{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 622,
  "global_step": 1036,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 1.2006484270095825,
      "learning_rate": 2.5e-06,
      "loss": 5.0998,
      "step": 1
    },
    {
      "epoch": 0.0,
      "eval_loss": 5.15006685256958,
      "eval_runtime": 28.119,
      "eval_samples_per_second": 3.023,
      "eval_steps_per_second": 1.529,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.6314904689788818,
      "learning_rate": 5e-06,
      "loss": 5.6374,
      "step": 2
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.3538857698440552,
      "learning_rate": 7.5e-06,
      "loss": 5.0656,
      "step": 3
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.9769891500473022,
      "learning_rate": 1e-05,
      "loss": 5.1906,
      "step": 4
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.4604706764221191,
      "learning_rate": 1.25e-05,
      "loss": 5.1107,
      "step": 5
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.2502201795578003,
      "learning_rate": 1.5e-05,
      "loss": 4.8447,
      "step": 6
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.1797655820846558,
      "learning_rate": 1.7500000000000002e-05,
      "loss": 5.0042,
      "step": 7
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.3177151679992676,
      "learning_rate": 2e-05,
      "loss": 4.878,
      "step": 8
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.1382358074188232,
      "learning_rate": 2.2499999999999998e-05,
      "loss": 5.106,
      "step": 9
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.1523857116699219,
      "learning_rate": 2.5e-05,
      "loss": 5.241,
      "step": 10
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.994879961013794,
      "learning_rate": 2.75e-05,
      "loss": 4.4475,
      "step": 11
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.0009862184524536,
      "learning_rate": 3e-05,
      "loss": 4.615,
      "step": 12
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.1249079704284668,
      "learning_rate": 3.2500000000000004e-05,
      "loss": 4.3469,
      "step": 13
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.1036263704299927,
      "learning_rate": 3.5000000000000004e-05,
      "loss": 4.7748,
      "step": 14
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.1240339279174805,
      "learning_rate": 3.75e-05,
      "loss": 4.8696,
      "step": 15
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.9816920757293701,
      "learning_rate": 4e-05,
      "loss": 4.5595,
      "step": 16
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.1084593534469604,
      "learning_rate": 4.25e-05,
      "loss": 4.2922,
      "step": 17
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.997968316078186,
      "learning_rate": 4.4999999999999996e-05,
      "loss": 4.3057,
      "step": 18
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.0401185750961304,
      "learning_rate": 4.75e-05,
      "loss": 4.364,
      "step": 19
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.0804249048233032,
      "learning_rate": 5e-05,
      "loss": 4.8663,
      "step": 20
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.2384392023086548,
      "learning_rate": 5.25e-05,
      "loss": 4.2992,
      "step": 21
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.231845736503601,
      "learning_rate": 5.5e-05,
      "loss": 4.8637,
      "step": 22
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.242187261581421,
      "learning_rate": 5.75e-05,
      "loss": 3.865,
      "step": 23
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.2873363494873047,
      "learning_rate": 6e-05,
      "loss": 3.8585,
      "step": 24
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.1170119047164917,
      "learning_rate": 6.25e-05,
      "loss": 3.8948,
      "step": 25
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.1938223838806152,
      "learning_rate": 6.500000000000001e-05,
      "loss": 3.8009,
      "step": 26
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.3068759441375732,
      "learning_rate": 6.75e-05,
      "loss": 3.6811,
      "step": 27
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.280858039855957,
      "learning_rate": 7.000000000000001e-05,
      "loss": 3.9975,
      "step": 28
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.3869880437850952,
      "learning_rate": 7.25e-05,
      "loss": 3.5562,
      "step": 29
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.3908005952835083,
      "learning_rate": 7.5e-05,
      "loss": 3.296,
      "step": 30
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.622360348701477,
      "learning_rate": 7.75e-05,
      "loss": 3.8233,
      "step": 31
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.5797035694122314,
      "learning_rate": 8e-05,
      "loss": 3.6274,
      "step": 32
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.816627025604248,
      "learning_rate": 8.25e-05,
      "loss": 3.3872,
      "step": 33
    },
    {
      "epoch": 0.03,
      "grad_norm": 2.1191844940185547,
      "learning_rate": 8.5e-05,
      "loss": 3.5982,
      "step": 34
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.9559025764465332,
      "learning_rate": 8.75e-05,
      "loss": 3.5972,
      "step": 35
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.6593563556671143,
      "learning_rate": 8.999999999999999e-05,
      "loss": 3.3982,
      "step": 36
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.0360360145568848,
      "learning_rate": 9.25e-05,
      "loss": 3.5185,
      "step": 37
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.9218908548355103,
      "learning_rate": 9.5e-05,
      "loss": 3.5998,
      "step": 38
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.6969695091247559,
      "learning_rate": 9.750000000000001e-05,
      "loss": 3.2894,
      "step": 39
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.6107378005981445,
      "learning_rate": 0.0001,
      "loss": 3.3074,
      "step": 40
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.6574798822402954,
      "learning_rate": 0.0001025,
      "loss": 3.4148,
      "step": 41
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.8693658113479614,
      "learning_rate": 0.000105,
      "loss": 3.4726,
      "step": 42
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.4956166744232178,
      "learning_rate": 0.0001075,
      "loss": 2.9462,
      "step": 43
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.6204662322998047,
      "learning_rate": 0.00011,
      "loss": 3.1303,
      "step": 44
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.8228565454483032,
      "learning_rate": 0.00011250000000000001,
      "loss": 3.9787,
      "step": 45
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.6097328662872314,
      "learning_rate": 0.000115,
      "loss": 2.9556,
      "step": 46
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.843204140663147,
      "learning_rate": 0.0001175,
      "loss": 3.0469,
      "step": 47
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.8468220233917236,
      "learning_rate": 0.00012,
      "loss": 3.1549,
      "step": 48
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.8635084629058838,
      "learning_rate": 0.0001225,
      "loss": 3.2309,
      "step": 49
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.9674320220947266,
      "learning_rate": 0.000125,
      "loss": 3.264,
      "step": 50
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.514456868171692,
      "learning_rate": 0.0001275,
      "loss": 2.895,
      "step": 51
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.8026971817016602,
      "learning_rate": 0.00013000000000000002,
      "loss": 2.9692,
      "step": 52
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.872318983078003,
      "learning_rate": 0.00013250000000000002,
      "loss": 3.7424,
      "step": 53
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.7249802350997925,
      "learning_rate": 0.000135,
      "loss": 2.941,
      "step": 54
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.677011489868164,
      "learning_rate": 0.0001375,
      "loss": 3.3312,
      "step": 55
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.9033194780349731,
      "learning_rate": 0.00014000000000000001,
      "loss": 2.5066,
      "step": 56
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.9987455606460571,
      "learning_rate": 0.0001425,
      "loss": 3.4693,
      "step": 57
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.9788786172866821,
      "learning_rate": 0.000145,
      "loss": 2.8805,
      "step": 58
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.7737230062484741,
      "learning_rate": 0.0001475,
      "loss": 2.8758,
      "step": 59
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.7360577583312988,
      "learning_rate": 0.00015,
      "loss": 3.4,
      "step": 60
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.982911229133606,
      "learning_rate": 0.0001525,
      "loss": 2.779,
      "step": 61
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.9741348028182983,
      "learning_rate": 0.000155,
      "loss": 3.3381,
      "step": 62
    },
    {
      "epoch": 0.06,
      "grad_norm": 2.060384750366211,
      "learning_rate": 0.0001575,
      "loss": 3.6285,
      "step": 63
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.6591861248016357,
      "learning_rate": 0.00016,
      "loss": 3.1691,
      "step": 64
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.8308738470077515,
      "learning_rate": 0.00016250000000000002,
      "loss": 3.3354,
      "step": 65
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.8891240358352661,
      "learning_rate": 0.000165,
      "loss": 3.2649,
      "step": 66
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.697945237159729,
      "learning_rate": 0.0001675,
      "loss": 2.8464,
      "step": 67
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.6797360181808472,
      "learning_rate": 0.00017,
      "loss": 3.3645,
      "step": 68
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.964936375617981,
      "learning_rate": 0.0001725,
      "loss": 2.6052,
      "step": 69
    },
    {
      "epoch": 0.07,
      "grad_norm": 2.081937313079834,
      "learning_rate": 0.000175,
      "loss": 3.3901,
      "step": 70
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.8318378925323486,
      "learning_rate": 0.0001775,
      "loss": 3.2655,
      "step": 71
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.757249116897583,
      "learning_rate": 0.00017999999999999998,
      "loss": 3.254,
      "step": 72
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.8252121210098267,
      "learning_rate": 0.0001825,
      "loss": 2.5852,
      "step": 73
    },
    {
      "epoch": 0.07,
      "grad_norm": 2.0867161750793457,
      "learning_rate": 0.000185,
      "loss": 2.8474,
      "step": 74
    },
    {
      "epoch": 0.07,
      "grad_norm": 2.1990573406219482,
      "learning_rate": 0.0001875,
      "loss": 2.5813,
      "step": 75
    },
    {
      "epoch": 0.07,
      "grad_norm": 2.1143388748168945,
      "learning_rate": 0.00019,
      "loss": 3.1759,
      "step": 76
    },
    {
      "epoch": 0.07,
      "grad_norm": 2.1437313556671143,
      "learning_rate": 0.00019250000000000002,
      "loss": 2.7499,
      "step": 77
    },
    {
      "epoch": 0.08,
      "grad_norm": 2.0984110832214355,
      "learning_rate": 0.00019500000000000002,
      "loss": 2.6248,
      "step": 78
    },
    {
      "epoch": 0.08,
      "grad_norm": 2.28216290473938,
      "learning_rate": 0.0001975,
      "loss": 2.5725,
      "step": 79
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.966034173965454,
      "learning_rate": 0.0002,
      "loss": 3.0547,
      "step": 80
    },
    {
      "epoch": 0.08,
      "grad_norm": 2.117560625076294,
      "learning_rate": 0.00020250000000000002,
      "loss": 3.1316,
      "step": 81
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.8826152086257935,
      "learning_rate": 0.000205,
      "loss": 2.8674,
      "step": 82
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.997240424156189,
      "learning_rate": 0.0002075,
      "loss": 2.4458,
      "step": 83
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.6871142387390137,
      "learning_rate": 0.00021,
      "loss": 2.814,
      "step": 84
    },
    {
      "epoch": 0.08,
      "grad_norm": 2.066634178161621,
      "learning_rate": 0.0002125,
      "loss": 3.2066,
      "step": 85
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.9600077867507935,
      "learning_rate": 0.000215,
      "loss": 3.2272,
      "step": 86
    },
    {
      "epoch": 0.08,
      "grad_norm": 2.1469485759735107,
      "learning_rate": 0.0002175,
      "loss": 3.4447,
      "step": 87
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.944505214691162,
      "learning_rate": 0.00022,
      "loss": 2.5898,
      "step": 88
    },
    {
      "epoch": 0.09,
      "grad_norm": 2.3109183311462402,
      "learning_rate": 0.00022250000000000001,
      "loss": 2.771,
      "step": 89
    },
    {
      "epoch": 0.09,
      "grad_norm": 2.174560070037842,
      "learning_rate": 0.00022500000000000002,
      "loss": 3.0631,
      "step": 90
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.8415343761444092,
      "learning_rate": 0.0002275,
      "loss": 3.123,
      "step": 91
    },
    {
      "epoch": 0.09,
      "grad_norm": 2.104952096939087,
      "learning_rate": 0.00023,
      "loss": 3.1166,
      "step": 92
    },
    {
      "epoch": 0.09,
      "grad_norm": 2.3860461711883545,
      "learning_rate": 0.0002325,
      "loss": 3.048,
      "step": 93
    },
    {
      "epoch": 0.09,
      "grad_norm": 2.132197856903076,
      "learning_rate": 0.000235,
      "loss": 2.7315,
      "step": 94
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.7788770198822021,
      "learning_rate": 0.0002375,
      "loss": 3.127,
      "step": 95
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.8100048303604126,
      "learning_rate": 0.00024,
      "loss": 2.3229,
      "step": 96
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.7176936864852905,
      "learning_rate": 0.00024249999999999999,
      "loss": 2.8595,
      "step": 97
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.920607089996338,
      "learning_rate": 0.000245,
      "loss": 2.6285,
      "step": 98
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.8976895809173584,
      "learning_rate": 0.0002475,
      "loss": 2.3511,
      "step": 99
    },
    {
      "epoch": 0.1,
      "grad_norm": 2.0189640522003174,
      "learning_rate": 0.00025,
      "loss": 2.7418,
      "step": 100
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.9701827764511108,
      "learning_rate": 0.00024999993182517067,
      "loss": 2.0604,
      "step": 101
    },
    {
      "epoch": 0.1,
      "grad_norm": 3.252211570739746,
      "learning_rate": 0.00024999972730075704,
      "loss": 2.6496,
      "step": 102
    },
    {
      "epoch": 0.1,
      "grad_norm": 2.1773853302001953,
      "learning_rate": 0.0002499993864269822,
      "loss": 2.8563,
      "step": 103
    },
    {
      "epoch": 0.1,
      "grad_norm": 2.173621654510498,
      "learning_rate": 0.00024999890920421796,
      "loss": 2.5704,
      "step": 104
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.8827345371246338,
      "learning_rate": 0.0002499982956329849,
      "loss": 2.7194,
      "step": 105
    },
    {
      "epoch": 0.1,
      "grad_norm": 2.4898743629455566,
      "learning_rate": 0.00024999754571395234,
      "loss": 3.4996,
      "step": 106
    },
    {
      "epoch": 0.1,
      "grad_norm": 2.006110191345215,
      "learning_rate": 0.00024999665944793815,
      "loss": 2.7179,
      "step": 107
    },
    {
      "epoch": 0.1,
      "grad_norm": 2.0414044857025146,
      "learning_rate": 0.00024999563683590924,
      "loss": 2.9323,
      "step": 108
    },
    {
      "epoch": 0.11,
      "grad_norm": 2.1107375621795654,
      "learning_rate": 0.0002499944778789809,
      "loss": 2.4901,
      "step": 109
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.8949179649353027,
      "learning_rate": 0.0002499931825784175,
      "loss": 2.6771,
      "step": 110
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.852339267730713,
      "learning_rate": 0.00024999175093563185,
      "loss": 2.5123,
      "step": 111
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.7306336164474487,
      "learning_rate": 0.0002499901829521856,
      "loss": 2.7333,
      "step": 112
    },
    {
      "epoch": 0.11,
      "grad_norm": 2.1602630615234375,
      "learning_rate": 0.000249988478629789,
      "loss": 3.2803,
      "step": 113
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.8918505907058716,
      "learning_rate": 0.0002499866379703013,
      "loss": 2.7467,
      "step": 114
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.9145008325576782,
      "learning_rate": 0.0002499846609757302,
      "loss": 2.7413,
      "step": 115
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.8250600099563599,
      "learning_rate": 0.00024998254764823215,
      "loss": 2.5852,
      "step": 116
    },
    {
      "epoch": 0.11,
      "grad_norm": 2.1568727493286133,
      "learning_rate": 0.0002499802979901124,
      "loss": 2.7673,
      "step": 117
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.7749338150024414,
      "learning_rate": 0.0002499779120038249,
      "loss": 2.4479,
      "step": 118
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.7486765384674072,
      "learning_rate": 0.0002499753896919723,
      "loss": 1.8766,
      "step": 119
    },
    {
      "epoch": 0.12,
      "grad_norm": 2.076768159866333,
      "learning_rate": 0.0002499727310573059,
      "loss": 1.8791,
      "step": 120
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.604339599609375,
      "learning_rate": 0.00024996993610272567,
      "loss": 2.0573,
      "step": 121
    },
    {
      "epoch": 0.12,
      "grad_norm": 2.116293430328369,
      "learning_rate": 0.00024996700483128037,
      "loss": 2.7077,
      "step": 122
    },
    {
      "epoch": 0.12,
      "grad_norm": 2.233560800552368,
      "learning_rate": 0.0002499639372461675,
      "loss": 3.1111,
      "step": 123
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.750420093536377,
      "learning_rate": 0.0002499607333507331,
      "loss": 2.4861,
      "step": 124
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.8790860176086426,
      "learning_rate": 0.000249957393148472,
      "loss": 2.5195,
      "step": 125
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.7548282146453857,
      "learning_rate": 0.0002499539166430276,
      "loss": 2.5072,
      "step": 126
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.8878798484802246,
      "learning_rate": 0.0002499503038381922,
      "loss": 2.3164,
      "step": 127
    },
    {
      "epoch": 0.12,
      "grad_norm": 2.040093421936035,
      "learning_rate": 0.00024994655473790657,
      "loss": 2.0806,
      "step": 128
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.9570492506027222,
      "learning_rate": 0.0002499426693462602,
      "loss": 2.7483,
      "step": 129
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.9189451932907104,
      "learning_rate": 0.0002499386476674913,
      "loss": 2.4365,
      "step": 130
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.840690016746521,
      "learning_rate": 0.0002499344897059867,
      "loss": 2.2675,
      "step": 131
    },
    {
      "epoch": 0.13,
      "grad_norm": 2.0497002601623535,
      "learning_rate": 0.0002499301954662818,
      "loss": 2.3665,
      "step": 132
    },
    {
      "epoch": 0.13,
      "grad_norm": 2.057999849319458,
      "learning_rate": 0.0002499257649530609,
      "loss": 2.2258,
      "step": 133
    },
    {
      "epoch": 0.13,
      "grad_norm": 2.1040830612182617,
      "learning_rate": 0.00024992119817115674,
      "loss": 2.9013,
      "step": 134
    },
    {
      "epoch": 0.13,
      "grad_norm": 2.1239287853240967,
      "learning_rate": 0.0002499164951255507,
      "loss": 2.4111,
      "step": 135
    },
    {
      "epoch": 0.13,
      "grad_norm": 2.246864080429077,
      "learning_rate": 0.00024991165582137286,
      "loss": 2.1288,
      "step": 136
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.854050874710083,
      "learning_rate": 0.000249906680263902,
      "loss": 2.308,
      "step": 137
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.7639249563217163,
      "learning_rate": 0.0002499015684585654,
      "loss": 2.1624,
      "step": 138
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.9622340202331543,
      "learning_rate": 0.0002498963204109389,
      "loss": 1.7293,
      "step": 139
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.9182345867156982,
      "learning_rate": 0.0002498909361267472,
      "loss": 2.5676,
      "step": 140
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.546036720275879,
      "learning_rate": 0.0002498854156118634,
      "loss": 2.0416,
      "step": 141
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.7901947498321533,
      "learning_rate": 0.0002498797588723093,
      "loss": 2.3867,
      "step": 142
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.682002305984497,
      "learning_rate": 0.00024987396591425517,
      "loss": 2.2,
      "step": 143
    },
    {
      "epoch": 0.14,
      "grad_norm": 2.1231274604797363,
      "learning_rate": 0.00024986803674402003,
      "loss": 2.7393,
      "step": 144
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.7583919763565063,
      "learning_rate": 0.0002498619713680714,
      "loss": 2.2841,
      "step": 145
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.6599968671798706,
      "learning_rate": 0.00024985576979302533,
      "loss": 1.9753,
      "step": 146
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.5316509008407593,
      "learning_rate": 0.00024984943202564655,
      "loss": 1.9288,
      "step": 147
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.8190314769744873,
      "learning_rate": 0.0002498429580728482,
      "loss": 2.2112,
      "step": 148
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.7357500791549683,
      "learning_rate": 0.00024983634794169214,
      "loss": 2.2501,
      "step": 149
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.8484951257705688,
      "learning_rate": 0.0002498296016393886,
      "loss": 1.7479,
      "step": 150
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.765267252922058,
      "learning_rate": 0.00024982271917329646,
      "loss": 2.5703,
      "step": 151
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.6502829790115356,
      "learning_rate": 0.0002498157005509231,
      "loss": 2.7001,
      "step": 152
    },
    {
      "epoch": 0.15,
      "grad_norm": 2.010906934738159,
      "learning_rate": 0.0002498085457799244,
      "loss": 2.8633,
      "step": 153
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.850074052810669,
      "learning_rate": 0.00024980125486810477,
      "loss": 2.2063,
      "step": 154
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.9126758575439453,
      "learning_rate": 0.00024979382782341713,
      "loss": 2.1976,
      "step": 155
    },
    {
      "epoch": 0.15,
      "grad_norm": 2.0436511039733887,
      "learning_rate": 0.00024978626465396286,
      "loss": 2.3515,
      "step": 156
    },
    {
      "epoch": 0.15,
      "grad_norm": 2.3156819343566895,
      "learning_rate": 0.0002497785653679919,
      "loss": 2.7545,
      "step": 157
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.9115028381347656,
      "learning_rate": 0.00024977072997390247,
      "loss": 2.206,
      "step": 158
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.8286609649658203,
      "learning_rate": 0.00024976275848024156,
      "loss": 2.371,
      "step": 159
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.6878124475479126,
      "learning_rate": 0.0002497546508957044,
      "loss": 2.1747,
      "step": 160
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.893005132675171,
      "learning_rate": 0.00024974640722913465,
      "loss": 1.7254,
      "step": 161
    },
    {
      "epoch": 0.16,
      "grad_norm": 2.11098313331604,
      "learning_rate": 0.0002497380274895246,
      "loss": 2.5951,
      "step": 162
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.7469336986541748,
      "learning_rate": 0.00024972951168601476,
      "loss": 2.2754,
      "step": 163
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.835694670677185,
      "learning_rate": 0.00024972085982789415,
      "loss": 2.1267,
      "step": 164
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.6754062175750732,
      "learning_rate": 0.0002497120719246002,
      "loss": 1.9974,
      "step": 165
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.8209201097488403,
      "learning_rate": 0.0002497031479857188,
      "loss": 2.5032,
      "step": 166
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.76840341091156,
      "learning_rate": 0.000249694088020984,
      "loss": 1.7389,
      "step": 167
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.552156925201416,
      "learning_rate": 0.00024968489204027863,
      "loss": 1.8085,
      "step": 168
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.761948585510254,
      "learning_rate": 0.0002496755600536334,
      "loss": 2.0805,
      "step": 169
    },
    {
      "epoch": 0.16,
      "grad_norm": 2.200204372406006,
      "learning_rate": 0.0002496660920712277,
      "loss": 2.0036,
      "step": 170
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.8660489320755005,
      "learning_rate": 0.0002496564881033892,
      "loss": 2.1293,
      "step": 171
    },
    {
      "epoch": 0.17,
      "grad_norm": 2.33247709274292,
      "learning_rate": 0.00024964674816059393,
      "loss": 2.3811,
      "step": 172
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.8773770332336426,
      "learning_rate": 0.00024963687225346604,
      "loss": 1.9161,
      "step": 173
    },
    {
      "epoch": 0.17,
      "grad_norm": 2.1089746952056885,
      "learning_rate": 0.0002496268603927783,
      "loss": 2.5818,
      "step": 174
    },
    {
      "epoch": 0.17,
      "grad_norm": 2.0784616470336914,
      "learning_rate": 0.00024961671258945156,
      "loss": 2.0916,
      "step": 175
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.8207252025604248,
      "learning_rate": 0.000249606428854555,
      "loss": 2.3204,
      "step": 176
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.9220213890075684,
      "learning_rate": 0.00024959600919930607,
      "loss": 2.5432,
      "step": 177
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.8817466497421265,
      "learning_rate": 0.0002495854536350706,
      "loss": 2.4326,
      "step": 178
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.8520444631576538,
      "learning_rate": 0.0002495747621733625,
      "loss": 1.9673,
      "step": 179
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.7097537517547607,
      "learning_rate": 0.00024956393482584397,
      "loss": 2.0189,
      "step": 180
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.8573436737060547,
      "learning_rate": 0.0002495529716043254,
      "loss": 2.0942,
      "step": 181
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.9761507511138916,
      "learning_rate": 0.00024954187252076564,
      "loss": 2.189,
      "step": 182
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.8549177646636963,
      "learning_rate": 0.00024953063758727137,
      "loss": 2.0821,
      "step": 183
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.7199348211288452,
      "learning_rate": 0.00024951926681609767,
      "loss": 1.9875,
      "step": 184
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.6276706457138062,
      "learning_rate": 0.00024950776021964775,
      "loss": 2.0707,
      "step": 185
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.5845235586166382,
      "learning_rate": 0.000249496117810473,
      "loss": 1.5413,
      "step": 186
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.7384014129638672,
      "learning_rate": 0.00024948433960127284,
      "loss": 2.1632,
      "step": 187
    },
    {
      "epoch": 0.18,
      "grad_norm": 2.0509490966796875,
      "learning_rate": 0.000249472425604895,
      "loss": 1.758,
      "step": 188
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.9983670711517334,
      "learning_rate": 0.0002494603758343352,
      "loss": 2.5845,
      "step": 189
    },
    {
      "epoch": 0.18,
      "grad_norm": 2.0809295177459717,
      "learning_rate": 0.0002494481903027373,
      "loss": 2.6146,
      "step": 190
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.7669973373413086,
      "learning_rate": 0.0002494358690233933,
      "loss": 2.0003,
      "step": 191
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.7728503942489624,
      "learning_rate": 0.0002494234120097431,
      "loss": 2.0072,
      "step": 192
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.9957607984542847,
      "learning_rate": 0.0002494108192753748,
      "loss": 2.2449,
      "step": 193
    },
    {
      "epoch": 0.19,
      "grad_norm": 2.0449464321136475,
      "learning_rate": 0.0002493980908340246,
      "loss": 2.0996,
      "step": 194
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.8009361028671265,
      "learning_rate": 0.0002493852266995766,
      "loss": 2.0876,
      "step": 195
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.9847387075424194,
      "learning_rate": 0.00024937222688606303,
      "loss": 1.8041,
      "step": 196
    },
    {
      "epoch": 0.19,
      "grad_norm": 2.0888640880584717,
      "learning_rate": 0.00024935909140766393,
      "loss": 2.4709,
      "step": 197
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.6796460151672363,
      "learning_rate": 0.0002493458202787075,
      "loss": 1.9681,
      "step": 198
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.6660256385803223,
      "learning_rate": 0.0002493324135136699,
      "loss": 2.5484,
      "step": 199
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.8919657468795776,
      "learning_rate": 0.0002493188711271751,
      "loss": 2.5249,
      "step": 200
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.8183636665344238,
      "learning_rate": 0.0002493051931339952,
      "loss": 1.8468,
      "step": 201
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.6800963878631592,
      "learning_rate": 0.0002492913795490501,
      "loss": 1.9648,
      "step": 202
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.8262184858322144,
      "learning_rate": 0.00024927743038740747,
      "loss": 2.0492,
      "step": 203
    },
    {
      "epoch": 0.2,
      "grad_norm": 2.1556880474090576,
      "learning_rate": 0.0002492633456642832,
      "loss": 2.4633,
      "step": 204
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.7076188325881958,
      "learning_rate": 0.0002492491253950408,
      "loss": 2.1612,
      "step": 205
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.8892126083374023,
      "learning_rate": 0.0002492347695951917,
      "loss": 1.9447,
      "step": 206
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.873687744140625,
      "learning_rate": 0.0002492202782803952,
      "loss": 2.7744,
      "step": 207
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.7962186336517334,
      "learning_rate": 0.0002492056514664583,
      "loss": 2.0622,
      "step": 208
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.9481604099273682,
      "learning_rate": 0.00024919088916933597,
      "loss": 1.9443,
      "step": 209
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.7974705696105957,
      "learning_rate": 0.0002491759914051308,
      "loss": 2.3544,
      "step": 210
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.5557405948638916,
      "learning_rate": 0.00024916095819009336,
      "loss": 2.2321,
      "step": 211
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.729710340499878,
      "learning_rate": 0.0002491457895406218,
      "loss": 1.9179,
      "step": 212
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.6658893823623657,
      "learning_rate": 0.00024913048547326193,
      "loss": 2.088,
      "step": 213
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.6439865827560425,
      "learning_rate": 0.0002491150460047075,
      "loss": 1.8413,
      "step": 214
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.595862865447998,
      "learning_rate": 0.00024909947115179983,
      "loss": 2.0446,
      "step": 215
    },
    {
      "epoch": 0.21,
      "grad_norm": 2.21343994140625,
      "learning_rate": 0.00024908376093152784,
      "loss": 2.841,
      "step": 216
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.7837032079696655,
      "learning_rate": 0.0002490679153610283,
      "loss": 1.8496,
      "step": 217
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.6079756021499634,
      "learning_rate": 0.00024905193445758545,
      "loss": 1.9118,
      "step": 218
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.9322482347488403,
      "learning_rate": 0.00024903581823863125,
      "loss": 2.2768,
      "step": 219
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.9454126358032227,
      "learning_rate": 0.0002490195667217452,
      "loss": 2.0224,
      "step": 220
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.8890690803527832,
      "learning_rate": 0.00024900317992465447,
      "loss": 1.9491,
      "step": 221
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.844327449798584,
      "learning_rate": 0.0002489866578652337,
      "loss": 2.3316,
      "step": 222
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.9561625719070435,
      "learning_rate": 0.00024897000056150505,
      "loss": 1.8929,
      "step": 223
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.812215805053711,
      "learning_rate": 0.0002489532080316383,
      "loss": 1.5681,
      "step": 224
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.7232577800750732,
      "learning_rate": 0.0002489362802939507,
      "loss": 2.2427,
      "step": 225
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.5524508953094482,
      "learning_rate": 0.00024891921736690703,
      "loss": 2.1043,
      "step": 226
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.5697345733642578,
      "learning_rate": 0.0002489020192691194,
      "loss": 1.9664,
      "step": 227
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.5127205848693848,
      "learning_rate": 0.0002488846860193475,
      "loss": 1.5816,
      "step": 228
    },
    {
      "epoch": 0.22,
      "grad_norm": 2.036536693572998,
      "learning_rate": 0.0002488672176364984,
      "loss": 2.7449,
      "step": 229
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.9850804805755615,
      "learning_rate": 0.0002488496141396265,
      "loss": 1.8061,
      "step": 230
    },
    {
      "epoch": 0.22,
      "grad_norm": 2.029428720474243,
      "learning_rate": 0.0002488318755479337,
      "loss": 1.7861,
      "step": 231
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.6909549236297607,
      "learning_rate": 0.00024881400188076923,
      "loss": 2.1058,
      "step": 232
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.7147257328033447,
      "learning_rate": 0.0002487959931576296,
      "loss": 1.6736,
      "step": 233
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.6619194746017456,
      "learning_rate": 0.00024877784939815863,
      "loss": 1.7423,
      "step": 234
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.7721213102340698,
      "learning_rate": 0.0002487595706221476,
      "loss": 1.4057,
      "step": 235
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.6098127365112305,
      "learning_rate": 0.00024874115684953485,
      "loss": 1.6113,
      "step": 236
    },
    {
      "epoch": 0.23,
      "grad_norm": 2.2047505378723145,
      "learning_rate": 0.00024872260810040607,
      "loss": 2.5387,
      "step": 237
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.6279314756393433,
      "learning_rate": 0.0002487039243949943,
      "loss": 1.9872,
      "step": 238
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.795262098312378,
      "learning_rate": 0.0002486851057536795,
      "loss": 1.4461,
      "step": 239
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.9134374856948853,
      "learning_rate": 0.00024866615219698915,
      "loss": 2.3969,
      "step": 240
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.6883046627044678,
      "learning_rate": 0.0002486470637455976,
      "loss": 2.2354,
      "step": 241
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.7430942058563232,
      "learning_rate": 0.00024862784042032666,
      "loss": 1.5288,
      "step": 242
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.6902698278427124,
      "learning_rate": 0.00024860848224214486,
      "loss": 2.102,
      "step": 243
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.5325380563735962,
      "learning_rate": 0.0002485889892321683,
      "loss": 1.3993,
      "step": 244
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.5996594429016113,
      "learning_rate": 0.00024856936141165963,
      "loss": 1.586,
      "step": 245
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.9675129652023315,
      "learning_rate": 0.00024854959880202905,
      "loss": 2.6326,
      "step": 246
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.9281418323516846,
      "learning_rate": 0.00024852970142483346,
      "loss": 2.3056,
      "step": 247
    },
    {
      "epoch": 0.24,
      "grad_norm": 2.0125837326049805,
      "learning_rate": 0.00024850966930177687,
      "loss": 2.2092,
      "step": 248
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.904648780822754,
      "learning_rate": 0.00024848950245471023,
      "loss": 2.1627,
      "step": 249
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.7093470096588135,
      "learning_rate": 0.00024846920090563156,
      "loss": 1.9851,
      "step": 250
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.7834309339523315,
      "learning_rate": 0.0002484487646766857,
      "loss": 2.112,
      "step": 251
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.5248178243637085,
      "learning_rate": 0.0002484281937901644,
      "loss": 1.426,
      "step": 252
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.4602184295654297,
      "learning_rate": 0.0002484074882685063,
      "loss": 1.3696,
      "step": 253
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.773917555809021,
      "learning_rate": 0.0002483866481342971,
      "loss": 2.1261,
      "step": 254
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.692901611328125,
      "learning_rate": 0.000248365673410269,
      "loss": 2.3204,
      "step": 255
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.8946104049682617,
      "learning_rate": 0.0002483445641193012,
      "loss": 1.8685,
      "step": 256
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.919169306755066,
      "learning_rate": 0.0002483233202844197,
      "loss": 2.0433,
      "step": 257
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.84755277633667,
      "learning_rate": 0.00024830194192879715,
      "loss": 2.2363,
      "step": 258
    },
    {
      "epoch": 0.25,
      "grad_norm": 2.112445116043091,
      "learning_rate": 0.00024828042907575304,
      "loss": 2.6892,
      "step": 259
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.6115814447402954,
      "learning_rate": 0.0002482587817487536,
      "loss": 1.8159,
      "step": 260
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.6731853485107422,
      "learning_rate": 0.00024823699997141154,
      "loss": 2.1162,
      "step": 261
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.5666356086730957,
      "learning_rate": 0.0002482150837674864,
      "loss": 1.8052,
      "step": 262
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.5038379430770874,
      "learning_rate": 0.0002481930331608844,
      "loss": 1.8459,
      "step": 263
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.5254186391830444,
      "learning_rate": 0.00024817084817565827,
      "loss": 1.9475,
      "step": 264
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.7081129550933838,
      "learning_rate": 0.0002481485288360072,
      "loss": 2.4121,
      "step": 265
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.9189531803131104,
      "learning_rate": 0.0002481260751662772,
      "loss": 2.4579,
      "step": 266
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.6706724166870117,
      "learning_rate": 0.00024810348719096065,
      "loss": 1.6386,
      "step": 267
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.539415717124939,
      "learning_rate": 0.0002480807649346964,
      "loss": 1.6492,
      "step": 268
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.747955560684204,
      "learning_rate": 0.00024805790842226985,
      "loss": 1.6512,
      "step": 269
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.6738513708114624,
      "learning_rate": 0.0002480349176786128,
      "loss": 1.7555,
      "step": 270
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.9724032878875732,
      "learning_rate": 0.0002480117927288035,
      "loss": 1.6291,
      "step": 271
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.6531236171722412,
      "learning_rate": 0.00024798853359806665,
      "loss": 1.6063,
      "step": 272
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.9296516180038452,
      "learning_rate": 0.0002479651403117732,
      "loss": 1.4847,
      "step": 273
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.7940136194229126,
      "learning_rate": 0.0002479416128954404,
      "loss": 2.4839,
      "step": 274
    },
    {
      "epoch": 0.27,
      "grad_norm": 2.010266065597534,
      "learning_rate": 0.00024791795137473204,
      "loss": 1.7243,
      "step": 275
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.8504762649536133,
      "learning_rate": 0.00024789415577545793,
      "loss": 1.9652,
      "step": 276
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.4306342601776123,
      "learning_rate": 0.0002478702261235743,
      "loss": 1.5664,
      "step": 277
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.6604520082473755,
      "learning_rate": 0.0002478461624451835,
      "loss": 2.0659,
      "step": 278
    },
    {
      "epoch": 0.27,
      "grad_norm": 2.0522496700286865,
      "learning_rate": 0.0002478219647665342,
      "loss": 1.9088,
      "step": 279
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.6429219245910645,
      "learning_rate": 0.0002477976331140211,
      "loss": 1.7491,
      "step": 280
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.397255301475525,
      "learning_rate": 0.00024777316751418515,
      "loss": 1.3155,
      "step": 281
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.5616679191589355,
      "learning_rate": 0.00024774856799371326,
      "loss": 1.8068,
      "step": 282
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.7590296268463135,
      "learning_rate": 0.00024772383457943864,
      "loss": 1.7305,
      "step": 283
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.6376055479049683,
      "learning_rate": 0.00024769896729834036,
      "loss": 1.7105,
      "step": 284
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.385542392730713,
      "learning_rate": 0.00024767396617754364,
      "loss": 1.605,
      "step": 285
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.827626347541809,
      "learning_rate": 0.0002476488312443195,
      "loss": 2.032,
      "step": 286
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.1502676010131836,
      "learning_rate": 0.00024762356252608527,
      "loss": 1.9228,
      "step": 287
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.5369927883148193,
      "learning_rate": 0.00024759816005040384,
      "loss": 1.3886,
      "step": 288
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.7193357944488525,
      "learning_rate": 0.0002475726238449842,
      "loss": 2.0007,
      "step": 289
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.561558246612549,
      "learning_rate": 0.00024754695393768114,
      "loss": 2.6965,
      "step": 290
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.2155981063842773,
      "learning_rate": 0.0002475211503564954,
      "loss": 1.678,
      "step": 291
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.59506356716156,
      "learning_rate": 0.00024749521312957337,
      "loss": 1.7036,
      "step": 292
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.5700795650482178,
      "learning_rate": 0.0002474691422852074,
      "loss": 1.5835,
      "step": 293
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.8655942678451538,
      "learning_rate": 0.00024744293785183537,
      "loss": 1.6957,
      "step": 294
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.434198260307312,
      "learning_rate": 0.00024741659985804115,
      "loss": 1.5576,
      "step": 295
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.4937015771865845,
      "learning_rate": 0.0002473901283325541,
      "loss": 1.7204,
      "step": 296
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.9438872337341309,
      "learning_rate": 0.00024736352330424923,
      "loss": 2.2574,
      "step": 297
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.8500808477401733,
      "learning_rate": 0.0002473367848021473,
      "loss": 2.2356,
      "step": 298
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.6820452213287354,
      "learning_rate": 0.00024730991285541455,
      "loss": 2.7996,
      "step": 299
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.6098867654800415,
      "learning_rate": 0.0002472829074933628,
      "loss": 2.0764,
      "step": 300
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.763376235961914,
      "learning_rate": 0.00024725576874544956,
      "loss": 2.1674,
      "step": 301
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.7412651777267456,
      "learning_rate": 0.0002472284966412776,
      "loss": 2.2091,
      "step": 302
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.690738320350647,
      "learning_rate": 0.00024720109121059524,
      "loss": 2.1094,
      "step": 303
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.3498783111572266,
      "learning_rate": 0.00024717355248329625,
      "loss": 1.8,
      "step": 304
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.6631073951721191,
      "learning_rate": 0.00024714588048941987,
      "loss": 2.251,
      "step": 305
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.7227246761322021,
      "learning_rate": 0.0002471180752591506,
      "loss": 1.8822,
      "step": 306
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.7856996059417725,
      "learning_rate": 0.00024709013682281826,
      "loss": 1.914,
      "step": 307
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.0128395557403564,
      "learning_rate": 0.0002470620652108981,
      "loss": 2.0113,
      "step": 308
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.4962693452835083,
      "learning_rate": 0.00024703386045401047,
      "loss": 1.7985,
      "step": 309
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.5838319063186646,
      "learning_rate": 0.0002470055225829211,
      "loss": 1.5443,
      "step": 310
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.6290487051010132,
      "learning_rate": 0.0002469770516285409,
      "loss": 1.6863,
      "step": 311
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.46754789352417,
      "learning_rate": 0.0002469484476219259,
      "loss": 1.6632,
      "step": 312
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.8494882583618164,
      "learning_rate": 0.00024691971059427717,
      "loss": 1.7439,
      "step": 313
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.868640661239624,
      "learning_rate": 0.000246890840576941,
      "loss": 2.0376,
      "step": 314
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.553097128868103,
      "learning_rate": 0.0002468618376014088,
      "loss": 1.9508,
      "step": 315
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.6028902530670166,
      "learning_rate": 0.000246832701699317,
      "loss": 1.5764,
      "step": 316
    },
    {
      "epoch": 0.31,
      "grad_norm": 2.1267387866973877,
      "learning_rate": 0.0002468034329024468,
      "loss": 1.8886,
      "step": 317
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.5905228853225708,
      "learning_rate": 0.00024677403124272456,
      "loss": 1.8134,
      "step": 318
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.4769141674041748,
      "learning_rate": 0.0002467444967522216,
      "loss": 1.2115,
      "step": 319
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.4753649234771729,
      "learning_rate": 0.000246714829463154,
      "loss": 1.653,
      "step": 320
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.6038315296173096,
      "learning_rate": 0.0002466850294078828,
      "loss": 2.1011,
      "step": 321
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.7277235984802246,
      "learning_rate": 0.00024665509661891385,
      "loss": 2.1318,
      "step": 322
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.6639180183410645,
      "learning_rate": 0.0002466250311288977,
      "loss": 1.6277,
      "step": 323
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.660370111465454,
      "learning_rate": 0.00024659483297062964,
      "loss": 2.112,
      "step": 324
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.7722567319869995,
      "learning_rate": 0.0002465645021770499,
      "loss": 2.0158,
      "step": 325
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.559173583984375,
      "learning_rate": 0.00024653403878124305,
      "loss": 1.8675,
      "step": 326
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.679623007774353,
      "learning_rate": 0.0002465034428164386,
      "loss": 1.844,
      "step": 327
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.72882878780365,
      "learning_rate": 0.00024647271431601055,
      "loss": 2.2323,
      "step": 328
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.4677948951721191,
      "learning_rate": 0.00024644185331347735,
      "loss": 1.8718,
      "step": 329
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.5333250761032104,
      "learning_rate": 0.00024641085984250223,
      "loss": 2.0585,
      "step": 330
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.307824730873108,
      "learning_rate": 0.0002463797339368927,
      "loss": 1.8447,
      "step": 331
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.6973696947097778,
      "learning_rate": 0.0002463484756306009,
      "loss": 1.5235,
      "step": 332
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.3276690244674683,
      "learning_rate": 0.0002463170849577232,
      "loss": 1.9472,
      "step": 333
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.557236671447754,
      "learning_rate": 0.0002462855619525005,
      "loss": 1.6783,
      "step": 334
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.6383414268493652,
      "learning_rate": 0.000246253906649318,
      "loss": 1.4636,
      "step": 335
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.4756358861923218,
      "learning_rate": 0.0002462221190827053,
      "loss": 1.4215,
      "step": 336
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.6746768951416016,
      "learning_rate": 0.000246190199287336,
      "loss": 1.7268,
      "step": 337
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.8532724380493164,
      "learning_rate": 0.00024615814729802833,
      "loss": 2.2173,
      "step": 338
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.8261704444885254,
      "learning_rate": 0.0002461259631497444,
      "loss": 1.6643,
      "step": 339
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.6665236949920654,
      "learning_rate": 0.0002460936468775907,
      "loss": 1.9251,
      "step": 340
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.5993882417678833,
      "learning_rate": 0.00024606119851681757,
      "loss": 1.4908,
      "step": 341
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.8725510835647583,
      "learning_rate": 0.00024602861810281966,
      "loss": 2.1784,
      "step": 342
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.803350567817688,
      "learning_rate": 0.0002459959056711357,
      "loss": 2.11,
      "step": 343
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.6295034885406494,
      "learning_rate": 0.00024596306125744815,
      "loss": 1.777,
      "step": 344
    },
    {
      "epoch": 0.33,
      "grad_norm": 2.189607858657837,
      "learning_rate": 0.00024593008489758375,
      "loss": 2.4662,
      "step": 345
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.835728645324707,
      "learning_rate": 0.0002458969766275129,
      "loss": 1.8764,
      "step": 346
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.6741111278533936,
      "learning_rate": 0.00024586373648335014,
      "loss": 1.8256,
      "step": 347
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.5066823959350586,
      "learning_rate": 0.0002458303645013536,
      "loss": 1.4491,
      "step": 348
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.439063549041748,
      "learning_rate": 0.00024579686071792543,
      "loss": 1.8896,
      "step": 349
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.2669867277145386,
      "learning_rate": 0.0002457632251696115,
      "loss": 1.7784,
      "step": 350
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.4590708017349243,
      "learning_rate": 0.00024572945789310123,
      "loss": 1.5708,
      "step": 351
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.7421554327011108,
      "learning_rate": 0.000245695558925228,
      "loss": 2.0778,
      "step": 352
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.5311520099639893,
      "learning_rate": 0.00024566152830296875,
      "loss": 1.4703,
      "step": 353
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.9303866624832153,
      "learning_rate": 0.0002456273660634438,
      "loss": 2.0816,
      "step": 354
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.5437086820602417,
      "learning_rate": 0.0002455930722439174,
      "loss": 1.4711,
      "step": 355
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.50086510181427,
      "learning_rate": 0.0002455586468817971,
      "loss": 1.8523,
      "step": 356
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.5427656173706055,
      "learning_rate": 0.00024552409001463393,
      "loss": 1.6157,
      "step": 357
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.7002308368682861,
      "learning_rate": 0.00024548940168012253,
      "loss": 1.724,
      "step": 358
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.5757559537887573,
      "learning_rate": 0.0002454545819161008,
      "loss": 1.663,
      "step": 359
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.4780895709991455,
      "learning_rate": 0.00024541963076055,
      "loss": 1.488,
      "step": 360
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.7047303915023804,
      "learning_rate": 0.00024538454825159486,
      "loss": 2.0954,
      "step": 361
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.9181485176086426,
      "learning_rate": 0.00024534933442750317,
      "loss": 1.5492,
      "step": 362
    },
    {
      "epoch": 0.35,
      "grad_norm": 2.150815010070801,
      "learning_rate": 0.0002453139893266861,
      "loss": 2.1316,
      "step": 363
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.663496732711792,
      "learning_rate": 0.00024527851298769803,
      "loss": 1.9913,
      "step": 364
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.6102313995361328,
      "learning_rate": 0.00024524290544923643,
      "loss": 1.8153,
      "step": 365
    },
    {
      "epoch": 0.35,
      "grad_norm": 2.003488302230835,
      "learning_rate": 0.0002452071667501419,
      "loss": 2.0315,
      "step": 366
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.4711098670959473,
      "learning_rate": 0.0002451712969293982,
      "loss": 1.4064,
      "step": 367
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.4077496528625488,
      "learning_rate": 0.0002451352960261319,
      "loss": 1.8027,
      "step": 368
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.4813429117202759,
      "learning_rate": 0.0002450991640796127,
      "loss": 1.9595,
      "step": 369
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.755618929862976,
      "learning_rate": 0.00024506290112925335,
      "loss": 1.8781,
      "step": 370
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.6948919296264648,
      "learning_rate": 0.00024502650721460926,
      "loss": 1.3217,
      "step": 371
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.6204948425292969,
      "learning_rate": 0.00024498998237537883,
      "loss": 1.7845,
      "step": 372
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.4599529504776,
      "learning_rate": 0.00024495332665140337,
      "loss": 1.8832,
      "step": 373
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.5336644649505615,
      "learning_rate": 0.00024491654008266666,
      "loss": 1.7582,
      "step": 374
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.6207972764968872,
      "learning_rate": 0.00024487962270929546,
      "loss": 1.6177,
      "step": 375
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.5989536046981812,
      "learning_rate": 0.0002448425745715592,
      "loss": 2.1292,
      "step": 376
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.7636903524398804,
      "learning_rate": 0.0002448053957098699,
      "loss": 1.8165,
      "step": 377
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.6132402420043945,
      "learning_rate": 0.0002447680861647821,
      "loss": 1.8227,
      "step": 378
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.8587868213653564,
      "learning_rate": 0.0002447306459769929,
      "loss": 2.365,
      "step": 379
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.4449355602264404,
      "learning_rate": 0.000244693075187342,
      "loss": 1.8724,
      "step": 380
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.2807527780532837,
      "learning_rate": 0.0002446553738368116,
      "loss": 1.7084,
      "step": 381
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.2705549001693726,
      "learning_rate": 0.0002446175419665261,
      "loss": 1.5953,
      "step": 382
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.444528341293335,
      "learning_rate": 0.00024457957961775253,
      "loss": 1.7351,
      "step": 383
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.6186937093734741,
      "learning_rate": 0.00024454148683189996,
      "loss": 2.1234,
      "step": 384
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.2942836284637451,
      "learning_rate": 0.00024450326365052,
      "loss": 1.5367,
      "step": 385
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.3657950162887573,
      "learning_rate": 0.0002444649101153064,
      "loss": 1.417,
      "step": 386
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.2977917194366455,
      "learning_rate": 0.0002444264262680951,
      "loss": 1.6442,
      "step": 387
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.5234405994415283,
      "learning_rate": 0.0002443878121508641,
      "loss": 1.7331,
      "step": 388
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.580356478691101,
      "learning_rate": 0.00024434906780573356,
      "loss": 1.5812,
      "step": 389
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.5635511875152588,
      "learning_rate": 0.0002443101932749658,
      "loss": 1.7743,
      "step": 390
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.4234613180160522,
      "learning_rate": 0.00024427118860096504,
      "loss": 1.9202,
      "step": 391
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.8113735914230347,
      "learning_rate": 0.00024423205382627746,
      "loss": 2.3395,
      "step": 392
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.5529452562332153,
      "learning_rate": 0.0002441927889935911,
      "loss": 1.6039,
      "step": 393
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.3958035707473755,
      "learning_rate": 0.000244153394145736,
      "loss": 1.5429,
      "step": 394
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.723780632019043,
      "learning_rate": 0.0002441138693256839,
      "loss": 1.7384,
      "step": 395
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.403164029121399,
      "learning_rate": 0.00024407421457654845,
      "loss": 0.9538,
      "step": 396
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.4130743741989136,
      "learning_rate": 0.00024403442994158487,
      "loss": 1.9526,
      "step": 397
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.4831724166870117,
      "learning_rate": 0.00024399451546419017,
      "loss": 1.6449,
      "step": 398
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.6771824359893799,
      "learning_rate": 0.00024395447118790293,
      "loss": 1.7758,
      "step": 399
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.5150790214538574,
      "learning_rate": 0.00024391429715640335,
      "loss": 1.9961,
      "step": 400
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.5657678842544556,
      "learning_rate": 0.0002438739934135131,
      "loss": 1.5219,
      "step": 401
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.4939355850219727,
      "learning_rate": 0.00024383356000319548,
      "loss": 1.5112,
      "step": 402
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.6075009107589722,
      "learning_rate": 0.0002437929969695551,
      "loss": 1.4834,
      "step": 403
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.6066113710403442,
      "learning_rate": 0.000243752304356838,
      "loss": 1.655,
      "step": 404
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.713575005531311,
      "learning_rate": 0.0002437114822094316,
      "loss": 1.8589,
      "step": 405
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.6600104570388794,
      "learning_rate": 0.00024367053057186455,
      "loss": 1.9319,
      "step": 406
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.8421694040298462,
      "learning_rate": 0.0002436294494888068,
      "loss": 2.1056,
      "step": 407
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.7146579027175903,
      "learning_rate": 0.0002435882390050695,
      "loss": 2.3835,
      "step": 408
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.558237075805664,
      "learning_rate": 0.0002435468991656049,
      "loss": 1.972,
      "step": 409
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.5981314182281494,
      "learning_rate": 0.0002435054300155064,
      "loss": 1.8716,
      "step": 410
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.376088261604309,
      "learning_rate": 0.00024346383160000847,
      "loss": 1.6165,
      "step": 411
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.5061323642730713,
      "learning_rate": 0.0002434221039644865,
      "loss": 1.6584,
      "step": 412
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.7581522464752197,
      "learning_rate": 0.00024338024715445688,
      "loss": 1.9282,
      "step": 413
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.3032132387161255,
      "learning_rate": 0.00024333826121557695,
      "loss": 1.1937,
      "step": 414
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.632356882095337,
      "learning_rate": 0.0002432961461936448,
      "loss": 2.075,
      "step": 415
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.2259700298309326,
      "learning_rate": 0.00024325390213459944,
      "loss": 1.1513,
      "step": 416
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.721431016921997,
      "learning_rate": 0.00024321152908452054,
      "loss": 1.5305,
      "step": 417
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.5428179502487183,
      "learning_rate": 0.00024316902708962848,
      "loss": 1.7885,
      "step": 418
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.6045695543289185,
      "learning_rate": 0.0002431263961962844,
      "loss": 1.6599,
      "step": 419
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.6771968603134155,
      "learning_rate": 0.00024308363645098984,
      "loss": 1.6666,
      "step": 420
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.5328516960144043,
      "learning_rate": 0.00024304074790038716,
      "loss": 1.7956,
      "step": 421
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.6807280778884888,
      "learning_rate": 0.00024299773059125896,
      "loss": 1.7099,
      "step": 422
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.528563380241394,
      "learning_rate": 0.00024295458457052844,
      "loss": 1.5751,
      "step": 423
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.2582366466522217,
      "learning_rate": 0.00024291130988525917,
      "loss": 1.5967,
      "step": 424
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.5818768739700317,
      "learning_rate": 0.00024286790658265507,
      "loss": 2.0187,
      "step": 425
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.647545576095581,
      "learning_rate": 0.00024282437471006033,
      "loss": 2.2781,
      "step": 426
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.1883927583694458,
      "learning_rate": 0.00024278071431495937,
      "loss": 1.7889,
      "step": 427
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.8100999593734741,
      "learning_rate": 0.00024273692544497688,
      "loss": 1.6325,
      "step": 428
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.4264518022537231,
      "learning_rate": 0.0002426930081478776,
      "loss": 1.8201,
      "step": 429
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.2940385341644287,
      "learning_rate": 0.00024264896247156643,
      "loss": 1.508,
      "step": 430
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.3940926790237427,
      "learning_rate": 0.00024260478846408823,
      "loss": 1.317,
      "step": 431
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.3203697204589844,
      "learning_rate": 0.00024256048617362792,
      "loss": 1.5154,
      "step": 432
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.5709110498428345,
      "learning_rate": 0.00024251605564851032,
      "loss": 1.4274,
      "step": 433
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.4669488668441772,
      "learning_rate": 0.00024247149693720012,
      "loss": 1.5945,
      "step": 434
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.9155356884002686,
      "learning_rate": 0.00024242681008830184,
      "loss": 1.727,
      "step": 435
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.4303950071334839,
      "learning_rate": 0.00024238199515055976,
      "loss": 1.215,
      "step": 436
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.5403741598129272,
      "learning_rate": 0.0002423370521728579,
      "loss": 1.2804,
      "step": 437
    },
    {
      "epoch": 0.42,
      "grad_norm": 2.0623433589935303,
      "learning_rate": 0.00024229198120421993,
      "loss": 1.5453,
      "step": 438
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.7625672817230225,
      "learning_rate": 0.00024224678229380913,
      "loss": 1.923,
      "step": 439
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.7964345216751099,
      "learning_rate": 0.00024220145549092842,
      "loss": 1.68,
      "step": 440
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.8445159196853638,
      "learning_rate": 0.00024215600084502006,
      "loss": 1.8517,
      "step": 441
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.7241743803024292,
      "learning_rate": 0.0002421104184056659,
      "loss": 1.8859,
      "step": 442
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.3813362121582031,
      "learning_rate": 0.00024206470822258713,
      "loss": 1.3435,
      "step": 443
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.6187509298324585,
      "learning_rate": 0.0002420188703456443,
      "loss": 1.9744,
      "step": 444
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.5142652988433838,
      "learning_rate": 0.00024197290482483725,
      "loss": 1.768,
      "step": 445
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.3997564315795898,
      "learning_rate": 0.000241926811710305,
      "loss": 1.6986,
      "step": 446
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.248969554901123,
      "learning_rate": 0.00024188059105232585,
      "loss": 1.2985,
      "step": 447
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.3714370727539062,
      "learning_rate": 0.00024183424290131716,
      "loss": 1.626,
      "step": 448
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.3766262531280518,
      "learning_rate": 0.0002417877673078353,
      "loss": 1.3913,
      "step": 449
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.621702790260315,
      "learning_rate": 0.00024174116432257583,
      "loss": 1.7584,
      "step": 450
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.6640499830245972,
      "learning_rate": 0.00024169443399637303,
      "loss": 1.7288,
      "step": 451
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.7537161111831665,
      "learning_rate": 0.00024164757638020032,
      "loss": 1.8502,
      "step": 452
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.614371657371521,
      "learning_rate": 0.00024160059152516982,
      "loss": 1.8407,
      "step": 453
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.3671376705169678,
      "learning_rate": 0.0002415534794825325,
      "loss": 1.7035,
      "step": 454
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.1287087202072144,
      "learning_rate": 0.00024150624030367797,
      "loss": 0.8822,
      "step": 455
    },
    {
      "epoch": 0.44,
      "grad_norm": 2.104029893875122,
      "learning_rate": 0.0002414588740401347,
      "loss": 2.4046,
      "step": 456
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.5074779987335205,
      "learning_rate": 0.00024141138074356962,
      "loss": 1.4848,
      "step": 457
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.8815460205078125,
      "learning_rate": 0.00024136376046578835,
      "loss": 1.759,
      "step": 458
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.5019757747650146,
      "learning_rate": 0.00024131601325873487,
      "loss": 1.4991,
      "step": 459
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.6193299293518066,
      "learning_rate": 0.00024126813917449175,
      "loss": 1.818,
      "step": 460
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.4965165853500366,
      "learning_rate": 0.0002412201382652799,
      "loss": 1.4364,
      "step": 461
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.6111191511154175,
      "learning_rate": 0.00024117201058345862,
      "loss": 1.7969,
      "step": 462
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.6606050729751587,
      "learning_rate": 0.00024112375618152537,
      "loss": 1.8442,
      "step": 463
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.7113983631134033,
      "learning_rate": 0.00024107537511211603,
      "loss": 1.6806,
      "step": 464
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.5785653591156006,
      "learning_rate": 0.00024102686742800446,
      "loss": 1.6221,
      "step": 465
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.715955138206482,
      "learning_rate": 0.0002409782331821027,
      "loss": 1.8423,
      "step": 466
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.7139517068862915,
      "learning_rate": 0.00024092947242746094,
      "loss": 1.9045,
      "step": 467
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.2441296577453613,
      "learning_rate": 0.00024088058521726718,
      "loss": 1.809,
      "step": 468
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.487850546836853,
      "learning_rate": 0.00024083157160484752,
      "loss": 1.3382,
      "step": 469
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.3797458410263062,
      "learning_rate": 0.00024078243164366586,
      "loss": 1.3186,
      "step": 470
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.5861365795135498,
      "learning_rate": 0.0002407331653873239,
      "loss": 2.0695,
      "step": 471
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.4714101552963257,
      "learning_rate": 0.00024068377288956116,
      "loss": 1.8704,
      "step": 472
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.6222436428070068,
      "learning_rate": 0.00024063425420425485,
      "loss": 2.0155,
      "step": 473
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.7147741317749023,
      "learning_rate": 0.00024058460938541982,
      "loss": 2.2635,
      "step": 474
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.7251660823822021,
      "learning_rate": 0.0002405348384872085,
      "loss": 2.3124,
      "step": 475
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.4105515480041504,
      "learning_rate": 0.00024048494156391086,
      "loss": 1.5038,
      "step": 476
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.4747495651245117,
      "learning_rate": 0.0002404349186699543,
      "loss": 1.8607,
      "step": 477
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.4962126016616821,
      "learning_rate": 0.00024038476985990364,
      "loss": 1.9751,
      "step": 478
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.5216854810714722,
      "learning_rate": 0.00024033449518846112,
      "loss": 1.6452,
      "step": 479
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.7051867246627808,
      "learning_rate": 0.0002402840947104662,
      "loss": 1.5337,
      "step": 480
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.6586822271347046,
      "learning_rate": 0.00024023356848089557,
      "loss": 1.5478,
      "step": 481
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.2107691764831543,
      "learning_rate": 0.00024018291655486306,
      "loss": 1.278,
      "step": 482
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.4238454103469849,
      "learning_rate": 0.00024013213898761975,
      "loss": 1.6255,
      "step": 483
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.3502719402313232,
      "learning_rate": 0.0002400812358345536,
      "loss": 1.3436,
      "step": 484
    },
    {
      "epoch": 0.47,
      "grad_norm": 2.2329463958740234,
      "learning_rate": 0.00024003020715118967,
      "loss": 2.3051,
      "step": 485
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.6358450651168823,
      "learning_rate": 0.00023997905299318983,
      "loss": 2.2955,
      "step": 486
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.5185836553573608,
      "learning_rate": 0.000239927773416353,
      "loss": 1.5445,
      "step": 487
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.5232542753219604,
      "learning_rate": 0.00023987636847661476,
      "loss": 1.6744,
      "step": 488
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.4239147901535034,
      "learning_rate": 0.00023982483823004747,
      "loss": 1.666,
      "step": 489
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.3807685375213623,
      "learning_rate": 0.00023977318273286018,
      "loss": 1.5856,
      "step": 490
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.573913812637329,
      "learning_rate": 0.00023972140204139858,
      "loss": 1.9832,
      "step": 491
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.556381344795227,
      "learning_rate": 0.0002396694962121449,
      "loss": 1.4315,
      "step": 492
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.7060798406600952,
      "learning_rate": 0.00023961746530171788,
      "loss": 1.7671,
      "step": 493
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.5967423915863037,
      "learning_rate": 0.0002395653093668727,
      "loss": 1.9734,
      "step": 494
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.390377402305603,
      "learning_rate": 0.00023951302846450087,
      "loss": 1.4402,
      "step": 495
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.330527663230896,
      "learning_rate": 0.00023946062265163033,
      "loss": 1.591,
      "step": 496
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.6965402364730835,
      "learning_rate": 0.00023940809198542512,
      "loss": 2.3242,
      "step": 497
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.51449716091156,
      "learning_rate": 0.0002393554365231856,
      "loss": 1.619,
      "step": 498
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.6005566120147705,
      "learning_rate": 0.00023930265632234816,
      "loss": 2.0559,
      "step": 499
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.4369415044784546,
      "learning_rate": 0.00023924975144048533,
      "loss": 1.4451,
      "step": 500
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.4336689710617065,
      "learning_rate": 0.00023919672193530565,
      "loss": 1.7534,
      "step": 501
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.449487566947937,
      "learning_rate": 0.00023914356786465348,
      "loss": 1.8109,
      "step": 502
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.439719319343567,
      "learning_rate": 0.00023909028928650918,
      "loss": 1.6258,
      "step": 503
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.702451229095459,
      "learning_rate": 0.00023903688625898887,
      "loss": 1.9753,
      "step": 504
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.1217689514160156,
      "learning_rate": 0.00023898335884034444,
      "loss": 1.0806,
      "step": 505
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.43269944190979,
      "learning_rate": 0.00023892970708896342,
      "loss": 2.2551,
      "step": 506
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.3758916854858398,
      "learning_rate": 0.00023887593106336903,
      "loss": 1.6455,
      "step": 507
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.348081350326538,
      "learning_rate": 0.00023882203082221996,
      "loss": 1.6481,
      "step": 508
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.44756019115448,
      "learning_rate": 0.00023876800642431048,
      "loss": 1.6608,
      "step": 509
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.5149917602539062,
      "learning_rate": 0.00023871385792857027,
      "loss": 1.6434,
      "step": 510
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.436800241470337,
      "learning_rate": 0.00023865958539406434,
      "loss": 1.9541,
      "step": 511
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.563488483428955,
      "learning_rate": 0.00023860518887999305,
      "loss": 1.921,
      "step": 512
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.2544896602630615,
      "learning_rate": 0.00023855066844569194,
      "loss": 1.285,
      "step": 513
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.1815474033355713,
      "learning_rate": 0.00023849602415063176,
      "loss": 1.5958,
      "step": 514
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.3179055452346802,
      "learning_rate": 0.00023844125605441835,
      "loss": 1.5887,
      "step": 515
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.3224471807479858,
      "learning_rate": 0.0002383863642167926,
      "loss": 1.3453,
      "step": 516
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.3743826150894165,
      "learning_rate": 0.00023833134869763041,
      "loss": 1.6145,
      "step": 517
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.7813752889633179,
      "learning_rate": 0.00023827620955694248,
      "loss": 1.5723,
      "step": 518
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.457780361175537,
      "learning_rate": 0.00023822094685487451,
      "loss": 1.5773,
      "step": 519
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.4680794477462769,
      "learning_rate": 0.00023816556065170693,
      "loss": 1.7291,
      "step": 520
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.5272331237792969,
      "learning_rate": 0.00023811005100785471,
      "loss": 1.6012,
      "step": 521
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.6784194707870483,
      "learning_rate": 0.00023805441798386778,
      "loss": 1.6109,
      "step": 522
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.5598907470703125,
      "learning_rate": 0.00023799866164043044,
      "loss": 2.0627,
      "step": 523
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.470335841178894,
      "learning_rate": 0.0002379427820383615,
      "loss": 1.8267,
      "step": 524
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.677306890487671,
      "learning_rate": 0.00023788677923861434,
      "loss": 2.2845,
      "step": 525
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.2650090456008911,
      "learning_rate": 0.00023783065330227666,
      "loss": 1.6407,
      "step": 526
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.6766515970230103,
      "learning_rate": 0.00023777440429057043,
      "loss": 1.7039,
      "step": 527
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.3705694675445557,
      "learning_rate": 0.00023771803226485197,
      "loss": 1.8251,
      "step": 528
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.3645520210266113,
      "learning_rate": 0.00023766153728661173,
      "loss": 1.0754,
      "step": 529
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.5679587125778198,
      "learning_rate": 0.00023760491941747419,
      "loss": 1.7131,
      "step": 530
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.4037436246871948,
      "learning_rate": 0.0002375481787191981,
      "loss": 1.7182,
      "step": 531
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.4753473997116089,
      "learning_rate": 0.000237491315253676,
      "loss": 1.9983,
      "step": 532
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.307569980621338,
      "learning_rate": 0.00023743432908293437,
      "loss": 1.4774,
      "step": 533
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.5248976945877075,
      "learning_rate": 0.00023737722026913357,
      "loss": 1.678,
      "step": 534
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.6276731491088867,
      "learning_rate": 0.00023731998887456778,
      "loss": 1.9258,
      "step": 535
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.5609790086746216,
      "learning_rate": 0.0002372626349616649,
      "loss": 1.7404,
      "step": 536
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.3915233612060547,
      "learning_rate": 0.0002372051585929863,
      "loss": 1.5489,
      "step": 537
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.6098966598510742,
      "learning_rate": 0.0002371475598312271,
      "loss": 1.7612,
      "step": 538
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.3052818775177002,
      "learning_rate": 0.00023708983873921586,
      "loss": 1.2804,
      "step": 539
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.3293620347976685,
      "learning_rate": 0.00023703199537991466,
      "loss": 1.4353,
      "step": 540
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.4364746809005737,
      "learning_rate": 0.00023697402981641877,
      "loss": 1.4233,
      "step": 541
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.7351473569869995,
      "learning_rate": 0.00023691594211195695,
      "loss": 1.9575,
      "step": 542
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.3488695621490479,
      "learning_rate": 0.00023685773232989105,
      "loss": 2.0196,
      "step": 543
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.455739140510559,
      "learning_rate": 0.0002367994005337162,
      "loss": 2.1026,
      "step": 544
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.2512130737304688,
      "learning_rate": 0.00023674094678706052,
      "loss": 1.3881,
      "step": 545
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.6788513660430908,
      "learning_rate": 0.0002366823711536852,
      "loss": 1.5795,
      "step": 546
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.0756875276565552,
      "learning_rate": 0.00023662367369748442,
      "loss": 1.3518,
      "step": 547
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.5534878969192505,
      "learning_rate": 0.00023656485448248516,
      "loss": 1.4272,
      "step": 548
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.276198387145996,
      "learning_rate": 0.00023650591357284728,
      "loss": 1.3384,
      "step": 549
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.4616369009017944,
      "learning_rate": 0.00023644685103286337,
      "loss": 1.6083,
      "step": 550
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.487240195274353,
      "learning_rate": 0.00023638766692695869,
      "loss": 1.2336,
      "step": 551
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.3095462322235107,
      "learning_rate": 0.00023632836131969105,
      "loss": 1.3381,
      "step": 552
    },
    {
      "epoch": 0.53,
      "grad_norm": 2.327471971511841,
      "learning_rate": 0.0002362689342757509,
      "loss": 2.192,
      "step": 553
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.9238693714141846,
      "learning_rate": 0.00023620938585996106,
      "loss": 1.6526,
      "step": 554
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.6897307634353638,
      "learning_rate": 0.00023614971613727684,
      "loss": 1.6713,
      "step": 555
    },
    {
      "epoch": 0.54,
      "grad_norm": 2.526841163635254,
      "learning_rate": 0.00023608992517278578,
      "loss": 2.0541,
      "step": 556
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.5254669189453125,
      "learning_rate": 0.00023603001303170763,
      "loss": 1.3508,
      "step": 557
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.57151198387146,
      "learning_rate": 0.0002359699797793945,
      "loss": 1.3197,
      "step": 558
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.5244841575622559,
      "learning_rate": 0.00023590982548133045,
      "loss": 1.4338,
      "step": 559
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.4043896198272705,
      "learning_rate": 0.0002358495502031316,
      "loss": 1.9264,
      "step": 560
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.4733903408050537,
      "learning_rate": 0.00023578915401054607,
      "loss": 1.59,
      "step": 561
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.7210224866867065,
      "learning_rate": 0.0002357286369694539,
      "loss": 1.7421,
      "step": 562
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.291479468345642,
      "learning_rate": 0.00023566799914586688,
      "loss": 0.9787,
      "step": 563
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.6497067213058472,
      "learning_rate": 0.00023560724060592858,
      "loss": 1.5709,
      "step": 564
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.6274982690811157,
      "learning_rate": 0.00023554636141591426,
      "loss": 1.5853,
      "step": 565
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.4311622381210327,
      "learning_rate": 0.00023548536164223077,
      "loss": 1.7003,
      "step": 566
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.6563549041748047,
      "learning_rate": 0.00023542424135141656,
      "loss": 1.4655,
      "step": 567
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.6957557201385498,
      "learning_rate": 0.00023536300061014132,
      "loss": 1.8792,
      "step": 568
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.2087243795394897,
      "learning_rate": 0.00023530163948520645,
      "loss": 1.262,
      "step": 569
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.3456618785858154,
      "learning_rate": 0.00023524015804354437,
      "loss": 1.6861,
      "step": 570
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.5029468536376953,
      "learning_rate": 0.00023517855635221896,
      "loss": 1.7885,
      "step": 571
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.4606847763061523,
      "learning_rate": 0.00023511683447842514,
      "loss": 1.4353,
      "step": 572
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.4862562417984009,
      "learning_rate": 0.00023505499248948898,
      "loss": 1.7917,
      "step": 573
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.3696837425231934,
      "learning_rate": 0.00023499303045286751,
      "loss": 1.2735,
      "step": 574
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.4719946384429932,
      "learning_rate": 0.00023493094843614877,
      "loss": 1.867,
      "step": 575
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.2641010284423828,
      "learning_rate": 0.0002348687465070517,
      "loss": 1.3491,
      "step": 576
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.2275723218917847,
      "learning_rate": 0.00023480642473342592,
      "loss": 1.5747,
      "step": 577
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.214805245399475,
      "learning_rate": 0.00023474398318325192,
      "loss": 1.5349,
      "step": 578
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.4233876466751099,
      "learning_rate": 0.0002346814219246407,
      "loss": 2.0182,
      "step": 579
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.363861322402954,
      "learning_rate": 0.00023461874102583395,
      "loss": 1.8423,
      "step": 580
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.2210170030593872,
      "learning_rate": 0.00023455594055520385,
      "loss": 1.1782,
      "step": 581
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.3371609449386597,
      "learning_rate": 0.0002344930205812529,
      "loss": 1.7994,
      "step": 582
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.424874186515808,
      "learning_rate": 0.00023442998117261416,
      "loss": 1.637,
      "step": 583
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.1814637184143066,
      "learning_rate": 0.00023436682239805074,
      "loss": 1.334,
      "step": 584
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.3004404306411743,
      "learning_rate": 0.00023430354432645606,
      "loss": 1.1703,
      "step": 585
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.332945704460144,
      "learning_rate": 0.00023424014702685375,
      "loss": 1.3147,
      "step": 586
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.5192474126815796,
      "learning_rate": 0.00023417663056839733,
      "loss": 1.2995,
      "step": 587
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.5328348875045776,
      "learning_rate": 0.00023411299502037043,
      "loss": 1.4266,
      "step": 588
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.7186061143875122,
      "learning_rate": 0.00023404924045218652,
      "loss": 1.2097,
      "step": 589
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.892555832862854,
      "learning_rate": 0.0002339853669333889,
      "loss": 1.9445,
      "step": 590
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.4742870330810547,
      "learning_rate": 0.0002339213745336506,
      "loss": 1.4,
      "step": 591
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.3957617282867432,
      "learning_rate": 0.00023385726332277441,
      "loss": 1.0397,
      "step": 592
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.274652123451233,
      "learning_rate": 0.00023379303337069262,
      "loss": 1.5659,
      "step": 593
    },
    {
      "epoch": 0.57,
      "grad_norm": 2.0992603302001953,
      "learning_rate": 0.00023372868474746714,
      "loss": 1.8078,
      "step": 594
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.4869073629379272,
      "learning_rate": 0.00023366421752328922,
      "loss": 1.3452,
      "step": 595
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.4044032096862793,
      "learning_rate": 0.00023359963176847957,
      "loss": 1.3954,
      "step": 596
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.3737423419952393,
      "learning_rate": 0.0002335349275534881,
      "loss": 1.2106,
      "step": 597
    },
    {
      "epoch": 0.58,
      "grad_norm": 2.022343873977661,
      "learning_rate": 0.00023347010494889406,
      "loss": 1.8948,
      "step": 598
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.3906071186065674,
      "learning_rate": 0.00023340516402540573,
      "loss": 1.5172,
      "step": 599
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.825240135192871,
      "learning_rate": 0.0002333401048538605,
      "loss": 1.0034,
      "step": 600
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.4067100286483765,
      "learning_rate": 0.00023327492750522475,
      "loss": 1.7208,
      "step": 601
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.222434401512146,
      "learning_rate": 0.0002332096320505937,
      "loss": 1.1459,
      "step": 602
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.7354257106781006,
      "learning_rate": 0.00023314421856119154,
      "loss": 1.5793,
      "step": 603
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.384118914604187,
      "learning_rate": 0.00023307868710837105,
      "loss": 1.0095,
      "step": 604
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.459686040878296,
      "learning_rate": 0.00023301303776361378,
      "loss": 1.528,
      "step": 605
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.4759612083435059,
      "learning_rate": 0.0002329472705985299,
      "loss": 1.378,
      "step": 606
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.5465447902679443,
      "learning_rate": 0.00023288138568485802,
      "loss": 1.5053,
      "step": 607
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.2388064861297607,
      "learning_rate": 0.0002328153830944652,
      "loss": 1.2588,
      "step": 608
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.4053964614868164,
      "learning_rate": 0.00023274926289934688,
      "loss": 1.014,
      "step": 609
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.5368363857269287,
      "learning_rate": 0.00023268302517162688,
      "loss": 1.6706,
      "step": 610
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.3434786796569824,
      "learning_rate": 0.00023261666998355702,
      "loss": 1.1686,
      "step": 611
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.3576295375823975,
      "learning_rate": 0.00023255019740751743,
      "loss": 1.5393,
      "step": 612
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.285962700843811,
      "learning_rate": 0.0002324836075160162,
      "loss": 1.7462,
      "step": 613
    },
    {
      "epoch": 0.59,
      "grad_norm": 2.13918399810791,
      "learning_rate": 0.0002324169003816894,
      "loss": 1.4822,
      "step": 614
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.7046306133270264,
      "learning_rate": 0.0002323500760773009,
      "loss": 1.1588,
      "step": 615
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.6164623498916626,
      "learning_rate": 0.00023228313467574263,
      "loss": 1.9175,
      "step": 616
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.4854605197906494,
      "learning_rate": 0.00023221607625003402,
      "loss": 1.7528,
      "step": 617
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.6400645971298218,
      "learning_rate": 0.00023214890087332218,
      "loss": 1.9289,
      "step": 618
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.5701171159744263,
      "learning_rate": 0.00023208160861888186,
      "loss": 1.3207,
      "step": 619
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.5051968097686768,
      "learning_rate": 0.00023201419956011526,
      "loss": 1.4028,
      "step": 620
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.1874357461929321,
      "learning_rate": 0.000231946673770552,
      "loss": 1.3708,
      "step": 621
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.3110204935073853,
      "learning_rate": 0.00023187903132384903,
      "loss": 1.3477,
      "step": 622
    },
    {
      "epoch": 0.6,
      "eval_loss": 1.6304495334625244,
      "eval_runtime": 28.33,
      "eval_samples_per_second": 3.0,
      "eval_steps_per_second": 1.518,
      "step": 622
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.3583319187164307,
      "learning_rate": 0.00023181127229379058,
      "loss": 1.6669,
      "step": 623
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.400608777999878,
      "learning_rate": 0.00023174339675428792,
      "loss": 1.1634,
      "step": 624
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.403623342514038,
      "learning_rate": 0.0002316754047793796,
      "loss": 1.4847,
      "step": 625
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.511208176612854,
      "learning_rate": 0.00023160729644323099,
      "loss": 1.3538,
      "step": 626
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.1673988103866577,
      "learning_rate": 0.00023153907182013456,
      "loss": 1.1773,
      "step": 627
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.5797700881958008,
      "learning_rate": 0.00023147073098450953,
      "loss": 1.6715,
      "step": 628
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.7258118391036987,
      "learning_rate": 0.00023140227401090185,
      "loss": 1.9853,
      "step": 629
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.3571829795837402,
      "learning_rate": 0.00023133370097398425,
      "loss": 1.624,
      "step": 630
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.462410807609558,
      "learning_rate": 0.00023126501194855597,
      "loss": 1.0793,
      "step": 631
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.2532066106796265,
      "learning_rate": 0.00023119620700954286,
      "loss": 1.3601,
      "step": 632
    },
    {
      "epoch": 0.61,
      "grad_norm": 2.077949285507202,
      "learning_rate": 0.0002311272862319971,
      "loss": 1.9824,
      "step": 633
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.5285255908966064,
      "learning_rate": 0.00023105824969109736,
      "loss": 1.3991,
      "step": 634
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.544295310974121,
      "learning_rate": 0.0002309890974621484,
      "loss": 1.8974,
      "step": 635
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.4555768966674805,
      "learning_rate": 0.00023091982962058138,
      "loss": 1.4607,
      "step": 636
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.2902973890304565,
      "learning_rate": 0.00023085044624195344,
      "loss": 1.7837,
      "step": 637
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.6322664022445679,
      "learning_rate": 0.0002307809474019478,
      "loss": 1.3603,
      "step": 638
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.295804738998413,
      "learning_rate": 0.00023071133317637357,
      "loss": 1.3386,
      "step": 639
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.3429510593414307,
      "learning_rate": 0.0002306416036411658,
      "loss": 1.2507,
      "step": 640
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.6754704713821411,
      "learning_rate": 0.0002305717588723852,
      "loss": 1.4455,
      "step": 641
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.1602115631103516,
      "learning_rate": 0.00023050179894621835,
      "loss": 1.6829,
      "step": 642
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.412318468093872,
      "learning_rate": 0.00023043172393897728,
      "loss": 1.5495,
      "step": 643
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.4912530183792114,
      "learning_rate": 0.00023036153392709963,
      "loss": 1.6909,
      "step": 644
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.4303503036499023,
      "learning_rate": 0.0002302912289871485,
      "loss": 1.5794,
      "step": 645
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.3841469287872314,
      "learning_rate": 0.00023022080919581227,
      "loss": 1.9355,
      "step": 646
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.5403145551681519,
      "learning_rate": 0.00023015027462990474,
      "loss": 1.6106,
      "step": 647
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.370869517326355,
      "learning_rate": 0.00023007962536636474,
      "loss": 1.4042,
      "step": 648
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.6991840600967407,
      "learning_rate": 0.00023000886148225635,
      "loss": 1.5327,
      "step": 649
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.7101843357086182,
      "learning_rate": 0.0002299379830547686,
      "loss": 1.4779,
      "step": 650
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.3180490732192993,
      "learning_rate": 0.00022986699016121545,
      "loss": 1.5081,
      "step": 651
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.180501937866211,
      "learning_rate": 0.0002297958828790358,
      "loss": 1.0716,
      "step": 652
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.322432279586792,
      "learning_rate": 0.00022972466128579327,
      "loss": 1.277,
      "step": 653
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.6292730569839478,
      "learning_rate": 0.00022965332545917623,
      "loss": 1.631,
      "step": 654
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.7111241817474365,
      "learning_rate": 0.00022958187547699748,
      "loss": 1.3783,
      "step": 655
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.3672375679016113,
      "learning_rate": 0.0002295103114171946,
      "loss": 1.0966,
      "step": 656
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.34515380859375,
      "learning_rate": 0.00022943863335782937,
      "loss": 1.1398,
      "step": 657
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.2631828784942627,
      "learning_rate": 0.00022936684137708806,
      "loss": 1.3126,
      "step": 658
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.538662314414978,
      "learning_rate": 0.00022929493555328117,
      "loss": 1.7226,
      "step": 659
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.458886981010437,
      "learning_rate": 0.00022922291596484337,
      "loss": 1.5284,
      "step": 660
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.525400996208191,
      "learning_rate": 0.00022915078269033346,
      "loss": 1.3691,
      "step": 661
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.4852648973464966,
      "learning_rate": 0.00022907853580843415,
      "loss": 1.6369,
      "step": 662
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.523781418800354,
      "learning_rate": 0.0002290061753979522,
      "loss": 1.4643,
      "step": 663
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.3667558431625366,
      "learning_rate": 0.00022893370153781817,
      "loss": 1.0672,
      "step": 664
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.8203182220458984,
      "learning_rate": 0.00022886111430708627,
      "loss": 2.1031,
      "step": 665
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.458509922027588,
      "learning_rate": 0.00022878841378493452,
      "loss": 1.5407,
      "step": 666
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.1977659463882446,
      "learning_rate": 0.0002287156000506644,
      "loss": 1.5075,
      "step": 667
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.1007845401763916,
      "learning_rate": 0.00022864267318370096,
      "loss": 0.9576,
      "step": 668
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.6353801488876343,
      "learning_rate": 0.00022856963326359266,
      "loss": 1.7101,
      "step": 669
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.335712194442749,
      "learning_rate": 0.00022849648037001115,
      "loss": 1.0678,
      "step": 670
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.375508189201355,
      "learning_rate": 0.0002284232145827515,
      "loss": 1.4913,
      "step": 671
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.211701512336731,
      "learning_rate": 0.00022834983598173182,
      "loss": 1.0438,
      "step": 672
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.7795249223709106,
      "learning_rate": 0.00022827634464699323,
      "loss": 1.9137,
      "step": 673
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.4882405996322632,
      "learning_rate": 0.00022820274065869992,
      "loss": 1.5704,
      "step": 674
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.3354179859161377,
      "learning_rate": 0.00022812902409713893,
      "loss": 1.3485,
      "step": 675
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.5397255420684814,
      "learning_rate": 0.00022805519504272005,
      "loss": 2.0343,
      "step": 676
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.4044562578201294,
      "learning_rate": 0.00022798125357597582,
      "loss": 1.7446,
      "step": 677
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.284012794494629,
      "learning_rate": 0.00022790719977756142,
      "loss": 0.9859,
      "step": 678
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.879234790802002,
      "learning_rate": 0.00022783303372825447,
      "loss": 1.7374,
      "step": 679
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.278829574584961,
      "learning_rate": 0.00022775875550895517,
      "loss": 1.4937,
      "step": 680
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.3762723207473755,
      "learning_rate": 0.00022768436520068596,
      "loss": 1.1091,
      "step": 681
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.4809486865997314,
      "learning_rate": 0.00022760986288459156,
      "loss": 1.6401,
      "step": 682
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.4666237831115723,
      "learning_rate": 0.00022753524864193892,
      "loss": 1.7877,
      "step": 683
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.3258100748062134,
      "learning_rate": 0.00022746052255411708,
      "loss": 1.2407,
      "step": 684
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.178125023841858,
      "learning_rate": 0.000227385684702637,
      "loss": 1.4674,
      "step": 685
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.3732388019561768,
      "learning_rate": 0.00022731073516913164,
      "loss": 1.4729,
      "step": 686
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.357108235359192,
      "learning_rate": 0.00022723567403535572,
      "loss": 1.428,
      "step": 687
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.293455719947815,
      "learning_rate": 0.00022716050138318574,
      "loss": 1.3115,
      "step": 688
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.1836376190185547,
      "learning_rate": 0.00022708521729461985,
      "loss": 1.2031,
      "step": 689
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.3905551433563232,
      "learning_rate": 0.00022700982185177765,
      "loss": 1.8465,
      "step": 690
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.2469663619995117,
      "learning_rate": 0.00022693431513690033,
      "loss": 1.5076,
      "step": 691
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.2081923484802246,
      "learning_rate": 0.00022685869723235042,
      "loss": 1.5242,
      "step": 692
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.2785173654556274,
      "learning_rate": 0.00022678296822061173,
      "loss": 1.3025,
      "step": 693
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.3872441053390503,
      "learning_rate": 0.00022670712818428923,
      "loss": 1.7145,
      "step": 694
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.1730290651321411,
      "learning_rate": 0.00022663117720610903,
      "loss": 1.1832,
      "step": 695
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.0906507968902588,
      "learning_rate": 0.00022655511536891825,
      "loss": 1.378,
      "step": 696
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.3207216262817383,
      "learning_rate": 0.00022647894275568492,
      "loss": 1.2161,
      "step": 697
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.4555737972259521,
      "learning_rate": 0.00022640265944949797,
      "loss": 1.4499,
      "step": 698
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.3717491626739502,
      "learning_rate": 0.000226326265533567,
      "loss": 1.7293,
      "step": 699
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.2536697387695312,
      "learning_rate": 0.00022624976109122225,
      "loss": 1.4178,
      "step": 700
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.550447940826416,
      "learning_rate": 0.0002261731462059146,
      "loss": 1.4674,
      "step": 701
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.2245453596115112,
      "learning_rate": 0.0002260964209612153,
      "loss": 1.2625,
      "step": 702
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.2388466596603394,
      "learning_rate": 0.0002260195854408161,
      "loss": 1.6174,
      "step": 703
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.4022188186645508,
      "learning_rate": 0.00022594263972852897,
      "loss": 1.0605,
      "step": 704
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.275978922843933,
      "learning_rate": 0.00022586558390828604,
      "loss": 1.3885,
      "step": 705
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.493241786956787,
      "learning_rate": 0.0002257884180641396,
      "loss": 1.5634,
      "step": 706
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.1711384057998657,
      "learning_rate": 0.00022571114228026198,
      "loss": 1.2732,
      "step": 707
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.2927380800247192,
      "learning_rate": 0.0002256337566409454,
      "loss": 1.4771,
      "step": 708
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.2375295162200928,
      "learning_rate": 0.0002255562612306018,
      "loss": 0.9908,
      "step": 709
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.1654285192489624,
      "learning_rate": 0.00022547865613376308,
      "loss": 0.9158,
      "step": 710
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.694615125656128,
      "learning_rate": 0.0002254009414350806,
      "loss": 1.517,
      "step": 711
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.110138177871704,
      "learning_rate": 0.0002253231172193254,
      "loss": 1.2596,
      "step": 712
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.5078182220458984,
      "learning_rate": 0.00022524518357138788,
      "loss": 1.6952,
      "step": 713
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.7366492748260498,
      "learning_rate": 0.00022516714057627783,
      "loss": 1.8556,
      "step": 714
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.4952809810638428,
      "learning_rate": 0.0002250889883191244,
      "loss": 1.5153,
      "step": 715
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.2906392812728882,
      "learning_rate": 0.00022501072688517576,
      "loss": 1.0992,
      "step": 716
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.26261568069458,
      "learning_rate": 0.0002249323563597994,
      "loss": 1.1001,
      "step": 717
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.437803030014038,
      "learning_rate": 0.00022485387682848155,
      "loss": 1.519,
      "step": 718
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.454239845275879,
      "learning_rate": 0.00022477528837682755,
      "loss": 1.258,
      "step": 719
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.403212547302246,
      "learning_rate": 0.00022469659109056143,
      "loss": 1.6505,
      "step": 720
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.2380965948104858,
      "learning_rate": 0.000224617785055526,
      "loss": 1.3611,
      "step": 721
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.41748046875,
      "learning_rate": 0.00022453887035768266,
      "loss": 1.8791,
      "step": 722
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.4731191396713257,
      "learning_rate": 0.00022445984708311132,
      "loss": 1.3579,
      "step": 723
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.688100814819336,
      "learning_rate": 0.00022438071531801038,
      "loss": 1.5332,
      "step": 724
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.7024562358856201,
      "learning_rate": 0.00022430147514869653,
      "loss": 1.7091,
      "step": 725
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.3762348890304565,
      "learning_rate": 0.00022422212666160478,
      "loss": 1.7395,
      "step": 726
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.4019901752471924,
      "learning_rate": 0.0002241426699432882,
      "loss": 1.0866,
      "step": 727
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.861675500869751,
      "learning_rate": 0.00022406310508041796,
      "loss": 2.1158,
      "step": 728
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.3458441495895386,
      "learning_rate": 0.00022398343215978322,
      "loss": 1.4403,
      "step": 729
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.5659409761428833,
      "learning_rate": 0.000223903651268291,
      "loss": 1.7634,
      "step": 730
    },
    {
      "epoch": 0.71,
      "grad_norm": 1.2884780168533325,
      "learning_rate": 0.00022382376249296598,
      "loss": 1.5543,
      "step": 731
    },
    {
      "epoch": 0.71,
      "grad_norm": 1.0735148191452026,
      "learning_rate": 0.00022374376592095076,
      "loss": 1.2122,
      "step": 732
    },
    {
      "epoch": 0.71,
      "grad_norm": 1.3894761800765991,
      "learning_rate": 0.00022366366163950523,
      "loss": 1.7313,
      "step": 733
    },
    {
      "epoch": 0.71,
      "grad_norm": 1.186671257019043,
      "learning_rate": 0.00022358344973600705,
      "loss": 0.9415,
      "step": 734
    },
    {
      "epoch": 0.71,
      "grad_norm": 1.2778778076171875,
      "learning_rate": 0.00022350313029795112,
      "loss": 1.871,
      "step": 735
    },
    {
      "epoch": 0.71,
      "grad_norm": 1.7073333263397217,
      "learning_rate": 0.0002234227034129496,
      "loss": 1.7262,
      "step": 736
    },
    {
      "epoch": 0.71,
      "grad_norm": 1.0448979139328003,
      "learning_rate": 0.00022334216916873196,
      "loss": 0.8838,
      "step": 737
    },
    {
      "epoch": 0.71,
      "grad_norm": 2.0367178916931152,
      "learning_rate": 0.00022326152765314472,
      "loss": 1.5082,
      "step": 738
    },
    {
      "epoch": 0.71,
      "grad_norm": 1.5965039730072021,
      "learning_rate": 0.0002231807789541515,
      "loss": 1.6947,
      "step": 739
    },
    {
      "epoch": 0.71,
      "grad_norm": 1.2586615085601807,
      "learning_rate": 0.00022309992315983263,
      "loss": 1.2724,
      "step": 740
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.498313546180725,
      "learning_rate": 0.0002230189603583855,
      "loss": 1.6083,
      "step": 741
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.2010595798492432,
      "learning_rate": 0.00022293789063812403,
      "loss": 1.0541,
      "step": 742
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.2845265865325928,
      "learning_rate": 0.00022285671408747888,
      "loss": 0.9787,
      "step": 743
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.5691664218902588,
      "learning_rate": 0.00022277543079499723,
      "loss": 1.9072,
      "step": 744
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.8554484844207764,
      "learning_rate": 0.00022269404084934269,
      "loss": 1.6881,
      "step": 745
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.4981218576431274,
      "learning_rate": 0.00022261254433929514,
      "loss": 1.2664,
      "step": 746
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.5050944089889526,
      "learning_rate": 0.00022253094135375075,
      "loss": 1.3197,
      "step": 747
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.4656778573989868,
      "learning_rate": 0.0002224492319817219,
      "loss": 1.5641,
      "step": 748
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.433406114578247,
      "learning_rate": 0.0002223674163123369,
      "loss": 1.6133,
      "step": 749
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.4999152421951294,
      "learning_rate": 0.00022228549443484004,
      "loss": 1.1804,
      "step": 750
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.386148452758789,
      "learning_rate": 0.00022220346643859153,
      "loss": 1.4288,
      "step": 751
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.305163025856018,
      "learning_rate": 0.00022212133241306722,
      "loss": 0.9577,
      "step": 752
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.3146147727966309,
      "learning_rate": 0.00022203909244785874,
      "loss": 1.4475,
      "step": 753
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.5211604833602905,
      "learning_rate": 0.00022195674663267324,
      "loss": 1.7606,
      "step": 754
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.4717007875442505,
      "learning_rate": 0.0002218742950573332,
      "loss": 1.638,
      "step": 755
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.1527776718139648,
      "learning_rate": 0.0002217917378117767,
      "loss": 1.3322,
      "step": 756
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.599092721939087,
      "learning_rate": 0.00022170907498605686,
      "loss": 1.6299,
      "step": 757
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.2002819776535034,
      "learning_rate": 0.0002216263066703421,
      "loss": 1.3424,
      "step": 758
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.3727798461914062,
      "learning_rate": 0.0002215434329549159,
      "loss": 1.7718,
      "step": 759
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.1956920623779297,
      "learning_rate": 0.0002214604539301767,
      "loss": 0.7773,
      "step": 760
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.3804597854614258,
      "learning_rate": 0.0002213773696866377,
      "loss": 1.3221,
      "step": 761
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.134540319442749,
      "learning_rate": 0.00022129418031492705,
      "loss": 1.1406,
      "step": 762
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.599867582321167,
      "learning_rate": 0.00022121088590578743,
      "loss": 1.5812,
      "step": 763
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.1916415691375732,
      "learning_rate": 0.0002211274865500762,
      "loss": 1.1639,
      "step": 764
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.4034931659698486,
      "learning_rate": 0.00022104398233876516,
      "loss": 1.6485,
      "step": 765
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.217153549194336,
      "learning_rate": 0.0002209603733629404,
      "loss": 1.2622,
      "step": 766
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.2423417568206787,
      "learning_rate": 0.00022087665971380244,
      "loss": 1.3384,
      "step": 767
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.2787408828735352,
      "learning_rate": 0.00022079284148266587,
      "loss": 1.3278,
      "step": 768
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.0672158002853394,
      "learning_rate": 0.0002207089187609594,
      "loss": 1.4183,
      "step": 769
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.4464133977890015,
      "learning_rate": 0.00022062489164022565,
      "loss": 2.0707,
      "step": 770
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.4786409139633179,
      "learning_rate": 0.00022054076021212128,
      "loss": 1.3817,
      "step": 771
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.325783133506775,
      "learning_rate": 0.00022045652456841651,
      "loss": 1.2509,
      "step": 772
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.3716827630996704,
      "learning_rate": 0.0002203721848009954,
      "loss": 1.5183,
      "step": 773
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.4279142618179321,
      "learning_rate": 0.00022028774100185555,
      "loss": 2.0254,
      "step": 774
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.3607648611068726,
      "learning_rate": 0.00022020319326310805,
      "loss": 1.5698,
      "step": 775
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.3437821865081787,
      "learning_rate": 0.00022011854167697725,
      "loss": 1.5756,
      "step": 776
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.1076477766036987,
      "learning_rate": 0.00022003378633580098,
      "loss": 1.0518,
      "step": 777
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.1801143884658813,
      "learning_rate": 0.00021994892733203007,
      "loss": 1.3424,
      "step": 778
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.5795953273773193,
      "learning_rate": 0.00021986396475822848,
      "loss": 1.5052,
      "step": 779
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.3943787813186646,
      "learning_rate": 0.00021977889870707318,
      "loss": 1.4324,
      "step": 780
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.5282405614852905,
      "learning_rate": 0.00021969372927135397,
      "loss": 1.3772,
      "step": 781
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.6440399885177612,
      "learning_rate": 0.00021960845654397349,
      "loss": 1.1698,
      "step": 782
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.5308705568313599,
      "learning_rate": 0.0002195230806179469,
      "loss": 1.4512,
      "step": 783
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.3149093389511108,
      "learning_rate": 0.00021943760158640212,
      "loss": 1.1414,
      "step": 784
    },
    {
      "epoch": 0.76,
      "grad_norm": 2.2626311779022217,
      "learning_rate": 0.00021935201954257938,
      "loss": 2.0695,
      "step": 785
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.4505892992019653,
      "learning_rate": 0.0002192663345798314,
      "loss": 1.6039,
      "step": 786
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.3184144496917725,
      "learning_rate": 0.00021918054679162302,
      "loss": 1.2249,
      "step": 787
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.313861608505249,
      "learning_rate": 0.00021909465627153142,
      "loss": 1.5637,
      "step": 788
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.7990531921386719,
      "learning_rate": 0.00021900866311324568,
      "loss": 1.7634,
      "step": 789
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.6349966526031494,
      "learning_rate": 0.00021892256741056694,
      "loss": 1.9673,
      "step": 790
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.3929771184921265,
      "learning_rate": 0.00021883636925740813,
      "loss": 1.3263,
      "step": 791
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.237173318862915,
      "learning_rate": 0.00021875006874779397,
      "loss": 1.5712,
      "step": 792
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.4733974933624268,
      "learning_rate": 0.00021866366597586082,
      "loss": 1.3899,
      "step": 793
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.4081130027770996,
      "learning_rate": 0.0002185771610358566,
      "loss": 1.6359,
      "step": 794
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.3193763494491577,
      "learning_rate": 0.00021849055402214064,
      "loss": 1.3589,
      "step": 795
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.4265916347503662,
      "learning_rate": 0.00021840384502918363,
      "loss": 1.4326,
      "step": 796
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.3795539140701294,
      "learning_rate": 0.00021831703415156752,
      "loss": 1.2139,
      "step": 797
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.7029786109924316,
      "learning_rate": 0.00021823012148398542,
      "loss": 1.5998,
      "step": 798
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.5365678071975708,
      "learning_rate": 0.00021814310712124134,
      "loss": 1.9567,
      "step": 799
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.8077536821365356,
      "learning_rate": 0.00021805599115825038,
      "loss": 1.8023,
      "step": 800
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.4349457025527954,
      "learning_rate": 0.0002179687736900383,
      "loss": 1.7043,
      "step": 801
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.450710654258728,
      "learning_rate": 0.00021788145481174183,
      "loss": 1.2796,
      "step": 802
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.2185072898864746,
      "learning_rate": 0.00021779403461860805,
      "loss": 1.1272,
      "step": 803
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.6366219520568848,
      "learning_rate": 0.00021770651320599473,
      "loss": 1.5251,
      "step": 804
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.445094347000122,
      "learning_rate": 0.0002176188906693699,
      "loss": 1.3265,
      "step": 805
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.5272247791290283,
      "learning_rate": 0.00021753116710431205,
      "loss": 0.9833,
      "step": 806
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.4574639797210693,
      "learning_rate": 0.00021744334260650984,
      "loss": 1.6412,
      "step": 807
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.1877381801605225,
      "learning_rate": 0.0002173554172717619,
      "loss": 1.3086,
      "step": 808
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.4031157493591309,
      "learning_rate": 0.000217267391195977,
      "loss": 1.587,
      "step": 809
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.2422850131988525,
      "learning_rate": 0.0002171792644751737,
      "loss": 1.334,
      "step": 810
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.3261579275131226,
      "learning_rate": 0.00021709103720548052,
      "loss": 1.2248,
      "step": 811
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.0835561752319336,
      "learning_rate": 0.00021700270948313532,
      "loss": 0.9224,
      "step": 812
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.3430688381195068,
      "learning_rate": 0.00021691428140448594,
      "loss": 1.2989,
      "step": 813
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.2712783813476562,
      "learning_rate": 0.00021682575306598934,
      "loss": 1.4027,
      "step": 814
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.2724041938781738,
      "learning_rate": 0.00021673712456421202,
      "loss": 1.2837,
      "step": 815
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.2153583765029907,
      "learning_rate": 0.00021664839599582974,
      "loss": 1.3666,
      "step": 816
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.2074302434921265,
      "learning_rate": 0.00021655956745762742,
      "loss": 1.4386,
      "step": 817
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.372488021850586,
      "learning_rate": 0.0002164706390464989,
      "loss": 1.626,
      "step": 818
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.1615793704986572,
      "learning_rate": 0.00021638161085944706,
      "loss": 1.1384,
      "step": 819
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.1077085733413696,
      "learning_rate": 0.00021629248299358367,
      "loss": 0.9159,
      "step": 820
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.2498594522476196,
      "learning_rate": 0.00021620325554612912,
      "loss": 1.2139,
      "step": 821
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.6090998649597168,
      "learning_rate": 0.00021611392861441243,
      "loss": 1.9134,
      "step": 822
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.2484623193740845,
      "learning_rate": 0.00021602450229587123,
      "loss": 1.2969,
      "step": 823
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.4213321208953857,
      "learning_rate": 0.0002159349766880515,
      "loss": 1.717,
      "step": 824
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.0547410249710083,
      "learning_rate": 0.0002158453518886075,
      "loss": 0.9849,
      "step": 825
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.2111304998397827,
      "learning_rate": 0.00021575562799530169,
      "loss": 1.2678,
      "step": 826
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.66452956199646,
      "learning_rate": 0.0002156658051060047,
      "loss": 1.4332,
      "step": 827
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.3838409185409546,
      "learning_rate": 0.0002155758833186951,
      "loss": 1.1844,
      "step": 828
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.5047062635421753,
      "learning_rate": 0.0002154858627314593,
      "loss": 1.7889,
      "step": 829
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.286152720451355,
      "learning_rate": 0.00021539574344249152,
      "loss": 1.1978,
      "step": 830
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.1742347478866577,
      "learning_rate": 0.00021530552555009365,
      "loss": 1.0202,
      "step": 831
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.7456132173538208,
      "learning_rate": 0.0002152152091526751,
      "loss": 1.0362,
      "step": 832
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.381914496421814,
      "learning_rate": 0.00021512479434875276,
      "loss": 1.4382,
      "step": 833
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.2699590921401978,
      "learning_rate": 0.00021503428123695082,
      "loss": 1.512,
      "step": 834
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.1738026142120361,
      "learning_rate": 0.00021494366991600076,
      "loss": 1.2607,
      "step": 835
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.9550883769989014,
      "learning_rate": 0.0002148529604847412,
      "loss": 1.8124,
      "step": 836
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.504042148590088,
      "learning_rate": 0.00021476215304211765,
      "loss": 1.4629,
      "step": 837
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.0610342025756836,
      "learning_rate": 0.00021467124768718269,
      "loss": 0.9717,
      "step": 838
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.478193759918213,
      "learning_rate": 0.00021458024451909564,
      "loss": 1.802,
      "step": 839
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.218469262123108,
      "learning_rate": 0.00021448914363712246,
      "loss": 1.3854,
      "step": 840
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.2758036851882935,
      "learning_rate": 0.00021439794514063577,
      "loss": 1.2681,
      "step": 841
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.3164936304092407,
      "learning_rate": 0.00021430664912911463,
      "loss": 1.4467,
      "step": 842
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.414792537689209,
      "learning_rate": 0.0002142152557021445,
      "loss": 1.5323,
      "step": 843
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.1610770225524902,
      "learning_rate": 0.00021412376495941705,
      "loss": 1.4057,
      "step": 844
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.2606096267700195,
      "learning_rate": 0.00021403217700073017,
      "loss": 1.19,
      "step": 845
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.1834993362426758,
      "learning_rate": 0.0002139404919259877,
      "loss": 1.5106,
      "step": 846
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.2146964073181152,
      "learning_rate": 0.00021384870983519956,
      "loss": 1.4995,
      "step": 847
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.208569049835205,
      "learning_rate": 0.0002137568308284813,
      "loss": 1.0893,
      "step": 848
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.429599642753601,
      "learning_rate": 0.00021366485500605437,
      "loss": 1.1302,
      "step": 849
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.868918538093567,
      "learning_rate": 0.0002135727824682457,
      "loss": 2.0047,
      "step": 850
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.1158243417739868,
      "learning_rate": 0.00021348061331548777,
      "loss": 1.1251,
      "step": 851
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.1816633939743042,
      "learning_rate": 0.00021338834764831843,
      "loss": 1.0859,
      "step": 852
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.3078497648239136,
      "learning_rate": 0.0002132959855673809,
      "loss": 1.3654,
      "step": 853
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.5226352214813232,
      "learning_rate": 0.00021320352717342335,
      "loss": 1.3744,
      "step": 854
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.194767713546753,
      "learning_rate": 0.0002131109725672992,
      "loss": 1.1332,
      "step": 855
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.292400598526001,
      "learning_rate": 0.00021301832184996686,
      "loss": 1.1676,
      "step": 856
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.307769536972046,
      "learning_rate": 0.00021292557512248933,
      "loss": 1.3764,
      "step": 857
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.5784074068069458,
      "learning_rate": 0.00021283273248603457,
      "loss": 1.8682,
      "step": 858
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.4524656534194946,
      "learning_rate": 0.00021273979404187507,
      "loss": 1.355,
      "step": 859
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.2774903774261475,
      "learning_rate": 0.0002126467598913878,
      "loss": 1.1275,
      "step": 860
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.364617943763733,
      "learning_rate": 0.00021255363013605419,
      "loss": 1.7862,
      "step": 861
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.259527325630188,
      "learning_rate": 0.0002124604048774599,
      "loss": 1.3798,
      "step": 862
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.6640812158584595,
      "learning_rate": 0.0002123670842172948,
      "loss": 1.8239,
      "step": 863
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.2190797328948975,
      "learning_rate": 0.0002122736682573528,
      "loss": 1.0124,
      "step": 864
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.341090202331543,
      "learning_rate": 0.00021218015709953175,
      "loss": 0.985,
      "step": 865
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.3625142574310303,
      "learning_rate": 0.00021208655084583344,
      "loss": 1.0811,
      "step": 866
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.861794352531433,
      "learning_rate": 0.0002119928495983633,
      "loss": 1.5199,
      "step": 867
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.5730104446411133,
      "learning_rate": 0.00021189905345933032,
      "loss": 1.4189,
      "step": 868
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.224785566329956,
      "learning_rate": 0.0002118051625310471,
      "loss": 1.2948,
      "step": 869
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.4107848405838013,
      "learning_rate": 0.00021171117691592964,
      "loss": 1.3929,
      "step": 870
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.16511070728302,
      "learning_rate": 0.00021161709671649721,
      "loss": 1.0999,
      "step": 871
    },
    {
      "epoch": 0.84,
      "grad_norm": 2.30039119720459,
      "learning_rate": 0.00021152292203537216,
      "loss": 1.6013,
      "step": 872
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.5170789957046509,
      "learning_rate": 0.00021142865297528004,
      "loss": 1.8839,
      "step": 873
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.2411909103393555,
      "learning_rate": 0.00021133428963904927,
      "loss": 1.4265,
      "step": 874
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.0655996799468994,
      "learning_rate": 0.00021123983212961112,
      "loss": 1.2643,
      "step": 875
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.1014069318771362,
      "learning_rate": 0.00021114528054999953,
      "loss": 1.0326,
      "step": 876
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.3042875528335571,
      "learning_rate": 0.0002110506350033512,
      "loss": 1.5919,
      "step": 877
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.9417317509651184,
      "learning_rate": 0.00021095589559290518,
      "loss": 0.8426,
      "step": 878
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.341861605644226,
      "learning_rate": 0.00021086106242200296,
      "loss": 1.2377,
      "step": 879
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.2200474739074707,
      "learning_rate": 0.0002107661355940883,
      "loss": 1.4793,
      "step": 880
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.4158042669296265,
      "learning_rate": 0.00021067111521270713,
      "loss": 1.752,
      "step": 881
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.4061779975891113,
      "learning_rate": 0.0002105760013815074,
      "loss": 1.089,
      "step": 882
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.3693805932998657,
      "learning_rate": 0.00021048079420423908,
      "loss": 1.4861,
      "step": 883
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.4483033418655396,
      "learning_rate": 0.00021038549378475386,
      "loss": 1.4553,
      "step": 884
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.283494234085083,
      "learning_rate": 0.00021029010022700518,
      "loss": 1.2053,
      "step": 885
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.5050575733184814,
      "learning_rate": 0.00021019461363504805,
      "loss": 1.2142,
      "step": 886
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.2599867582321167,
      "learning_rate": 0.00021009903411303902,
      "loss": 1.3242,
      "step": 887
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.2808127403259277,
      "learning_rate": 0.00021000336176523593,
      "loss": 1.1527,
      "step": 888
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.145213007926941,
      "learning_rate": 0.00020990759669599799,
      "loss": 1.1483,
      "step": 889
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.1295148134231567,
      "learning_rate": 0.0002098117390097854,
      "loss": 1.0159,
      "step": 890
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.159382939338684,
      "learning_rate": 0.00020971578881115948,
      "loss": 1.1626,
      "step": 891
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.6490403413772583,
      "learning_rate": 0.00020961974620478248,
      "loss": 1.7539,
      "step": 892
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.517959475517273,
      "learning_rate": 0.0002095236112954174,
      "loss": 1.6324,
      "step": 893
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.4953640699386597,
      "learning_rate": 0.0002094273841879279,
      "loss": 1.4384,
      "step": 894
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.3297251462936401,
      "learning_rate": 0.00020933106498727825,
      "loss": 1.1462,
      "step": 895
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.2726671695709229,
      "learning_rate": 0.00020923465379853323,
      "loss": 1.0989,
      "step": 896
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.1799503564834595,
      "learning_rate": 0.00020913815072685785,
      "loss": 0.9603,
      "step": 897
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.2199748754501343,
      "learning_rate": 0.00020904155587751742,
      "loss": 1.149,
      "step": 898
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.8773846626281738,
      "learning_rate": 0.00020894486935587732,
      "loss": 0.8965,
      "step": 899
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.669113039970398,
      "learning_rate": 0.00020884809126740295,
      "loss": 1.7323,
      "step": 900
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.3780359029769897,
      "learning_rate": 0.00020875122171765958,
      "loss": 1.47,
      "step": 901
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.1259912252426147,
      "learning_rate": 0.00020865426081231227,
      "loss": 1.3505,
      "step": 902
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.4640130996704102,
      "learning_rate": 0.00020855720865712568,
      "loss": 1.6051,
      "step": 903
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.0640982389450073,
      "learning_rate": 0.00020846006535796407,
      "loss": 1.109,
      "step": 904
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.1131609678268433,
      "learning_rate": 0.00020836283102079108,
      "loss": 1.1407,
      "step": 905
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.1612396240234375,
      "learning_rate": 0.00020826550575166964,
      "loss": 1.2,
      "step": 906
    },
    {
      "epoch": 0.88,
      "grad_norm": 1.6626241207122803,
      "learning_rate": 0.00020816808965676186,
      "loss": 1.8475,
      "step": 907
    },
    {
      "epoch": 0.88,
      "grad_norm": 1.438838005065918,
      "learning_rate": 0.000208070582842329,
      "loss": 1.4186,
      "step": 908
    },
    {
      "epoch": 0.88,
      "grad_norm": 1.4505665302276611,
      "learning_rate": 0.00020797298541473127,
      "loss": 1.6378,
      "step": 909
    },
    {
      "epoch": 0.88,
      "grad_norm": 1.3546316623687744,
      "learning_rate": 0.00020787529748042758,
      "loss": 1.3801,
      "step": 910
    },
    {
      "epoch": 0.88,
      "grad_norm": 1.5958383083343506,
      "learning_rate": 0.0002077775191459757,
      "loss": 1.7361,
      "step": 911
    },
    {
      "epoch": 0.88,
      "grad_norm": 1.6373250484466553,
      "learning_rate": 0.000207679650518032,
      "loss": 1.8638,
      "step": 912
    },
    {
      "epoch": 0.88,
      "grad_norm": 1.126513957977295,
      "learning_rate": 0.00020758169170335123,
      "loss": 1.2306,
      "step": 913
    },
    {
      "epoch": 0.88,
      "grad_norm": 1.2469696998596191,
      "learning_rate": 0.0002074836428087867,
      "loss": 1.4286,
      "step": 914
    },
    {
      "epoch": 0.88,
      "grad_norm": 1.3433685302734375,
      "learning_rate": 0.00020738550394128983,
      "loss": 1.1645,
      "step": 915
    },
    {
      "epoch": 0.88,
      "grad_norm": 1.274887204170227,
      "learning_rate": 0.0002072872752079102,
      "loss": 1.4313,
      "step": 916
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.2697557210922241,
      "learning_rate": 0.0002071889567157955,
      "loss": 1.1121,
      "step": 917
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.6256011724472046,
      "learning_rate": 0.00020709054857219122,
      "loss": 2.3053,
      "step": 918
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.4982554912567139,
      "learning_rate": 0.00020699205088444073,
      "loss": 1.5737,
      "step": 919
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.3380775451660156,
      "learning_rate": 0.00020689346375998504,
      "loss": 1.3681,
      "step": 920
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.1807115077972412,
      "learning_rate": 0.0002067947873063627,
      "loss": 1.121,
      "step": 921
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.6416897773742676,
      "learning_rate": 0.0002066960216312097,
      "loss": 1.585,
      "step": 922
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.310311198234558,
      "learning_rate": 0.00020659716684225942,
      "loss": 1.0295,
      "step": 923
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.283509373664856,
      "learning_rate": 0.00020649822304734234,
      "loss": 0.8625,
      "step": 924
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.6311004161834717,
      "learning_rate": 0.00020639919035438616,
      "loss": 1.5964,
      "step": 925
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.4839829206466675,
      "learning_rate": 0.00020630006887141533,
      "loss": 1.3802,
      "step": 926
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.6016814708709717,
      "learning_rate": 0.00020620085870655142,
      "loss": 1.3962,
      "step": 927
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.3555843830108643,
      "learning_rate": 0.00020610155996801256,
      "loss": 1.3931,
      "step": 928
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.5287679433822632,
      "learning_rate": 0.00020600217276411353,
      "loss": 1.0829,
      "step": 929
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.1010750532150269,
      "learning_rate": 0.00020590269720326568,
      "loss": 0.9906,
      "step": 930
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.2286770343780518,
      "learning_rate": 0.00020580313339397662,
      "loss": 1.099,
      "step": 931
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.4633378982543945,
      "learning_rate": 0.0002057034814448503,
      "loss": 1.2768,
      "step": 932
    },
    {
      "epoch": 0.9,
      "grad_norm": 2.0793206691741943,
      "learning_rate": 0.0002056037414645868,
      "loss": 1.5472,
      "step": 933
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.041421890258789,
      "learning_rate": 0.00020550391356198222,
      "loss": 1.0693,
      "step": 934
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.7149021625518799,
      "learning_rate": 0.00020540399784592852,
      "loss": 1.6992,
      "step": 935
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.1274030208587646,
      "learning_rate": 0.00020530399442541356,
      "loss": 1.1349,
      "step": 936
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.6209064722061157,
      "learning_rate": 0.00020520390340952074,
      "loss": 1.6395,
      "step": 937
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.0727425813674927,
      "learning_rate": 0.00020510372490742913,
      "loss": 1.0555,
      "step": 938
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.1689037084579468,
      "learning_rate": 0.00020500345902841311,
      "loss": 1.4163,
      "step": 939
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.1197943687438965,
      "learning_rate": 0.00020490310588184242,
      "loss": 1.669,
      "step": 940
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.0973910093307495,
      "learning_rate": 0.00020480266557718201,
      "loss": 1.2043,
      "step": 941
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.459172010421753,
      "learning_rate": 0.00020470213822399198,
      "loss": 1.6599,
      "step": 942
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.4092894792556763,
      "learning_rate": 0.00020460152393192717,
      "loss": 1.5244,
      "step": 943
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.0109457969665527,
      "learning_rate": 0.0002045008228107374,
      "loss": 1.0698,
      "step": 944
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.1727893352508545,
      "learning_rate": 0.00020440003497026722,
      "loss": 0.8874,
      "step": 945
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.4926542043685913,
      "learning_rate": 0.0002042991605204557,
      "loss": 1.1908,
      "step": 946
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.2154324054718018,
      "learning_rate": 0.00020419819957133642,
      "loss": 1.0055,
      "step": 947
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.1640907526016235,
      "learning_rate": 0.00020409715223303733,
      "loss": 1.0244,
      "step": 948
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.094460129737854,
      "learning_rate": 0.00020399601861578054,
      "loss": 1.086,
      "step": 949
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.0807240009307861,
      "learning_rate": 0.0002038947988298824,
      "loss": 1.0952,
      "step": 950
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.2182865142822266,
      "learning_rate": 0.00020379349298575312,
      "loss": 0.8286,
      "step": 951
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.2438262701034546,
      "learning_rate": 0.0002036921011938968,
      "loss": 1.449,
      "step": 952
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.0871261358261108,
      "learning_rate": 0.00020359062356491142,
      "loss": 1.005,
      "step": 953
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.3494447469711304,
      "learning_rate": 0.0002034890602094885,
      "loss": 1.599,
      "step": 954
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.3108654022216797,
      "learning_rate": 0.00020338741123841297,
      "loss": 1.3039,
      "step": 955
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.4159644842147827,
      "learning_rate": 0.00020328567676256335,
      "loss": 1.3015,
      "step": 956
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.267722249031067,
      "learning_rate": 0.00020318385689291126,
      "loss": 1.1788,
      "step": 957
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.2124974727630615,
      "learning_rate": 0.00020308195174052157,
      "loss": 1.3477,
      "step": 958
    },
    {
      "epoch": 0.93,
      "grad_norm": 1.7106714248657227,
      "learning_rate": 0.00020297996141655216,
      "loss": 1.7461,
      "step": 959
    },
    {
      "epoch": 0.93,
      "grad_norm": 1.3902013301849365,
      "learning_rate": 0.00020287788603225377,
      "loss": 1.3323,
      "step": 960
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.9589385390281677,
      "learning_rate": 0.00020277572569896995,
      "loss": 0.7713,
      "step": 961
    },
    {
      "epoch": 0.93,
      "grad_norm": 1.223151683807373,
      "learning_rate": 0.00020267348052813692,
      "loss": 1.2357,
      "step": 962
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.9997661709785461,
      "learning_rate": 0.0002025711506312834,
      "loss": 0.9083,
      "step": 963
    },
    {
      "epoch": 0.93,
      "grad_norm": 1.2039114236831665,
      "learning_rate": 0.0002024687361200306,
      "loss": 1.0277,
      "step": 964
    },
    {
      "epoch": 0.93,
      "grad_norm": 1.2620604038238525,
      "learning_rate": 0.00020236623710609202,
      "loss": 0.9147,
      "step": 965
    },
    {
      "epoch": 0.93,
      "grad_norm": 1.603683352470398,
      "learning_rate": 0.00020226365370127323,
      "loss": 1.3692,
      "step": 966
    },
    {
      "epoch": 0.93,
      "grad_norm": 1.1889077425003052,
      "learning_rate": 0.00020216098601747197,
      "loss": 1.3586,
      "step": 967
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.9763696193695068,
      "learning_rate": 0.00020205823416667788,
      "loss": 1.0737,
      "step": 968
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.9170948266983032,
      "learning_rate": 0.00020195539826097238,
      "loss": 1.7105,
      "step": 969
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.3122986555099487,
      "learning_rate": 0.00020185247841252856,
      "loss": 1.6161,
      "step": 970
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.4224249124526978,
      "learning_rate": 0.00020174947473361118,
      "loss": 1.5831,
      "step": 971
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.293426752090454,
      "learning_rate": 0.00020164638733657637,
      "loss": 1.3102,
      "step": 972
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.07170569896698,
      "learning_rate": 0.0002015432163338715,
      "loss": 0.8819,
      "step": 973
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.181535005569458,
      "learning_rate": 0.00020143996183803525,
      "loss": 1.1775,
      "step": 974
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.3837966918945312,
      "learning_rate": 0.00020133662396169743,
      "loss": 1.3791,
      "step": 975
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.1325147151947021,
      "learning_rate": 0.00020123320281757856,
      "loss": 0.9244,
      "step": 976
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.8120399713516235,
      "learning_rate": 0.0002011296985184903,
      "loss": 1.3342,
      "step": 977
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.260125994682312,
      "learning_rate": 0.0002010261111773347,
      "loss": 1.3665,
      "step": 978
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.342794418334961,
      "learning_rate": 0.00020092244090710472,
      "loss": 1.2215,
      "step": 979
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.5904815196990967,
      "learning_rate": 0.0002008186878208834,
      "loss": 1.071,
      "step": 980
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.2493515014648438,
      "learning_rate": 0.00020071485203184455,
      "loss": 1.5185,
      "step": 981
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.2379603385925293,
      "learning_rate": 0.00020061093365325184,
      "loss": 1.239,
      "step": 982
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.4626051187515259,
      "learning_rate": 0.00020050693279845914,
      "loss": 1.5534,
      "step": 983
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.3357350826263428,
      "learning_rate": 0.0002004028495809103,
      "loss": 1.8324,
      "step": 984
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.4206643104553223,
      "learning_rate": 0.0002002986841141391,
      "loss": 1.251,
      "step": 985
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.436003565788269,
      "learning_rate": 0.00020019443651176884,
      "loss": 1.7747,
      "step": 986
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.350991129875183,
      "learning_rate": 0.0002000901068875126,
      "loss": 1.5258,
      "step": 987
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.1044528484344482,
      "learning_rate": 0.00019998569535517276,
      "loss": 1.269,
      "step": 988
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.4368743896484375,
      "learning_rate": 0.00019988120202864125,
      "loss": 1.7985,
      "step": 989
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.080803394317627,
      "learning_rate": 0.00019977662702189904,
      "loss": 0.8469,
      "step": 990
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.4379607439041138,
      "learning_rate": 0.00019967197044901618,
      "loss": 1.48,
      "step": 991
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.1783312559127808,
      "learning_rate": 0.00019956723242415194,
      "loss": 1.4203,
      "step": 992
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.2120602130889893,
      "learning_rate": 0.0001994624130615542,
      "loss": 1.2925,
      "step": 993
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.0752936601638794,
      "learning_rate": 0.00019935751247555954,
      "loss": 1.4643,
      "step": 994
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.1408238410949707,
      "learning_rate": 0.0001992525307805934,
      "loss": 1.3049,
      "step": 995
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.1272218227386475,
      "learning_rate": 0.0001991474680911694,
      "loss": 1.0908,
      "step": 996
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.525564432144165,
      "learning_rate": 0.0001990423245218897,
      "loss": 1.2493,
      "step": 997
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.3659435510635376,
      "learning_rate": 0.0001989371001874446,
      "loss": 1.402,
      "step": 998
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.344510793685913,
      "learning_rate": 0.00019883179520261245,
      "loss": 1.519,
      "step": 999
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.455416202545166,
      "learning_rate": 0.0001987264096822598,
      "loss": 1.6428,
      "step": 1000
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.206539273262024,
      "learning_rate": 0.00019862094374134074,
      "loss": 1.2198,
      "step": 1001
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.2967417240142822,
      "learning_rate": 0.0001985153974948973,
      "loss": 1.4467,
      "step": 1002
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.3688677549362183,
      "learning_rate": 0.00019840977105805896,
      "loss": 1.5175,
      "step": 1003
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.4224023818969727,
      "learning_rate": 0.00019830406454604289,
      "loss": 1.3388,
      "step": 1004
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.1016898155212402,
      "learning_rate": 0.00019819827807415332,
      "loss": 1.256,
      "step": 1005
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.9830572605133057,
      "learning_rate": 0.00019809241175778195,
      "loss": 1.0293,
      "step": 1006
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.3320682048797607,
      "learning_rate": 0.00019798646571240742,
      "loss": 1.6151,
      "step": 1007
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.203979730606079,
      "learning_rate": 0.00019788044005359534,
      "loss": 1.0101,
      "step": 1008
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.121233344078064,
      "learning_rate": 0.0001977743348969983,
      "loss": 0.9686,
      "step": 1009
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.2444918155670166,
      "learning_rate": 0.0001976681503583555,
      "loss": 1.5227,
      "step": 1010
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.322102665901184,
      "learning_rate": 0.0001975618865534927,
      "loss": 1.6012,
      "step": 1011
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.2791893482208252,
      "learning_rate": 0.0001974555435983222,
      "loss": 1.3951,
      "step": 1012
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.2599291801452637,
      "learning_rate": 0.00019734912160884255,
      "loss": 1.5789,
      "step": 1013
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.2614623308181763,
      "learning_rate": 0.0001972426207011387,
      "loss": 1.2646,
      "step": 1014
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.9194993376731873,
      "learning_rate": 0.0001971360409913814,
      "loss": 1.1464,
      "step": 1015
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.739387035369873,
      "learning_rate": 0.00019702938259582754,
      "loss": 1.6735,
      "step": 1016
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.1382012367248535,
      "learning_rate": 0.00019692264563081984,
      "loss": 1.2768,
      "step": 1017
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.9404625296592712,
      "learning_rate": 0.00019681583021278673,
      "loss": 1.1835,
      "step": 1018
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.2209988832473755,
      "learning_rate": 0.00019670893645824212,
      "loss": 0.9123,
      "step": 1019
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.1911715269088745,
      "learning_rate": 0.00019660196448378542,
      "loss": 1.4704,
      "step": 1020
    },
    {
      "epoch": 0.99,
      "grad_norm": 1.1453022956848145,
      "learning_rate": 0.00019649491440610137,
      "loss": 1.1527,
      "step": 1021
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.9455304145812988,
      "learning_rate": 0.0001963877863419599,
      "loss": 0.9554,
      "step": 1022
    },
    {
      "epoch": 0.99,
      "grad_norm": 1.5460333824157715,
      "learning_rate": 0.00019628058040821607,
      "loss": 1.5023,
      "step": 1023
    },
    {
      "epoch": 0.99,
      "grad_norm": 1.0515046119689941,
      "learning_rate": 0.00019617329672180975,
      "loss": 1.3552,
      "step": 1024
    },
    {
      "epoch": 0.99,
      "grad_norm": 1.4096556901931763,
      "learning_rate": 0.00019606593539976574,
      "loss": 1.3264,
      "step": 1025
    },
    {
      "epoch": 0.99,
      "grad_norm": 1.2442529201507568,
      "learning_rate": 0.00019595849655919343,
      "loss": 0.8898,
      "step": 1026
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.8789858818054199,
      "learning_rate": 0.00019585098031728684,
      "loss": 0.9477,
      "step": 1027
    },
    {
      "epoch": 0.99,
      "grad_norm": 1.2422945499420166,
      "learning_rate": 0.0001957433867913244,
      "loss": 1.1831,
      "step": 1028
    },
    {
      "epoch": 0.99,
      "grad_norm": 1.2705669403076172,
      "learning_rate": 0.00019563571609866883,
      "loss": 1.4243,
      "step": 1029
    },
    {
      "epoch": 0.99,
      "grad_norm": 1.4680469036102295,
      "learning_rate": 0.000195527968356767,
      "loss": 1.2629,
      "step": 1030
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.369444727897644,
      "learning_rate": 0.0001954201436831499,
      "loss": 1.6357,
      "step": 1031
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.4526840448379517,
      "learning_rate": 0.00019531224219543233,
      "loss": 1.3395,
      "step": 1032
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.1711682081222534,
      "learning_rate": 0.00019520426401131304,
      "loss": 1.0378,
      "step": 1033
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.085681676864624,
      "learning_rate": 0.0001950962092485742,
      "loss": 1.3059,
      "step": 1034
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.2222062349319458,
      "learning_rate": 0.00019498807802508173,
      "loss": 1.1306,
      "step": 1035
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.879671573638916,
      "learning_rate": 0.00019487987045878485,
      "loss": 0.7041,
      "step": 1036
    }
  ],
  "logging_steps": 1,
  "max_steps": 3108,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1.472235393377108e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}