "epoch": 1.1136,
      "grad_norm": 0.9945847392082214,
      "learning_rate": 0.000372,
      "loss": 1.1681,
      "step": 696
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.8993807435035706,
      "learning_rate": 0.00037079999999999996,
      "loss": 1.2373,
      "step": 697
    },
    {
      "epoch": 1.1168,
      "grad_norm": 1.0664983987808228,
      "learning_rate": 0.0003696,
      "loss": 1.0773,
      "step": 698
    },
    {
      "epoch": 1.1184,
      "grad_norm": 1.7917791604995728,
      "learning_rate": 0.00036839999999999996,
      "loss": 1.4875,
      "step": 699
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.8973643183708191,
      "learning_rate": 0.0003672,
      "loss": 1.1206,
      "step": 700
    },
    {
      "epoch": 1.1216,
      "grad_norm": 1.2886801958084106,
      "learning_rate": 0.00036599999999999995,
      "loss": 0.9865,
      "step": 701
    },
    {
      "epoch": 1.1232,
      "grad_norm": 1.4595153331756592,
      "learning_rate": 0.0003648,
      "loss": 1.388,
      "step": 702
    },
    {
      "epoch": 1.1248,
      "grad_norm": 1.1467390060424805,
      "learning_rate": 0.00036359999999999995,
      "loss": 0.9407,
      "step": 703
    },
    {
      "epoch": 1.1264,
      "grad_norm": 1.0743904113769531,
      "learning_rate": 0.00036239999999999997,
      "loss": 1.0897,
      "step": 704
    },
    {
      "epoch": 1.1280000000000001,
      "grad_norm": 0.9008836150169373,
      "learning_rate": 0.00036119999999999994,
      "loss": 0.9555,
      "step": 705
    },
    {
      "epoch": 1.1296,
      "grad_norm": 1.016830325126648,
      "learning_rate": 0.00035999999999999997,
      "loss": 1.0143,
      "step": 706
    },
    {
      "epoch": 1.1312,
      "grad_norm": 1.8725807666778564,
      "learning_rate": 0.00035879999999999994,
      "loss": 1.4681,
      "step": 707
    },
    {
      "epoch": 1.1328,
      "grad_norm": 1.0106738805770874,
      "learning_rate": 0.00035759999999999996,
      "loss": 1.0107,
      "step": 708
    },
    {
      "epoch": 1.1344,
      "grad_norm": 1.0680439472198486,
      "learning_rate": 0.00035639999999999994,
      "loss": 0.8917,
      "step": 709
    },
    {
      "epoch": 1.1360000000000001,
      "grad_norm": 0.9104785323143005,
      "learning_rate": 0.00035519999999999996,
      "loss": 1.1333,
      "step": 710
    },
    {
      "epoch": 1.1376,
      "grad_norm": 1.0638065338134766,
      "learning_rate": 0.00035399999999999993,
      "loss": 0.9465,
      "step": 711
    },
    {
      "epoch": 1.1392,
      "grad_norm": 0.7474643588066101,
      "learning_rate": 0.00035279999999999996,
      "loss": 0.7617,
      "step": 712
    },
    {
      "epoch": 1.1408,
      "grad_norm": 0.8218055963516235,
      "learning_rate": 0.0003515999999999999,
      "loss": 1.0069,
      "step": 713
    },
    {
      "epoch": 1.1424,
      "grad_norm": 1.1213569641113281,
      "learning_rate": 0.00035039999999999995,
      "loss": 1.0228,
      "step": 714
    },
    {
      "epoch": 1.144,
      "grad_norm": 1.0540097951889038,
      "learning_rate": 0.0003491999999999999,
      "loss": 1.2662,
      "step": 715
    },
    {
      "epoch": 1.1456,
      "grad_norm": 1.019623875617981,
      "learning_rate": 0.00034799999999999995,
      "loss": 1.2725,
      "step": 716
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.9017633199691772,
      "learning_rate": 0.0003467999999999999,
      "loss": 0.8269,
      "step": 717
    },
    {
      "epoch": 1.1488,
      "grad_norm": 3.7017982006073,
      "learning_rate": 0.00034559999999999994,
      "loss": 2.0837,
      "step": 718
    },
    {
      "epoch": 1.1504,
      "grad_norm": 1.0694856643676758,
      "learning_rate": 0.00034439999999999997,
      "loss": 1.0984,
      "step": 719
    },
    {
      "epoch": 1.152,
      "grad_norm": 1.0227575302124023,
      "learning_rate": 0.00034319999999999994,
      "loss": 1.1509,
      "step": 720
    },
    {
      "epoch": 1.1536,
      "grad_norm": 0.8948163390159607,
      "learning_rate": 0.00034199999999999996,
      "loss": 0.8165,
      "step": 721
    },
    {
      "epoch": 1.1552,
      "grad_norm": 1.575053334236145,
      "learning_rate": 0.00034079999999999994,
      "loss": 1.4505,
      "step": 722
    },
    {
      "epoch": 1.1568,
      "grad_norm": 1.6160234212875366,
      "learning_rate": 0.00033959999999999996,
      "loss": 1.5606,
      "step": 723
    },
    {
      "epoch": 1.1584,
      "grad_norm": 1.469820499420166,
      "learning_rate": 0.00033839999999999993,
      "loss": 1.3345,
      "step": 724
    },
    {
      "epoch": 1.16,
      "grad_norm": 2.6582064628601074,
      "learning_rate": 0.0003372,
      "loss": 1.9885,
      "step": 725
    },
    {
      "epoch": 1.1616,
      "grad_norm": 8.827315330505371,
      "learning_rate": 0.000336,
      "loss": 2.7075,
      "step": 726
    },
    {
      "epoch": 1.1632,
      "grad_norm": 5.201417922973633,
      "learning_rate": 0.0003348,
      "loss": 2.0887,
      "step": 727
    },
    {
      "epoch": 1.1648,
      "grad_norm": 2.5593836307525635,
      "learning_rate": 0.0003336,
      "loss": 1.3964,
      "step": 728
    },
    {
      "epoch": 1.1663999999999999,
      "grad_norm": 2.476527452468872,
      "learning_rate": 0.0003324,
      "loss": 1.6703,
      "step": 729
    },
    {
      "epoch": 1.168,
      "grad_norm": 1.3854165077209473,
      "learning_rate": 0.0003312,
      "loss": 1.4116,
      "step": 730
    },
    {
      "epoch": 1.1696,
      "grad_norm": 1.7695822715759277,
      "learning_rate": 0.00033,
      "loss": 1.7663,
      "step": 731
    },
    {
      "epoch": 1.1712,
      "grad_norm": 1.7809518575668335,
      "learning_rate": 0.0003288,
      "loss": 2.2482,
      "step": 732
    },
    {
      "epoch": 1.1728,
      "grad_norm": 1.5759507417678833,
      "learning_rate": 0.0003276,
      "loss": 1.0389,
      "step": 733
    },
    {
      "epoch": 1.1743999999999999,
      "grad_norm": 3.968517780303955,
      "learning_rate": 0.0003264,
      "loss": 1.1723,
      "step": 734
    },
    {
      "epoch": 1.176,
      "grad_norm": 3.551710367202759,
      "learning_rate": 0.0003252,
      "loss": 1.1122,
      "step": 735
    },
    {
      "epoch": 1.1776,
      "grad_norm": 3.917438507080078,
      "learning_rate": 0.000324,
      "loss": 1.3001,
      "step": 736
    },
    {
      "epoch": 1.1792,
      "grad_norm": 2.960092306137085,
      "learning_rate": 0.0003228,
      "loss": 1.3047,
      "step": 737
    },
    {
      "epoch": 1.1808,
      "grad_norm": 2.588700532913208,
      "learning_rate": 0.0003216,
      "loss": 1.0372,
      "step": 738
    },
    {
      "epoch": 1.1824,
      "grad_norm": 2.2377519607543945,
      "learning_rate": 0.0003204,
      "loss": 1.2321,
      "step": 739
    },
    {
      "epoch": 1.184,
      "grad_norm": 1.3439960479736328,
      "learning_rate": 0.0003192,
      "loss": 1.2945,
      "step": 740
    },
    {
      "epoch": 1.1856,
      "grad_norm": 0.8878504037857056,
      "learning_rate": 0.000318,
      "loss": 1.085,
      "step": 741
    },
    {
      "epoch": 1.1872,
      "grad_norm": 0.8282541632652283,
      "learning_rate": 0.0003168,
      "loss": 1.1087,
      "step": 742
    },
    {
      "epoch": 1.1888,
      "grad_norm": 0.8017875552177429,
      "learning_rate": 0.0003156,
      "loss": 0.9878,
      "step": 743
    },
    {
      "epoch": 1.1904,
      "grad_norm": 1.347516655921936,
      "learning_rate": 0.0003144,
      "loss": 1.0613,
      "step": 744
    },
    {
      "epoch": 1.192,
      "grad_norm": 1.9655871391296387,
      "learning_rate": 0.00031319999999999997,
      "loss": 1.0696,
      "step": 745
    },
    {
      "epoch": 1.1936,
      "grad_norm": 1.588913083076477,
      "learning_rate": 0.000312,
      "loss": 1.2005,
      "step": 746
    },
    {
      "epoch": 1.1952,
      "grad_norm": 1.29644775390625,
      "learning_rate": 0.00031079999999999997,
      "loss": 0.904,
      "step": 747
    },
    {
      "epoch": 1.1968,
      "grad_norm": 2.8275885581970215,
      "learning_rate": 0.0003096,
      "loss": 1.5441,
      "step": 748
    },
    {
      "epoch": 1.1984,
      "grad_norm": 1.0542739629745483,
      "learning_rate": 0.00030839999999999996,
      "loss": 0.9999,
      "step": 749
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.7280116677284241,
      "learning_rate": 0.0003072,
      "loss": 0.8393,
      "step": 750
    },
    {
      "epoch": 1.2016,
      "grad_norm": 0.7934659123420715,
      "learning_rate": 0.00030599999999999996,
      "loss": 1.1939,
      "step": 751
    },
    {
      "epoch": 1.2032,
      "grad_norm": 0.8406733274459839,
      "learning_rate": 0.0003048,
      "loss": 0.9562,
      "step": 752
    },
    {
      "epoch": 1.2048,
      "grad_norm": 1.0488728284835815,
      "learning_rate": 0.00030359999999999995,
      "loss": 1.2521,
      "step": 753
    },
    {
      "epoch": 1.2064,
      "grad_norm": 0.8879828453063965,
      "learning_rate": 0.0003024,
      "loss": 0.7971,
      "step": 754
    },
    {
      "epoch": 1.208,
      "grad_norm": 1.3260725736618042,
      "learning_rate": 0.00030119999999999995,
      "loss": 1.2188,
      "step": 755
    },
    {
      "epoch": 1.2096,
      "grad_norm": 1.1654318571090698,
      "learning_rate": 0.0003,
      "loss": 0.9387,
      "step": 756
    },
    {
      "epoch": 1.2112,
      "grad_norm": 1.351473093032837,
      "learning_rate": 0.0002988,
      "loss": 1.0464,
      "step": 757
    },
    {
      "epoch": 1.2128,
      "grad_norm": 0.9511071443557739,
      "learning_rate": 0.00029759999999999997,
      "loss": 0.9436,
      "step": 758
    },
    {
      "epoch": 1.2144,
      "grad_norm": 1.179603934288025,
      "learning_rate": 0.0002964,
      "loss": 1.1701,
      "step": 759
    },
    {
      "epoch": 1.216,
      "grad_norm": 0.8080942034721375,
      "learning_rate": 0.00029519999999999997,
      "loss": 1.3787,
      "step": 760
    },
    {
      "epoch": 1.2176,
      "grad_norm": 0.7412335872650146,
      "learning_rate": 0.000294,
      "loss": 1.2793,
      "step": 761
    },
    {
      "epoch": 1.2192,
      "grad_norm": 0.9035298824310303,
      "learning_rate": 0.00029279999999999996,
      "loss": 1.117,
      "step": 762
    },
    {
      "epoch": 1.2208,
      "grad_norm": 1.026508092880249,
      "learning_rate": 0.0002916,
      "loss": 1.0756,
      "step": 763
    },
    {
      "epoch": 1.2224,
      "grad_norm": 1.2814981937408447,
      "learning_rate": 0.00029039999999999996,
      "loss": 1.0611,
      "step": 764
    },
    {
      "epoch": 1.224,
      "grad_norm": 1.475760579109192,
      "learning_rate": 0.0002892,
      "loss": 1.1253,
      "step": 765
    },
    {
      "epoch": 1.2256,
      "grad_norm": 1.2571303844451904,
      "learning_rate": 0.00028799999999999995,
      "loss": 1.585,
      "step": 766
    },
    {
      "epoch": 1.2272,
      "grad_norm": 1.2124806642532349,
      "learning_rate": 0.0002868,
      "loss": 1.3838,
      "step": 767
    },
    {
      "epoch": 1.2288000000000001,
      "grad_norm": 2.581066131591797,
      "learning_rate": 0.00028559999999999995,
      "loss": 1.7297,
      "step": 768
    },
    {
      "epoch": 1.2304,
      "grad_norm": 1.0715489387512207,
      "learning_rate": 0.0002844,
      "loss": 1.0821,
      "step": 769
    },
    {
      "epoch": 1.232,
      "grad_norm": 1.242422342300415,
      "learning_rate": 0.00028319999999999994,
      "loss": 1.5755,
      "step": 770
    },
    {
      "epoch": 1.2336,
      "grad_norm": 1.0124776363372803,
      "learning_rate": 0.00028199999999999997,
      "loss": 1.4446,
      "step": 771
    },
    {
      "epoch": 1.2352,
      "grad_norm": 1.3531243801116943,
      "learning_rate": 0.0002808,
      "loss": 1.1064,
      "step": 772
    },
    {
      "epoch": 1.2368000000000001,
      "grad_norm": 2.615983724594116,
      "learning_rate": 0.00027959999999999997,
      "loss": 1.9305,
      "step": 773
    },
    {
      "epoch": 1.2384,
      "grad_norm": 1.1576447486877441,
      "learning_rate": 0.0002784,
      "loss": 1.2717,
      "step": 774
    },
    {
      "epoch": 1.24,
      "grad_norm": 1.79608154296875,
      "learning_rate": 0.0002772,
      "loss": 1.9539,
      "step": 775
    },
    {
      "epoch": 1.2416,
      "grad_norm": 14.315869331359863,
      "learning_rate": 0.000276,
      "loss": 3.4927,
      "step": 776
    },
    {
      "epoch": 1.2432,
      "grad_norm": 4.236233234405518,
      "learning_rate": 0.0002748,
      "loss": 1.8021,
      "step": 777
    },
    {
      "epoch": 1.2448,
      "grad_norm": 1.4815818071365356,
      "learning_rate": 0.0002736,
      "loss": 1.5814,
      "step": 778
    },
    {
      "epoch": 1.2464,
      "grad_norm": 3.584576368331909,
      "learning_rate": 0.0002724,
      "loss": 2.2016,
      "step": 779
    },
    {
      "epoch": 1.248,
      "grad_norm": 5.444706916809082,
      "learning_rate": 0.0002712,
      "loss": 2.1992,
      "step": 780
    },
    {
      "epoch": 1.2496,
      "grad_norm": 1.9935747385025024,
      "learning_rate": 0.00027,
      "loss": 1.603,
      "step": 781
    },
    {
      "epoch": 1.2511999999999999,
      "grad_norm": 1.7511041164398193,
      "learning_rate": 0.0002688,
      "loss": 1.2715,
      "step": 782
    },
    {
      "epoch": 1.2528000000000001,
      "grad_norm": 3.652242422103882,
      "learning_rate": 0.0002676,
      "loss": 1.2038,
      "step": 783
    },
    {
      "epoch": 1.2544,
      "grad_norm": 4.156230449676514,
      "learning_rate": 0.00026639999999999997,
      "loss": 1.1785,
      "step": 784
    },
    {
      "epoch": 1.256,
      "grad_norm": 3.3570737838745117,
      "learning_rate": 0.0002652,
      "loss": 1.2539,
      "step": 785
    },
    {
      "epoch": 1.2576,
      "grad_norm": 2.836935043334961,
      "learning_rate": 0.00026399999999999997,
      "loss": 1.1734,
      "step": 786
    },
    {
      "epoch": 1.2591999999999999,
      "grad_norm": 2.9423153400421143,
      "learning_rate": 0.0002628,
      "loss": 1.2193,
      "step": 787
    },
    {
      "epoch": 1.2608,
      "grad_norm": 2.4489452838897705,
      "learning_rate": 0.00026159999999999996,
      "loss": 1.291,
      "step": 788
    },
    {
      "epoch": 1.2624,
      "grad_norm": 1.2637799978256226,
      "learning_rate": 0.0002604,
      "loss": 0.8955,
      "step": 789
    },
    {
      "epoch": 1.264,
      "grad_norm": 0.8530015349388123,
      "learning_rate": 0.00025919999999999996,
      "loss": 1.0469,
      "step": 790
    },
    {
      "epoch": 1.2656,
      "grad_norm": 0.9832066297531128,
      "learning_rate": 0.000258,
      "loss": 0.8632,
      "step": 791
    },
    {
      "epoch": 1.2671999999999999,
      "grad_norm": 0.5436220765113831,
      "learning_rate": 0.00025679999999999995,
      "loss": 0.9851,
      "step": 792
    },
    {
      "epoch": 1.2688,
      "grad_norm": 1.093661904335022,
      "learning_rate": 0.0002556,
      "loss": 1.4339,
      "step": 793
    },
    {
      "epoch": 1.2704,
      "grad_norm": 0.7037041187286377,
      "learning_rate": 0.00025439999999999995,
      "loss": 1.1739,
      "step": 794
    },
    {
      "epoch": 1.272,
      "grad_norm": 0.6901881098747253,
      "learning_rate": 0.0002532,
      "loss": 0.9546,
      "step": 795
    },
    {
      "epoch": 1.2736,
      "grad_norm": 1.5446933507919312,
      "learning_rate": 0.00025199999999999995,
      "loss": 1.336,
      "step": 796
    },
    {
      "epoch": 1.2752,
      "grad_norm": 1.7216750383377075,
      "learning_rate": 0.00025079999999999997,
      "loss": 0.8747,
      "step": 797
    },
    {
      "epoch": 1.2768,
      "grad_norm": 1.2335959672927856,
      "learning_rate": 0.00024959999999999994,
      "loss": 0.969,
      "step": 798
    },
    {
      "epoch": 1.2784,
      "grad_norm": 1.483788251876831,
      "learning_rate": 0.00024839999999999997,
      "loss": 0.8357,
      "step": 799
    },
    {
      "epoch": 1.28,
      "grad_norm": 2.874842882156372,
      "learning_rate": 0.0002472,
      "loss": 1.2983,
      "step": 800
    },
    {
      "epoch": 1.2816,
      "grad_norm": 0.7263085246086121,
      "learning_rate": 0.00024599999999999996,
      "loss": 1.2608,
      "step": 801
    },
    {
      "epoch": 1.2832,
      "grad_norm": 1.0308623313903809,
      "learning_rate": 0.0002448,
      "loss": 1.117,
      "step": 802
    },
    {
      "epoch": 1.2848,
      "grad_norm": 0.9009158611297607,
      "learning_rate": 0.00024359999999999999,
      "loss": 1.0626,
      "step": 803
    },
    {
      "epoch": 1.2864,
      "grad_norm": 1.3145204782485962,
      "learning_rate": 0.00024239999999999998,
      "loss": 0.9173,
      "step": 804
    },
    {
      "epoch": 1.288,
      "grad_norm": 1.034488320350647,
      "learning_rate": 0.00024119999999999998,
      "loss": 1.1901,
      "step": 805
    },
    {
      "epoch": 1.2896,
      "grad_norm": 0.8789196610450745,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.6968,
      "step": 806
    },
    {
      "epoch": 1.2912,
      "grad_norm": 0.9626047015190125,
      "learning_rate": 0.0002388,
      "loss": 1.0333,
      "step": 807
    },
    {
      "epoch": 1.2928,
      "grad_norm": 0.7256068587303162,
      "learning_rate": 0.0002376,
      "loss": 0.8452,
      "step": 808
    },
    {
      "epoch": 1.2944,
      "grad_norm": 0.8389018774032593,
      "learning_rate": 0.0002364,
      "loss": 1.2869,
      "step": 809
    },
    {
      "epoch": 1.296,
      "grad_norm": 0.7672526836395264,
      "learning_rate": 0.0002352,
      "loss": 1.1781,
      "step": 810
    },
    {
      "epoch": 1.2976,
      "grad_norm": 0.9246567487716675,
      "learning_rate": 0.000234,
      "loss": 1.2095,
      "step": 811
    },
    {
      "epoch": 1.2992,
      "grad_norm": 0.7383604049682617,
      "learning_rate": 0.0002328,
      "loss": 1.2276,
      "step": 812
    },
    {
      "epoch": 1.3008,
      "grad_norm": 1.0119376182556152,
      "learning_rate": 0.0002316,
      "loss": 1.0281,
      "step": 813
    },
    {
      "epoch": 1.3024,
      "grad_norm": 1.0825129747390747,
      "learning_rate": 0.0002304,
      "loss": 1.3149,
      "step": 814
    },
    {
      "epoch": 1.304,
      "grad_norm": 0.8355119824409485,
      "learning_rate": 0.0002292,
      "loss": 1.2918,
      "step": 815
    },
    {
      "epoch": 1.3056,
      "grad_norm": 0.6260655522346497,
      "learning_rate": 0.00022799999999999999,
      "loss": 0.936,
      "step": 816
    },
    {
      "epoch": 1.3072,
      "grad_norm": 1.1685816049575806,
      "learning_rate": 0.00022679999999999998,
      "loss": 1.2178,
      "step": 817
    },
    {
      "epoch": 1.3088,
      "grad_norm": 1.188957691192627,
      "learning_rate": 0.00022559999999999998,
      "loss": 1.1981,
      "step": 818
    },
    {
      "epoch": 1.3104,
      "grad_norm": 1.8030787706375122,
      "learning_rate": 0.00022439999999999998,
      "loss": 1.247,
      "step": 819
    },
    {
      "epoch": 1.312,
      "grad_norm": 1.1914401054382324,
      "learning_rate": 0.00022319999999999998,
      "loss": 1.0994,
      "step": 820
    },
    {
      "epoch": 1.3136,
      "grad_norm": 1.3656498193740845,
      "learning_rate": 0.00022199999999999998,
      "loss": 1.3633,
      "step": 821
    },
    {
      "epoch": 1.3152,
      "grad_norm": 1.4209223985671997,
      "learning_rate": 0.00022079999999999997,
      "loss": 1.2058,
      "step": 822
    },
    {
      "epoch": 1.3168,
      "grad_norm": 1.3564426898956299,
      "learning_rate": 0.00021959999999999997,
      "loss": 1.6574,
      "step": 823
    },
    {
      "epoch": 1.3184,
      "grad_norm": 2.096992015838623,
      "learning_rate": 0.00021839999999999997,
      "loss": 1.9215,
      "step": 824
    },
    {
      "epoch": 1.32,
      "grad_norm": 1.984554409980774,
      "learning_rate": 0.00021719999999999997,
      "loss": 1.8319,
      "step": 825
    },
    {
      "epoch": 1.3216,
      "grad_norm": 24.8841495513916,
      "learning_rate": 0.00021599999999999996,
      "loss": 4.94,
      "step": 826
    },
    {
      "epoch": 1.3232,
      "grad_norm": 3.756709575653076,
      "learning_rate": 0.00021479999999999996,
      "loss": 1.9233,
      "step": 827
    },
    {
      "epoch": 1.3248,
      "grad_norm": 1.6268775463104248,
      "learning_rate": 0.00021359999999999996,
      "loss": 1.2712,
      "step": 828
    },
    {
      "epoch": 1.3264,
      "grad_norm": 3.466218948364258,
      "learning_rate": 0.00021239999999999996,
      "loss": 1.7201,
      "step": 829
    },
    {
      "epoch": 1.328,
      "grad_norm": 1.621490716934204,
      "learning_rate": 0.00021119999999999996,
      "loss": 1.3709,
      "step": 830
    },
    {
      "epoch": 1.3296000000000001,
      "grad_norm": 1.1984294652938843,
      "learning_rate": 0.00020999999999999998,
      "loss": 1.4106,
      "step": 831
    },
    {
      "epoch": 1.3312,
      "grad_norm": 2.042137384414673,
      "learning_rate": 0.00020879999999999998,
      "loss": 1.1315,
      "step": 832
    },
    {
      "epoch": 1.3328,
      "grad_norm": 1.9946519136428833,
      "learning_rate": 0.00020759999999999998,
      "loss": 1.0351,
      "step": 833
    },
    {
      "epoch": 1.3344,
      "grad_norm": 2.1006603240966797,
      "learning_rate": 0.00020639999999999998,
      "loss": 1.2888,
      "step": 834
    },
    {
      "epoch": 1.336,
      "grad_norm": 1.6302366256713867,
      "learning_rate": 0.0002052,
      "loss": 1.224,
      "step": 835
    },
    {
      "epoch": 1.3376000000000001,
      "grad_norm": 2.526308059692383,
      "learning_rate": 0.000204,
      "loss": 1.0775,
      "step": 836
    },
    {
      "epoch": 1.3392,
      "grad_norm": 2.1584436893463135,
      "learning_rate": 0.0002028,
      "loss": 1.2341,
      "step": 837
    },
    {
      "epoch": 1.3408,
      "grad_norm": 1.6390632390975952,
      "learning_rate": 0.0002016,
      "loss": 1.0818,
      "step": 838
    },
    {
      "epoch": 1.3424,
      "grad_norm": 1.6563626527786255,
      "learning_rate": 0.0002004,
      "loss": 0.9795,
      "step": 839
    },
    {
      "epoch": 1.3439999999999999,
      "grad_norm": 0.9004166126251221,
      "learning_rate": 0.0001992,
      "loss": 1.0898,
      "step": 840
    },
    {
      "epoch": 1.3456000000000001,
      "grad_norm": 0.8950443863868713,
      "learning_rate": 0.000198,
      "loss": 0.9329,
      "step": 841
    },
    {
      "epoch": 1.3472,
      "grad_norm": 0.8782767057418823,
      "learning_rate": 0.00019679999999999999,
      "loss": 0.9925,
      "step": 842
    },
    {
      "epoch": 1.3488,
      "grad_norm": 0.851058304309845,
      "learning_rate": 0.00019559999999999998,
      "loss": 1.0908,
      "step": 843
    },
    {
      "epoch": 1.3504,
      "grad_norm": 5.08306360244751,
      "learning_rate": 0.00019439999999999998,
      "loss": 1.6817,
      "step": 844
    },
    {
      "epoch": 1.3519999999999999,
      "grad_norm": 1.4701097011566162,
      "learning_rate": 0.00019319999999999998,
      "loss": 0.9395,
      "step": 845
    },
    {
      "epoch": 1.3536000000000001,
      "grad_norm": 0.7278363704681396,
      "learning_rate": 0.00019199999999999998,
      "loss": 1.0193,
      "step": 846
    },
    {
      "epoch": 1.3552,
      "grad_norm": 0.9397822618484497,
      "learning_rate": 0.00019079999999999998,
      "loss": 0.7025,
      "step": 847
    },
    {
      "epoch": 1.3568,
      "grad_norm": 1.3745059967041016,
      "learning_rate": 0.00018959999999999997,
      "loss": 0.9564,
      "step": 848
    },
    {
      "epoch": 1.3584,
      "grad_norm": 1.193011999130249,
      "learning_rate": 0.00018839999999999997,
      "loss": 1.1551,
      "step": 849
    },
    {
      "epoch": 1.3599999999999999,
      "grad_norm": 0.889456570148468,
      "learning_rate": 0.0001872,
      "loss": 0.9999,
      "step": 850
    },
    {
      "epoch": 1.3616,
      "grad_norm": 1.0669219493865967,
      "learning_rate": 0.000186,
      "loss": 0.8591,
      "step": 851
    },
    {
      "epoch": 1.3632,
      "grad_norm": 0.9589920043945312,
      "learning_rate": 0.0001848,
      "loss": 0.9345,
      "step": 852
    },
    {
      "epoch": 1.3648,
      "grad_norm": 0.7081511616706848,
      "learning_rate": 0.0001836,
      "loss": 0.7226,
      "step": 853
    },
    {
      "epoch": 1.3664,
      "grad_norm": 0.7487833499908447,
      "learning_rate": 0.0001824,
      "loss": 1.0986,
      "step": 854
    },
    {
      "epoch": 1.3679999999999999,
      "grad_norm": 0.9651376605033875,
      "learning_rate": 0.00018119999999999999,
      "loss": 0.8951,
      "step": 855
    },
    {
      "epoch": 1.3696,
      "grad_norm": 0.9545527100563049,
      "learning_rate": 0.00017999999999999998,
      "loss": 1.0129,
      "step": 856
    },
    {
      "epoch": 1.3712,
      "grad_norm": 1.016617774963379,
      "learning_rate": 0.00017879999999999998,
      "loss": 1.0022,
      "step": 857
    },
    {
      "epoch": 1.3728,
      "grad_norm": 1.4556984901428223,
      "learning_rate": 0.00017759999999999998,
      "loss": 1.317,
      "step": 858
    },
    {
      "epoch": 1.3744,
      "grad_norm": 0.790810763835907,
      "learning_rate": 0.00017639999999999998,
      "loss": 1.2533,
      "step": 859
    },
    {
      "epoch": 1.376,
      "grad_norm": 0.9381358027458191,
      "learning_rate": 0.00017519999999999998,
      "loss": 0.9896,
      "step": 860
    },
    {
      "epoch": 1.3776,
      "grad_norm": 0.8054640889167786,
      "learning_rate": 0.00017399999999999997,
      "loss": 0.9845,
      "step": 861
    },
    {
      "epoch": 1.3792,
      "grad_norm": 0.8973929286003113,
      "learning_rate": 0.00017279999999999997,
      "loss": 1.4839,
      "step": 862
    },
    {
      "epoch": 1.3808,
      "grad_norm": 1.5383504629135132,
      "learning_rate": 0.00017159999999999997,
      "loss": 1.0332,
      "step": 863
    },
    {
      "epoch": 1.3824,
      "grad_norm": 0.842438817024231,
      "learning_rate": 0.00017039999999999997,
      "loss": 1.3733,
      "step": 864
    },
    {
      "epoch": 1.384,
      "grad_norm": 1.3562208414077759,
      "learning_rate": 0.00016919999999999997,
      "loss": 1.3423,
      "step": 865
    },
    {
      "epoch": 1.3856,
      "grad_norm": 0.8287788033485413,
      "learning_rate": 0.000168,
      "loss": 1.0022,
      "step": 866
    },
    {
      "epoch": 1.3872,
      "grad_norm": 1.4219720363616943,
      "learning_rate": 0.0001668,
      "loss": 1.2173,
      "step": 867
    },
    {
      "epoch": 1.3888,
      "grad_norm": 1.0144832134246826,
      "learning_rate": 0.0001656,
      "loss": 1.3067,
      "step": 868
    },
    {
      "epoch": 1.3904,
      "grad_norm": 1.1086362600326538,
      "learning_rate": 0.0001644,
      "loss": 1.4451,
      "step": 869
    },
    {
      "epoch": 1.392,
      "grad_norm": 1.1874333620071411,
      "learning_rate": 0.0001632,
      "loss": 1.4588,
      "step": 870
    },
    {
      "epoch": 1.3936,
      "grad_norm": 1.2281416654586792,
      "learning_rate": 0.000162,
      "loss": 1.564,
      "step": 871
    },
    {
      "epoch": 1.3952,
      "grad_norm": 1.024274468421936,
      "learning_rate": 0.0001608,
      "loss": 1.3052,
      "step": 872
    },
    {
      "epoch": 1.3968,
      "grad_norm": 1.0952630043029785,
      "learning_rate": 0.0001596,
      "loss": 1.4589,
      "step": 873
    },
    {
      "epoch": 1.3984,
      "grad_norm": 1.3578640222549438,
      "learning_rate": 0.0001584,
      "loss": 1.8693,
      "step": 874
    },
    {
      "epoch": 1.4,
      "grad_norm": 1.3908575773239136,
      "learning_rate": 0.0001572,
      "loss": 2.0153,
      "step": 875
    },
    {
      "epoch": 1.4016,
      "grad_norm": 3.5373663902282715,
      "learning_rate": 0.000156,
      "loss": 1.6021,
      "step": 876
    },
    {
      "epoch": 1.4032,
      "grad_norm": 4.33758020401001,
      "learning_rate": 0.0001548,
      "loss": 1.6701,
      "step": 877
    },
    {
      "epoch": 1.4048,
      "grad_norm": 0.7620241641998291,
      "learning_rate": 0.0001536,
      "loss": 1.1137,
      "step": 878
    },
    {
      "epoch": 1.4064,
      "grad_norm": 3.931720733642578,
      "learning_rate": 0.0001524,
      "loss": 1.7832,
      "step": 879
    },
    {
      "epoch": 1.408,
      "grad_norm": 4.786645412445068,
      "learning_rate": 0.0001512,
      "loss": 1.9583,
      "step": 880
    },
    {
      "epoch": 1.4096,
      "grad_norm": 1.036521553993225,
      "learning_rate": 0.00015,
      "loss": 1.0819,
      "step": 881
    },
    {
      "epoch": 1.4112,
      "grad_norm": 4.162290096282959,
      "learning_rate": 0.00014879999999999998,
      "loss": 2.0566,
      "step": 882
    },
    {
      "epoch": 1.4128,
      "grad_norm": 2.359104633331299,
      "learning_rate": 0.00014759999999999998,
      "loss": 1.6004,
      "step": 883
    },
    {
      "epoch": 1.4144,
      "grad_norm": 1.0020617246627808,
      "learning_rate": 0.00014639999999999998,
      "loss": 1.3709,
      "step": 884
    },
    {
      "epoch": 1.416,
      "grad_norm": 1.8424478769302368,
      "learning_rate": 0.00014519999999999998,
      "loss": 1.2331,
      "step": 885
    },
    {
      "epoch": 1.4176,
      "grad_norm": 1.1963813304901123,
      "learning_rate": 0.00014399999999999998,
      "loss": 1.0713,
      "step": 886
    },
    {
      "epoch": 1.4192,
      "grad_norm": 1.5633505582809448,
      "learning_rate": 0.00014279999999999997,
      "loss": 1.0691,
      "step": 887
    },
    {
      "epoch": 1.4208,
      "grad_norm": 1.1638294458389282,
      "learning_rate": 0.00014159999999999997,
      "loss": 0.9451,
      "step": 888
    },
    {
      "epoch": 1.4224,
      "grad_norm": 1.7012614011764526,
      "learning_rate": 0.0001404,
      "loss": 0.8284,
      "step": 889
    },
    {
      "epoch": 1.424,
      "grad_norm": 1.2200738191604614,
      "learning_rate": 0.0001392,
      "loss": 0.8185,
      "step": 890
    },
    {
      "epoch": 1.4256,
      "grad_norm": 1.1766223907470703,
      "learning_rate": 0.000138,
      "loss": 0.9226,
      "step": 891
    },
    {
      "epoch": 1.4272,
      "grad_norm": 0.9217773675918579,
      "learning_rate": 0.0001368,
      "loss": 0.8685,
      "step": 892
    },
    {
      "epoch": 1.4288,
      "grad_norm": 0.8851170539855957,
      "learning_rate": 0.0001356,
      "loss": 0.7968,
      "step": 893
    },
    {
      "epoch": 1.4304000000000001,
      "grad_norm": 0.8986585140228271,
      "learning_rate": 0.0001344,
      "loss": 0.9926,
      "step": 894
    },
    {
      "epoch": 1.432,
      "grad_norm": 0.760133683681488,
      "learning_rate": 0.00013319999999999999,
      "loss": 0.761,
      "step": 895
    },
    {
      "epoch": 1.4336,
      "grad_norm": 1.3854573965072632,
      "learning_rate": 0.00013199999999999998,
      "loss": 1.3453,
      "step": 896
    },
    {
      "epoch": 1.4352,
      "grad_norm": 2.538731098175049,
      "learning_rate": 0.00013079999999999998,
      "loss": 1.405,
      "step": 897
    },
    {
      "epoch": 1.4368,
      "grad_norm": 0.9111962914466858,
      "learning_rate": 0.00012959999999999998,
      "loss": 1.0013,
      "step": 898
    },
    {
      "epoch": 1.4384000000000001,
      "grad_norm": 0.9045060873031616,
      "learning_rate": 0.00012839999999999998,
      "loss": 1.0312,
      "step": 899
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.8330265879631042,
      "learning_rate": 0.00012719999999999997,
      "loss": 1.1871,
      "step": 900
    },
    {
      "epoch": 1.4416,
      "grad_norm": 0.8071021437644958,
      "learning_rate": 0.00012599999999999997,
      "loss": 1.4302,
      "step": 901
    },
    {
      "epoch": 1.4432,
      "grad_norm": 1.5638083219528198,
      "learning_rate": 0.00012479999999999997,
      "loss": 1.2347,
      "step": 902
    },
    {
      "epoch": 1.4447999999999999,
      "grad_norm": 1.674561619758606,
      "learning_rate": 0.0001236,
      "loss": 1.197,
      "step": 903
    },
    {
      "epoch": 1.4464000000000001,
      "grad_norm": 1.1858874559402466,
      "learning_rate": 0.0001224,
      "loss": 0.8886,
      "step": 904
    },
    {
      "epoch": 1.448,
      "grad_norm": 0.8216437101364136,
      "learning_rate": 0.00012119999999999999,
      "loss": 0.8166,
      "step": 905
    },
    {
      "epoch": 1.4496,
      "grad_norm": 0.7633309364318848,
      "learning_rate": 0.00011999999999999999,
      "loss": 1.0883,
      "step": 906
    },
    {
      "epoch": 1.4512,
      "grad_norm": 0.7902513742446899,
      "learning_rate": 0.0001188,
      "loss": 1.1526,
      "step": 907
    },
    {
      "epoch": 1.4527999999999999,
      "grad_norm": 0.6996155381202698,
      "learning_rate": 0.0001176,
      "loss": 0.8232,
      "step": 908
    },
    {
      "epoch": 1.4544000000000001,
      "grad_norm": 1.0348126888275146,
      "learning_rate": 0.0001164,
      "loss": 1.5505,
      "step": 909
    },
    {
      "epoch": 1.456,
      "grad_norm": 0.9336599707603455,
      "learning_rate": 0.0001152,
      "loss": 1.1751,
      "step": 910
    },
    {
      "epoch": 1.4576,
      "grad_norm": 0.8460116386413574,
      "learning_rate": 0.00011399999999999999,
      "loss": 1.1737,
      "step": 911
    },
    {
      "epoch": 1.4592,
      "grad_norm": 0.8113738894462585,
      "learning_rate": 0.00011279999999999999,
      "loss": 1.0484,
      "step": 912
    },
    {
      "epoch": 1.4607999999999999,
      "grad_norm": 0.8300177454948425,
      "learning_rate": 0.00011159999999999999,
      "loss": 0.7955,
      "step": 913
    },
    {
      "epoch": 1.4624,
      "grad_norm": 1.2062090635299683,
      "learning_rate": 0.00011039999999999999,
      "loss": 1.3775,
      "step": 914
    },
    {
      "epoch": 1.464,
      "grad_norm": 1.4255359172821045,
      "learning_rate": 0.00010919999999999998,
      "loss": 1.3335,
      "step": 915
    },
    {
      "epoch": 1.4656,
      "grad_norm": 1.0286868810653687,
      "learning_rate": 0.00010799999999999998,
      "loss": 1.1146,
      "step": 916
    },
    {
      "epoch": 1.4672,
      "grad_norm": 1.0137372016906738,
      "learning_rate": 0.00010679999999999998,
      "loss": 1.0296,
      "step": 917
    },
    {
      "epoch": 1.4687999999999999,
      "grad_norm": 0.9231035113334656,
      "learning_rate": 0.00010559999999999998,
      "loss": 1.0707,
      "step": 918
    },
    {
      "epoch": 1.4704,
      "grad_norm": 0.7874758839607239,
      "learning_rate": 0.00010439999999999999,
      "loss": 0.8588,
      "step": 919
    },
    {
      "epoch": 1.472,
      "grad_norm": 0.9133099913597107,
      "learning_rate": 0.00010319999999999999,
      "loss": 1.3255,
      "step": 920
    },
    {
      "epoch": 1.4736,
      "grad_norm": 1.030543565750122,
      "learning_rate": 0.000102,
      "loss": 1.5101,
      "step": 921
    },
    {
      "epoch": 1.4752,
      "grad_norm": 0.9055063724517822,
      "learning_rate": 0.0001008,
      "loss": 0.9935,
      "step": 922
    },
    {
      "epoch": 1.4768,
      "grad_norm": 1.4548567533493042,
      "learning_rate": 9.96e-05,
      "loss": 1.5182,
      "step": 923
    },
    {
      "epoch": 1.4784,
      "grad_norm": 1.2355753183364868,
      "learning_rate": 9.839999999999999e-05,
      "loss": 1.5844,
      "step": 924
    },
    {
      "epoch": 1.48,
      "grad_norm": 2.3780314922332764,
      "learning_rate": 9.719999999999999e-05,
      "loss": 1.9339,
      "step": 925
    },
    {
      "epoch": 1.4816,
      "grad_norm": 1.8571350574493408,
      "learning_rate": 9.599999999999999e-05,
      "loss": 1.1195,
      "step": 926
    },
    {
      "epoch": 1.4832,
      "grad_norm": 3.069366455078125,
      "learning_rate": 9.479999999999999e-05,
      "loss": 1.4374,
      "step": 927
    },
    {
      "epoch": 1.4848,
      "grad_norm": 1.4385106563568115,
      "learning_rate": 9.36e-05,
      "loss": 1.0552,
      "step": 928
    },
    {
      "epoch": 1.4864,
      "grad_norm": 2.345099449157715,
      "learning_rate": 9.24e-05,
      "loss": 1.7314,
      "step": 929
    },
    {
      "epoch": 1.488,
      "grad_norm": 0.7494111657142639,
      "learning_rate": 9.12e-05,
      "loss": 0.8883,
      "step": 930
    },
    {
      "epoch": 1.4896,
      "grad_norm": 0.8890445232391357,
      "learning_rate": 8.999999999999999e-05,
      "loss": 0.9971,
      "step": 931
    },
    {
      "epoch": 1.4912,
      "grad_norm": 0.8567062616348267,
      "learning_rate": 8.879999999999999e-05,
      "loss": 0.9844,
      "step": 932
    },
    {
      "epoch": 1.4928,
      "grad_norm": 2.1418657302856445,
      "learning_rate": 8.759999999999999e-05,
      "loss": 1.2059,
      "step": 933
    },
    {
      "epoch": 1.4944,
      "grad_norm": 0.9829572439193726,
      "learning_rate": 8.639999999999999e-05,
      "loss": 1.2171,
      "step": 934
    },
    {
      "epoch": 1.496,
      "grad_norm": 0.8156057000160217,
      "learning_rate": 8.519999999999998e-05,
      "loss": 0.9593,
      "step": 935
    },
    {
      "epoch": 1.4976,
      "grad_norm": 1.4363347291946411,
      "learning_rate": 8.4e-05,
      "loss": 0.9601,
      "step": 936
    },
    {
      "epoch": 1.4992,
      "grad_norm": 1.0625683069229126,
      "learning_rate": 8.28e-05,
      "loss": 1.0812,
      "step": 937
    },
    {
      "epoch": 1.5008,
      "grad_norm": 0.9699941873550415,
      "learning_rate": 8.16e-05,
      "loss": 0.9473,
      "step": 938
    },
    {
      "epoch": 1.5024,
      "grad_norm": 0.772868812084198,
      "learning_rate": 8.04e-05,
      "loss": 0.7998,
      "step": 939
    },
    {
      "epoch": 1.504,
      "grad_norm": 0.9699268341064453,
      "learning_rate": 7.92e-05,
      "loss": 0.9162,
      "step": 940
    },
    {
      "epoch": 1.5056,
      "grad_norm": 0.9714852571487427,
      "learning_rate": 7.8e-05,
      "loss": 0.7546,
      "step": 941
    },
    {
      "epoch": 1.5072,
      "grad_norm": 0.7131572365760803,
      "learning_rate": 7.68e-05,
      "loss": 0.9188,
      "step": 942
    },
    {
      "epoch": 1.5088,
      "grad_norm": 0.855786919593811,
      "learning_rate": 7.56e-05,
      "loss": 1.0208,
      "step": 943
    },
    {
      "epoch": 1.5104,
      "grad_norm": 0.8761354684829712,
      "learning_rate": 7.439999999999999e-05,
      "loss": 0.9603,
      "step": 944
    },
    {
      "epoch": 1.512,
      "grad_norm": 0.8056983351707458,
      "learning_rate": 7.319999999999999e-05,
      "loss": 1.0588,
      "step": 945
    },
    {
      "epoch": 1.5135999999999998,
      "grad_norm": 1.0384901762008667,
      "learning_rate": 7.199999999999999e-05,
      "loss": 0.8915,
      "step": 946
    },
    {
      "epoch": 1.5152,
      "grad_norm": 0.6739591956138611,
      "learning_rate": 7.079999999999999e-05,
      "loss": 0.7157,
      "step": 947
    },
    {
      "epoch": 1.5168,
      "grad_norm": 0.7567741274833679,
      "learning_rate": 6.96e-05,
      "loss": 0.733,
      "step": 948
    },
    {
      "epoch": 1.5184,
      "grad_norm": 1.2941476106643677,
      "learning_rate": 6.84e-05,
      "loss": 1.2629,
      "step": 949
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.7849064469337463,
      "learning_rate": 6.72e-05,
      "loss": 1.0687,
      "step": 950
    },
    {
      "epoch": 1.5215999999999998,
      "grad_norm": 0.7167961001396179,
      "learning_rate": 6.599999999999999e-05,
      "loss": 0.9125,
      "step": 951
    },
    {
      "epoch": 1.5232,
      "grad_norm": 1.4623581171035767,
      "learning_rate": 6.479999999999999e-05,
      "loss": 1.0419,
      "step": 952
    },
    {
      "epoch": 1.5248,
      "grad_norm": 1.0260752439498901,
      "learning_rate": 6.359999999999999e-05,
      "loss": 0.759,
      "step": 953
    },
    {
      "epoch": 1.5264,
      "grad_norm": 0.6854276657104492,
      "learning_rate": 6.239999999999999e-05,
      "loss": 1.1125,
      "step": 954
    },
    {
      "epoch": 1.528,
      "grad_norm": 0.776077151298523,
      "learning_rate": 6.12e-05,
      "loss": 0.7962,
      "step": 955
    },
    {
      "epoch": 1.5295999999999998,
      "grad_norm": 0.7898908257484436,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 1.0421,
      "step": 956
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.7632326483726501,
      "learning_rate": 5.88e-05,
      "loss": 0.8491,
      "step": 957
    },
    {
      "epoch": 1.5328,
      "grad_norm": 0.7451068162918091,
      "learning_rate": 5.76e-05,
      "loss": 0.9336,
      "step": 958
    },
    {
      "epoch": 1.5344,
      "grad_norm": 0.7700297832489014,
      "learning_rate": 5.6399999999999995e-05,
      "loss": 0.9,
      "step": 959
    },
    {
      "epoch": 1.536,
      "grad_norm": 0.879389762878418,
      "learning_rate": 5.519999999999999e-05,
      "loss": 1.1537,
      "step": 960
    },
    {
      "epoch": 1.5375999999999999,
      "grad_norm": 0.9828069806098938,
      "learning_rate": 5.399999999999999e-05,
      "loss": 0.9058,
      "step": 961
    },
    {
      "epoch": 1.5392000000000001,
      "grad_norm": 0.9066524505615234,
      "learning_rate": 5.279999999999999e-05,
      "loss": 1.1331,
      "step": 962
    },
    {
      "epoch": 1.5408,
      "grad_norm": 0.7387616038322449,
      "learning_rate": 5.1599999999999994e-05,
      "loss": 0.9451,
      "step": 963
    },
    {
      "epoch": 1.5424,
      "grad_norm": 1.3451749086380005,
      "learning_rate": 5.04e-05,
      "loss": 1.1639,
      "step": 964
    },
    {
      "epoch": 1.544,
      "grad_norm": 1.0739102363586426,
      "learning_rate": 4.9199999999999997e-05,
      "loss": 1.4125,
      "step": 965
    },
    {
      "epoch": 1.5455999999999999,
      "grad_norm": 0.9140891432762146,
      "learning_rate": 4.7999999999999994e-05,
      "loss": 1.1883,
      "step": 966
    },
    {
      "epoch": 1.5472000000000001,
      "grad_norm": 1.1968399286270142,
      "learning_rate": 4.68e-05,
      "loss": 1.1692,
      "step": 967
    },
    {
      "epoch": 1.5488,
      "grad_norm": 0.9506883025169373,
      "learning_rate": 4.56e-05,
      "loss": 0.9939,
      "step": 968
    },
    {
      "epoch": 1.5504,
      "grad_norm": 1.3204574584960938,
      "learning_rate": 4.4399999999999995e-05,
      "loss": 0.9103,
      "step": 969
    },
    {
      "epoch": 1.552,
      "grad_norm": 1.3693453073501587,
      "learning_rate": 4.319999999999999e-05,
      "loss": 1.279,
      "step": 970
    },
    {
      "epoch": 1.5535999999999999,
      "grad_norm": 1.286435842514038,
      "learning_rate": 4.2e-05,
      "loss": 1.2637,
      "step": 971
    },
    {
      "epoch": 1.5552000000000001,
      "grad_norm": 1.6549150943756104,
      "learning_rate": 4.08e-05,
      "loss": 1.5989,
      "step": 972
    },
    {
      "epoch": 1.5568,
      "grad_norm": 1.1446410417556763,
      "learning_rate": 3.96e-05,
      "loss": 1.5587,
      "step": 973
    },
    {
      "epoch": 1.5584,
      "grad_norm": 1.1329030990600586,
      "learning_rate": 3.84e-05,
      "loss": 1.5915,
      "step": 974
    },
    {
      "epoch": 1.56,
      "grad_norm": 6.757206916809082,
      "learning_rate": 3.7199999999999996e-05,
      "loss": 3.078,
      "step": 975
    },
    {
      "epoch": 1.5615999999999999,
      "grad_norm": 3.0250446796417236,
      "learning_rate": 3.5999999999999994e-05,
      "loss": 1.5766,
      "step": 976
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 1.796055793762207,
      "learning_rate": 3.48e-05,
      "loss": 1.4036,
      "step": 977
    },
    {
      "epoch": 1.5648,
      "grad_norm": 2.8835206031799316,
      "learning_rate": 3.36e-05,
      "loss": 1.4255,
      "step": 978
    },
    {
      "epoch": 1.5664,
      "grad_norm": 3.2792022228240967,
      "learning_rate": 3.2399999999999995e-05,
      "loss": 1.5776,
      "step": 979
    },
    {
      "epoch": 1.568,
      "grad_norm": 0.9129714369773865,
      "learning_rate": 3.119999999999999e-05,
      "loss": 1.1509,
      "step": 980
    },
    {
      "epoch": 1.5695999999999999,
      "grad_norm": 1.325767993927002,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 1.3454,
      "step": 981
    },
    {
      "epoch": 1.5712000000000002,
      "grad_norm": 0.8600112795829773,
      "learning_rate": 2.88e-05,
      "loss": 0.8939,
      "step": 982
    },
    {
      "epoch": 1.5728,
      "grad_norm": 0.6780167818069458,
      "learning_rate": 2.7599999999999997e-05,
      "loss": 0.7801,
      "step": 983
    },
    {
      "epoch": 1.5744,
      "grad_norm": 1.1986228227615356,
      "learning_rate": 2.6399999999999995e-05,
      "loss": 1.0117,
      "step": 984
    },
    {
      "epoch": 1.576,
      "grad_norm": 1.3320322036743164,
      "learning_rate": 2.52e-05,
      "loss": 1.0323,
      "step": 985
    },
    {
      "epoch": 1.5776,
      "grad_norm": 0.6832807064056396,
      "learning_rate": 2.3999999999999997e-05,
      "loss": 1.1191,
      "step": 986
    },
    {
      "epoch": 1.5792000000000002,
      "grad_norm": 3.410607099533081,
      "learning_rate": 2.28e-05,
      "loss": 1.7286,
      "step": 987
    },
    {
      "epoch": 1.5808,
      "grad_norm": 0.7802388668060303,
      "learning_rate": 2.1599999999999996e-05,
      "loss": 0.7245,
      "step": 988
    },
    {
      "epoch": 1.5824,
      "grad_norm": 0.8889452815055847,
      "learning_rate": 2.04e-05,
      "loss": 0.8384,
      "step": 989
    },
    {
      "epoch": 1.584,
      "grad_norm": 0.7941523194313049,
      "learning_rate": 1.92e-05,
      "loss": 1.0049,
      "step": 990
    },
    {
      "epoch": 1.5856,
      "grad_norm": 0.8905624747276306,
      "learning_rate": 1.7999999999999997e-05,
      "loss": 0.7337,
      "step": 991
    },
    {
      "epoch": 1.5872000000000002,
      "grad_norm": 1.1984586715698242,
      "learning_rate": 1.68e-05,
      "loss": 0.8216,
      "step": 992
    },
    {
      "epoch": 1.5888,
      "grad_norm": 0.8467238545417786,
      "learning_rate": 1.5599999999999996e-05,
      "loss": 0.807,
      "step": 993
    },
    {
      "epoch": 1.5904,
      "grad_norm": 1.230843424797058,
      "learning_rate": 1.44e-05,
      "loss": 1.1562,
      "step": 994
    },
    {
      "epoch": 1.592,
      "grad_norm": 1.0017879009246826,
      "learning_rate": 1.3199999999999997e-05,
      "loss": 0.971,
      "step": 995
    },
    {
      "epoch": 1.5936,
      "grad_norm": 1.124283790588379,
      "learning_rate": 1.1999999999999999e-05,
      "loss": 0.8834,
      "step": 996
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.9626112580299377,
      "learning_rate": 1.0799999999999998e-05,
      "loss": 1.0106,
      "step": 997
    },
    {
      "epoch": 1.5968,
      "grad_norm": 1.2808488607406616,
      "learning_rate": 9.6e-06,
      "loss": 0.9121,
      "step": 998
    },
    {
      "epoch": 1.5984,
      "grad_norm": 0.904681921005249,
      "learning_rate": 8.4e-06,
      "loss": 0.8199,
      "step": 999
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.8806717395782471,
      "learning_rate": 7.2e-06,
      "loss": 0.972,
      "step": 1000
    },
    {
      "epoch": 1.6,
      "eval_cer": 0.3090127193119299,
      "eval_loss": 1.500978946685791,
      "eval_runtime": 162.132,
      "eval_samples_per_second": 19.342,
      "eval_steps_per_second": 1.209,
      "eval_wer": 0.4351051665913261,
      "step": 1000
    },
    {
      "epoch": 1.6,
      "step": 1000,
      "total_flos": 6.212261523683712e+18,
      "train_loss": 3.1903547011613846,
      "train_runtime": 2112.9078,
      "train_samples_per_second": 15.145,
      "train_steps_per_second": 0.473
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.212261523683712e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}