{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.08769177437454527,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.769177437454527e-05,
      "grad_norm": 7.59375,
      "learning_rate": 3e-06,
      "loss": 4.1911,
      "step": 1
    },
    {
      "epoch": 0.00017538354874909053,
      "grad_norm": 7.6875,
      "learning_rate": 6e-06,
      "loss": 4.1743,
      "step": 2
    },
    {
      "epoch": 0.0002630753231236358,
      "grad_norm": 8.1875,
      "learning_rate": 9e-06,
      "loss": 4.1165,
      "step": 3
    },
    {
      "epoch": 0.00035076709749818106,
      "grad_norm": 8.0,
      "learning_rate": 1.2e-05,
      "loss": 4.2652,
      "step": 4
    },
    {
      "epoch": 0.00043845887187272633,
      "grad_norm": 7.9375,
      "learning_rate": 1.5e-05,
      "loss": 4.1691,
      "step": 5
    },
    {
      "epoch": 0.0005261506462472716,
      "grad_norm": 8.0625,
      "learning_rate": 1.8e-05,
      "loss": 4.2089,
      "step": 6
    },
    {
      "epoch": 0.0006138424206218169,
      "grad_norm": 8.25,
      "learning_rate": 2.1000000000000002e-05,
      "loss": 4.1677,
      "step": 7
    },
    {
      "epoch": 0.0007015341949963621,
      "grad_norm": 7.8125,
      "learning_rate": 2.4e-05,
      "loss": 4.1564,
      "step": 8
    },
    {
      "epoch": 0.0007892259693709074,
      "grad_norm": 7.75,
      "learning_rate": 2.7e-05,
      "loss": 4.1722,
      "step": 9
    },
    {
      "epoch": 0.0008769177437454527,
      "grad_norm": 7.3125,
      "learning_rate": 3e-05,
      "loss": 4.1554,
      "step": 10
    },
    {
      "epoch": 0.0009646095181199979,
      "grad_norm": 7.53125,
      "learning_rate": 3.2999999999999996e-05,
      "loss": 4.1511,
      "step": 11
    },
    {
      "epoch": 0.0010523012924945432,
      "grad_norm": 7.1875,
      "learning_rate": 3.6e-05,
      "loss": 4.152,
      "step": 12
    },
    {
      "epoch": 0.0011399930668690886,
      "grad_norm": 6.8125,
      "learning_rate": 3.9e-05,
      "loss": 4.1581,
      "step": 13
    },
    {
      "epoch": 0.0012276848412436337,
      "grad_norm": 6.625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 4.0485,
      "step": 14
    },
    {
      "epoch": 0.001315376615618179,
      "grad_norm": 6.28125,
      "learning_rate": 4.4999999999999996e-05,
      "loss": 4.083,
      "step": 15
    },
    {
      "epoch": 0.0014030683899927242,
      "grad_norm": 5.625,
      "learning_rate": 4.8e-05,
      "loss": 4.075,
      "step": 16
    },
    {
      "epoch": 0.0014907601643672696,
      "grad_norm": 5.625,
      "learning_rate": 5.1000000000000006e-05,
      "loss": 4.0571,
      "step": 17
    },
    {
      "epoch": 0.0015784519387418148,
      "grad_norm": 5.53125,
      "learning_rate": 5.4e-05,
      "loss": 4.0213,
      "step": 18
    },
    {
      "epoch": 0.0016661437131163602,
      "grad_norm": 4.5625,
      "learning_rate": 5.7e-05,
      "loss": 4.0436,
      "step": 19
    },
    {
      "epoch": 0.0017538354874909053,
      "grad_norm": 4.6875,
      "learning_rate": 6e-05,
      "loss": 4.0306,
      "step": 20
    },
    {
      "epoch": 0.0018415272618654507,
      "grad_norm": 4.125,
      "learning_rate": 6.3e-05,
      "loss": 4.0364,
      "step": 21
    },
    {
      "epoch": 0.0019292190362399958,
      "grad_norm": 3.875,
      "learning_rate": 6.599999999999999e-05,
      "loss": 3.9774,
      "step": 22
    },
    {
      "epoch": 0.002016910810614541,
      "grad_norm": 3.421875,
      "learning_rate": 6.9e-05,
      "loss": 3.9513,
      "step": 23
    },
    {
      "epoch": 0.0021046025849890864,
      "grad_norm": 3.21875,
      "learning_rate": 7.2e-05,
      "loss": 3.9453,
      "step": 24
    },
    {
      "epoch": 0.0021922943593636317,
      "grad_norm": 2.734375,
      "learning_rate": 7.500000000000001e-05,
      "loss": 3.9338,
      "step": 25
    },
    {
      "epoch": 0.002279986133738177,
      "grad_norm": 2.375,
      "learning_rate": 7.8e-05,
      "loss": 3.9318,
      "step": 26
    },
    {
      "epoch": 0.0023676779081127225,
      "grad_norm": 1.921875,
      "learning_rate": 8.1e-05,
      "loss": 3.8758,
      "step": 27
    },
    {
      "epoch": 0.0024553696824872674,
      "grad_norm": 1.828125,
      "learning_rate": 8.400000000000001e-05,
      "loss": 3.8885,
      "step": 28
    },
    {
      "epoch": 0.002543061456861813,
      "grad_norm": 1.765625,
      "learning_rate": 8.7e-05,
      "loss": 3.8412,
      "step": 29
    },
    {
      "epoch": 0.002630753231236358,
      "grad_norm": 1.65625,
      "learning_rate": 8.999999999999999e-05,
      "loss": 3.8114,
      "step": 30
    },
    {
      "epoch": 0.0027184450056109036,
      "grad_norm": 1.5234375,
      "learning_rate": 9.3e-05,
      "loss": 3.778,
      "step": 31
    },
    {
      "epoch": 0.0028061367799854485,
      "grad_norm": 1.515625,
      "learning_rate": 9.6e-05,
      "loss": 3.7355,
      "step": 32
    },
    {
      "epoch": 0.002893828554359994,
      "grad_norm": 1.5,
      "learning_rate": 9.900000000000001e-05,
      "loss": 3.7612,
      "step": 33
    },
    {
      "epoch": 0.0029815203287345392,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010200000000000001,
      "loss": 3.7869,
      "step": 34
    },
    {
      "epoch": 0.0030692121031090846,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010500000000000002,
      "loss": 3.6998,
      "step": 35
    },
    {
      "epoch": 0.0031569038774836296,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000108,
      "loss": 3.645,
      "step": 36
    },
    {
      "epoch": 0.003244595651858175,
      "grad_norm": 1.265625,
      "learning_rate": 0.000111,
      "loss": 3.6778,
      "step": 37
    },
    {
      "epoch": 0.0033322874262327203,
      "grad_norm": 1.28125,
      "learning_rate": 0.000114,
      "loss": 3.6011,
      "step": 38
    },
    {
      "epoch": 0.0034199792006072657,
      "grad_norm": 1.28125,
      "learning_rate": 0.000117,
      "loss": 3.6147,
      "step": 39
    },
    {
      "epoch": 0.0035076709749818106,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00012,
      "loss": 3.5427,
      "step": 40
    },
    {
      "epoch": 0.003595362749356356,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000123,
      "loss": 3.5252,
      "step": 41
    },
    {
      "epoch": 0.0036830545237309014,
      "grad_norm": 1.140625,
      "learning_rate": 0.000126,
      "loss": 3.5946,
      "step": 42
    },
    {
      "epoch": 0.0037707462981054467,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000129,
      "loss": 3.498,
      "step": 43
    },
    {
      "epoch": 0.0038584380724799917,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013199999999999998,
      "loss": 3.4045,
      "step": 44
    },
    {
      "epoch": 0.0039461298468545375,
      "grad_norm": 1.46875,
      "learning_rate": 0.000135,
      "loss": 3.4418,
      "step": 45
    },
    {
      "epoch": 0.004033821621229082,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000138,
      "loss": 3.3624,
      "step": 46
    },
    {
      "epoch": 0.004121513395603627,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000141,
      "loss": 3.3476,
      "step": 47
    },
    {
      "epoch": 0.004209205169978173,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000144,
      "loss": 3.2416,
      "step": 48
    },
    {
      "epoch": 0.004296896944352718,
      "grad_norm": 1.171875,
      "learning_rate": 0.000147,
      "loss": 3.263,
      "step": 49
    },
    {
      "epoch": 0.0043845887187272635,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015000000000000001,
      "loss": 3.2391,
      "step": 50
    },
    {
      "epoch": 0.004472280493101809,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000153,
      "loss": 3.1366,
      "step": 51
    },
    {
      "epoch": 0.004559972267476354,
      "grad_norm": 0.75390625,
      "learning_rate": 0.000156,
      "loss": 3.103,
      "step": 52
    },
    {
      "epoch": 0.0046476640418509,
      "grad_norm": 0.77734375,
      "learning_rate": 0.000159,
      "loss": 3.1938,
      "step": 53
    },
    {
      "epoch": 0.004735355816225445,
      "grad_norm": 0.69140625,
      "learning_rate": 0.000162,
      "loss": 3.1503,
      "step": 54
    },
    {
      "epoch": 0.0048230475905999895,
      "grad_norm": 0.70703125,
      "learning_rate": 0.000165,
      "loss": 3.1695,
      "step": 55
    },
    {
      "epoch": 0.004910739364974535,
      "grad_norm": 0.67578125,
      "learning_rate": 0.00016800000000000002,
      "loss": 3.0348,
      "step": 56
    },
    {
      "epoch": 0.00499843113934908,
      "grad_norm": 0.68359375,
      "learning_rate": 0.000171,
      "loss": 3.0016,
      "step": 57
    },
    {
      "epoch": 0.005086122913723626,
      "grad_norm": 0.640625,
      "learning_rate": 0.000174,
      "loss": 2.9536,
      "step": 58
    },
    {
      "epoch": 0.005173814688098171,
      "grad_norm": 0.59765625,
      "learning_rate": 0.000177,
      "loss": 2.9378,
      "step": 59
    },
    {
      "epoch": 0.005261506462472716,
      "grad_norm": 0.56640625,
      "learning_rate": 0.00017999999999999998,
      "loss": 2.9442,
      "step": 60
    },
    {
      "epoch": 0.005349198236847262,
      "grad_norm": 0.55078125,
      "learning_rate": 0.000183,
      "loss": 2.8868,
      "step": 61
    },
    {
      "epoch": 0.005436890011221807,
      "grad_norm": 0.55078125,
      "learning_rate": 0.000186,
      "loss": 2.8555,
      "step": 62
    },
    {
      "epoch": 0.005524581785596352,
      "grad_norm": 0.55078125,
      "learning_rate": 0.000189,
      "loss": 2.8699,
      "step": 63
    },
    {
      "epoch": 0.005612273559970897,
      "grad_norm": 0.59375,
      "learning_rate": 0.000192,
      "loss": 2.9004,
      "step": 64
    },
    {
      "epoch": 0.005699965334345442,
      "grad_norm": 0.625,
      "learning_rate": 0.00019500000000000002,
      "loss": 2.8415,
      "step": 65
    },
    {
      "epoch": 0.005787657108719988,
      "grad_norm": 0.5703125,
      "learning_rate": 0.00019800000000000002,
      "loss": 2.756,
      "step": 66
    },
    {
      "epoch": 0.005875348883094533,
      "grad_norm": 0.54296875,
      "learning_rate": 0.000201,
      "loss": 2.8199,
      "step": 67
    },
    {
      "epoch": 0.0059630406574690785,
      "grad_norm": 0.52734375,
      "learning_rate": 0.00020400000000000003,
      "loss": 2.753,
      "step": 68
    },
    {
      "epoch": 0.006050732431843624,
      "grad_norm": 0.46484375,
      "learning_rate": 0.00020700000000000002,
      "loss": 2.677,
      "step": 69
    },
    {
      "epoch": 0.006138424206218169,
      "grad_norm": 0.5546875,
      "learning_rate": 0.00021000000000000004,
      "loss": 2.7099,
      "step": 70
    },
    {
      "epoch": 0.006226115980592715,
      "grad_norm": 0.5859375,
      "learning_rate": 0.00021299999999999997,
      "loss": 2.6848,
      "step": 71
    },
    {
      "epoch": 0.006313807754967259,
      "grad_norm": 0.439453125,
      "learning_rate": 0.000216,
      "loss": 2.6327,
      "step": 72
    },
    {
      "epoch": 0.0064014995293418045,
      "grad_norm": 0.43359375,
      "learning_rate": 0.00021899999999999998,
      "loss": 2.6438,
      "step": 73
    },
    {
      "epoch": 0.00648919130371635,
      "grad_norm": 0.44921875,
      "learning_rate": 0.000222,
      "loss": 2.6331,
      "step": 74
    },
    {
      "epoch": 0.006576883078090895,
      "grad_norm": 0.4296875,
      "learning_rate": 0.000225,
      "loss": 2.616,
      "step": 75
    },
    {
      "epoch": 0.006664574852465441,
      "grad_norm": 0.3828125,
      "learning_rate": 0.000228,
      "loss": 2.4532,
      "step": 76
    },
    {
      "epoch": 0.006752266626839986,
      "grad_norm": 0.318359375,
      "learning_rate": 0.000231,
      "loss": 2.4668,
      "step": 77
    },
    {
      "epoch": 0.006839958401214531,
      "grad_norm": 0.3203125,
      "learning_rate": 0.000234,
      "loss": 2.4524,
      "step": 78
    },
    {
      "epoch": 0.006927650175589077,
      "grad_norm": 0.353515625,
      "learning_rate": 0.00023700000000000001,
      "loss": 2.4424,
      "step": 79
    },
    {
      "epoch": 0.007015341949963621,
      "grad_norm": 0.30859375,
      "learning_rate": 0.00024,
      "loss": 2.403,
      "step": 80
    },
    {
      "epoch": 0.007103033724338167,
      "grad_norm": 0.294921875,
      "learning_rate": 0.00024300000000000002,
      "loss": 2.4171,
      "step": 81
    },
    {
      "epoch": 0.007190725498712712,
      "grad_norm": 0.298828125,
      "learning_rate": 0.000246,
      "loss": 2.4317,
      "step": 82
    },
    {
      "epoch": 0.007278417273087257,
      "grad_norm": 0.259765625,
      "learning_rate": 0.00024900000000000004,
      "loss": 2.3678,
      "step": 83
    },
    {
      "epoch": 0.007366109047461803,
      "grad_norm": 0.341796875,
      "learning_rate": 0.000252,
      "loss": 2.3322,
      "step": 84
    },
    {
      "epoch": 0.007453800821836348,
      "grad_norm": 0.310546875,
      "learning_rate": 0.000255,
      "loss": 2.2875,
      "step": 85
    },
    {
      "epoch": 0.0075414925962108935,
      "grad_norm": 0.248046875,
      "learning_rate": 0.000258,
      "loss": 2.2755,
      "step": 86
    },
    {
      "epoch": 0.007629184370585439,
      "grad_norm": 0.2578125,
      "learning_rate": 0.000261,
      "loss": 2.245,
      "step": 87
    },
    {
      "epoch": 0.007716876144959983,
      "grad_norm": 0.220703125,
      "learning_rate": 0.00026399999999999997,
      "loss": 2.1879,
      "step": 88
    },
    {
      "epoch": 0.007804567919334529,
      "grad_norm": 0.279296875,
      "learning_rate": 0.000267,
      "loss": 2.263,
      "step": 89
    },
    {
      "epoch": 0.007892259693709075,
      "grad_norm": 0.240234375,
      "learning_rate": 0.00027,
      "loss": 2.2339,
      "step": 90
    },
    {
      "epoch": 0.00797995146808362,
      "grad_norm": 0.23828125,
      "learning_rate": 0.000273,
      "loss": 2.2315,
      "step": 91
    },
    {
      "epoch": 0.008067643242458164,
      "grad_norm": 0.2119140625,
      "learning_rate": 0.000276,
      "loss": 2.1785,
      "step": 92
    },
    {
      "epoch": 0.00815533501683271,
      "grad_norm": 0.216796875,
      "learning_rate": 0.000279,
      "loss": 2.1301,
      "step": 93
    },
    {
      "epoch": 0.008243026791207255,
      "grad_norm": 0.203125,
      "learning_rate": 0.000282,
      "loss": 2.1555,
      "step": 94
    },
    {
      "epoch": 0.008330718565581801,
      "grad_norm": 0.2734375,
      "learning_rate": 0.000285,
      "loss": 2.0664,
      "step": 95
    },
    {
      "epoch": 0.008418410339956345,
      "grad_norm": 0.212890625,
      "learning_rate": 0.000288,
      "loss": 2.0808,
      "step": 96
    },
    {
      "epoch": 0.008506102114330892,
      "grad_norm": 0.2060546875,
      "learning_rate": 0.000291,
      "loss": 1.9987,
      "step": 97
    },
    {
      "epoch": 0.008593793888705436,
      "grad_norm": 0.2177734375,
      "learning_rate": 0.000294,
      "loss": 2.0368,
      "step": 98
    },
    {
      "epoch": 0.008681485663079982,
      "grad_norm": 0.1953125,
      "learning_rate": 0.000297,
      "loss": 2.0081,
      "step": 99
    },
    {
      "epoch": 0.008769177437454527,
      "grad_norm": 0.28515625,
      "learning_rate": 0.00030000000000000003,
      "loss": 2.0451,
      "step": 100
    },
    {
      "epoch": 0.008856869211829071,
      "grad_norm": 0.1875,
      "learning_rate": 0.00030300000000000005,
      "loss": 1.9185,
      "step": 101
    },
    {
      "epoch": 0.008944560986203618,
      "grad_norm": 0.18359375,
      "learning_rate": 0.000306,
      "loss": 1.9602,
      "step": 102
    },
    {
      "epoch": 0.009032252760578162,
      "grad_norm": 0.197265625,
      "learning_rate": 0.000309,
      "loss": 1.9397,
      "step": 103
    },
    {
      "epoch": 0.009119944534952708,
      "grad_norm": 0.271484375,
      "learning_rate": 0.000312,
      "loss": 1.9686,
      "step": 104
    },
    {
      "epoch": 0.009207636309327253,
      "grad_norm": 0.2314453125,
      "learning_rate": 0.000315,
      "loss": 1.903,
      "step": 105
    },
    {
      "epoch": 0.0092953280837018,
      "grad_norm": 0.1923828125,
      "learning_rate": 0.000318,
      "loss": 1.9059,
      "step": 106
    },
    {
      "epoch": 0.009383019858076344,
      "grad_norm": 0.189453125,
      "learning_rate": 0.000321,
      "loss": 1.8646,
      "step": 107
    },
    {
      "epoch": 0.00947071163245089,
      "grad_norm": 0.1826171875,
      "learning_rate": 0.000324,
      "loss": 1.8592,
      "step": 108
    },
    {
      "epoch": 0.009558403406825434,
      "grad_norm": 0.2314453125,
      "learning_rate": 0.000327,
      "loss": 1.8891,
      "step": 109
    },
    {
      "epoch": 0.009646095181199979,
      "grad_norm": 0.1474609375,
      "learning_rate": 0.00033,
      "loss": 1.7582,
      "step": 110
    },
    {
      "epoch": 0.009733786955574525,
      "grad_norm": 0.1708984375,
      "learning_rate": 0.000333,
      "loss": 1.7959,
      "step": 111
    },
    {
      "epoch": 0.00982147872994907,
      "grad_norm": 0.2255859375,
      "learning_rate": 0.00033600000000000004,
      "loss": 1.8503,
      "step": 112
    },
    {
      "epoch": 0.009909170504323616,
      "grad_norm": 0.1640625,
      "learning_rate": 0.000339,
      "loss": 1.8528,
      "step": 113
    },
    {
      "epoch": 0.00999686227869816,
      "grad_norm": 0.1474609375,
      "learning_rate": 0.000342,
      "loss": 1.7961,
      "step": 114
    },
    {
      "epoch": 0.010084554053072707,
      "grad_norm": 0.1337890625,
      "learning_rate": 0.00034500000000000004,
      "loss": 1.7692,
      "step": 115
    },
    {
      "epoch": 0.010172245827447251,
      "grad_norm": 0.1376953125,
      "learning_rate": 0.000348,
      "loss": 1.781,
      "step": 116
    },
    {
      "epoch": 0.010259937601821796,
      "grad_norm": 0.1669921875,
      "learning_rate": 0.000351,
      "loss": 1.7401,
      "step": 117
    },
    {
      "epoch": 0.010347629376196342,
      "grad_norm": 0.1328125,
      "learning_rate": 0.000354,
      "loss": 1.7695,
      "step": 118
    },
    {
      "epoch": 0.010435321150570886,
      "grad_norm": 0.1337890625,
      "learning_rate": 0.000357,
      "loss": 1.744,
      "step": 119
    },
    {
      "epoch": 0.010523012924945433,
      "grad_norm": 0.166015625,
      "learning_rate": 0.00035999999999999997,
      "loss": 1.7182,
      "step": 120
    },
    {
      "epoch": 0.010610704699319977,
      "grad_norm": 0.1474609375,
      "learning_rate": 0.000363,
      "loss": 1.7426,
      "step": 121
    },
    {
      "epoch": 0.010698396473694523,
      "grad_norm": 0.1416015625,
      "learning_rate": 0.000366,
      "loss": 1.6804,
      "step": 122
    },
    {
      "epoch": 0.010786088248069068,
      "grad_norm": 0.12353515625,
      "learning_rate": 0.000369,
      "loss": 1.753,
      "step": 123
    },
    {
      "epoch": 0.010873780022443614,
      "grad_norm": 0.14453125,
      "learning_rate": 0.000372,
      "loss": 1.7363,
      "step": 124
    },
    {
      "epoch": 0.010961471796818159,
      "grad_norm": 0.138671875,
      "learning_rate": 0.000375,
      "loss": 1.7182,
      "step": 125
    },
    {
      "epoch": 0.011049163571192703,
      "grad_norm": 0.158203125,
      "learning_rate": 0.000378,
      "loss": 1.6622,
      "step": 126
    },
    {
      "epoch": 0.01113685534556725,
      "grad_norm": 0.1376953125,
      "learning_rate": 0.000381,
      "loss": 1.6381,
      "step": 127
    },
    {
      "epoch": 0.011224547119941794,
      "grad_norm": 0.1142578125,
      "learning_rate": 0.000384,
      "loss": 1.6721,
      "step": 128
    },
    {
      "epoch": 0.01131223889431634,
      "grad_norm": 0.1669921875,
      "learning_rate": 0.00038700000000000003,
      "loss": 1.6262,
      "step": 129
    },
    {
      "epoch": 0.011399930668690885,
      "grad_norm": 0.1376953125,
      "learning_rate": 0.00039000000000000005,
      "loss": 1.7461,
      "step": 130
    },
    {
      "epoch": 0.011487622443065431,
      "grad_norm": 0.2353515625,
      "learning_rate": 0.000393,
      "loss": 1.6759,
      "step": 131
    },
    {
      "epoch": 0.011575314217439975,
      "grad_norm": 0.208984375,
      "learning_rate": 0.00039600000000000003,
      "loss": 1.6878,
      "step": 132
    },
    {
      "epoch": 0.011663005991814522,
      "grad_norm": 0.29296875,
      "learning_rate": 0.00039900000000000005,
      "loss": 1.6794,
      "step": 133
    },
    {
      "epoch": 0.011750697766189066,
      "grad_norm": 0.2080078125,
      "learning_rate": 0.000402,
      "loss": 1.6118,
      "step": 134
    },
    {
      "epoch": 0.01183838954056361,
      "grad_norm": 0.1982421875,
      "learning_rate": 0.00040500000000000003,
      "loss": 1.6834,
      "step": 135
    },
    {
      "epoch": 0.011926081314938157,
      "grad_norm": 0.1474609375,
      "learning_rate": 0.00040800000000000005,
      "loss": 1.6754,
      "step": 136
    },
    {
      "epoch": 0.012013773089312701,
      "grad_norm": 0.1572265625,
      "learning_rate": 0.000411,
      "loss": 1.6149,
      "step": 137
    },
    {
      "epoch": 0.012101464863687248,
      "grad_norm": 0.10546875,
      "learning_rate": 0.00041400000000000003,
      "loss": 1.7056,
      "step": 138
    },
    {
      "epoch": 0.012189156638061792,
      "grad_norm": 0.1650390625,
      "learning_rate": 0.00041700000000000005,
      "loss": 1.6293,
      "step": 139
    },
    {
      "epoch": 0.012276848412436338,
      "grad_norm": 0.140625,
      "learning_rate": 0.00042000000000000007,
      "loss": 1.6483,
      "step": 140
    },
    {
      "epoch": 0.012364540186810883,
      "grad_norm": 0.1181640625,
      "learning_rate": 0.000423,
      "loss": 1.6171,
      "step": 141
    },
    {
      "epoch": 0.01245223196118543,
      "grad_norm": 0.12255859375,
      "learning_rate": 0.00042599999999999995,
      "loss": 1.6129,
      "step": 142
    },
    {
      "epoch": 0.012539923735559974,
      "grad_norm": 0.1337890625,
      "learning_rate": 0.00042899999999999997,
      "loss": 1.6066,
      "step": 143
    },
    {
      "epoch": 0.012627615509934518,
      "grad_norm": 0.154296875,
      "learning_rate": 0.000432,
      "loss": 1.5817,
      "step": 144
    },
    {
      "epoch": 0.012715307284309064,
      "grad_norm": 0.1162109375,
      "learning_rate": 0.000435,
      "loss": 1.666,
      "step": 145
    },
    {
      "epoch": 0.012802999058683609,
      "grad_norm": 0.1123046875,
      "learning_rate": 0.00043799999999999997,
      "loss": 1.6477,
      "step": 146
    },
    {
      "epoch": 0.012890690833058155,
      "grad_norm": 0.1533203125,
      "learning_rate": 0.000441,
      "loss": 1.5617,
      "step": 147
    },
    {
      "epoch": 0.0129783826074327,
      "grad_norm": 0.154296875,
      "learning_rate": 0.000444,
      "loss": 1.5307,
      "step": 148
    },
    {
      "epoch": 0.013066074381807246,
      "grad_norm": 0.09716796875,
      "learning_rate": 0.00044699999999999997,
      "loss": 1.5576,
      "step": 149
    },
    {
      "epoch": 0.01315376615618179,
      "grad_norm": 0.1533203125,
      "learning_rate": 0.00045,
      "loss": 1.5612,
      "step": 150
    },
    {
      "epoch": 0.013241457930556335,
      "grad_norm": 0.1064453125,
      "learning_rate": 0.000453,
      "loss": 1.5933,
      "step": 151
    },
    {
      "epoch": 0.013329149704930881,
      "grad_norm": 0.140625,
      "learning_rate": 0.000456,
      "loss": 1.5505,
      "step": 152
    },
    {
      "epoch": 0.013416841479305426,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.000459,
      "loss": 1.4969,
      "step": 153
    },
    {
      "epoch": 0.013504533253679972,
      "grad_norm": 0.1318359375,
      "learning_rate": 0.000462,
      "loss": 1.5668,
      "step": 154
    },
    {
      "epoch": 0.013592225028054516,
      "grad_norm": 0.095703125,
      "learning_rate": 0.000465,
      "loss": 1.5287,
      "step": 155
    },
    {
      "epoch": 0.013679916802429063,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.000468,
      "loss": 1.545,
      "step": 156
    },
    {
      "epoch": 0.013767608576803607,
      "grad_norm": 0.1298828125,
      "learning_rate": 0.000471,
      "loss": 1.615,
      "step": 157
    },
    {
      "epoch": 0.013855300351178153,
      "grad_norm": 0.134765625,
      "learning_rate": 0.00047400000000000003,
      "loss": 1.5476,
      "step": 158
    },
    {
      "epoch": 0.013942992125552698,
      "grad_norm": 0.10546875,
      "learning_rate": 0.000477,
      "loss": 1.5439,
      "step": 159
    },
    {
      "epoch": 0.014030683899927242,
      "grad_norm": 0.189453125,
      "learning_rate": 0.00048,
      "loss": 1.5647,
      "step": 160
    },
    {
      "epoch": 0.014118375674301789,
      "grad_norm": 0.11669921875,
      "learning_rate": 0.00048300000000000003,
      "loss": 1.6254,
      "step": 161
    },
    {
      "epoch": 0.014206067448676333,
      "grad_norm": 0.10791015625,
      "learning_rate": 0.00048600000000000005,
      "loss": 1.4983,
      "step": 162
    },
    {
      "epoch": 0.01429375922305088,
      "grad_norm": 0.11474609375,
      "learning_rate": 0.0004890000000000001,
      "loss": 1.5548,
      "step": 163
    },
    {
      "epoch": 0.014381450997425424,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.000492,
      "loss": 1.4867,
      "step": 164
    },
    {
      "epoch": 0.01446914277179997,
      "grad_norm": 0.1162109375,
      "learning_rate": 0.000495,
      "loss": 1.6054,
      "step": 165
    },
    {
      "epoch": 0.014556834546174515,
      "grad_norm": 0.09716796875,
      "learning_rate": 0.0004980000000000001,
      "loss": 1.553,
      "step": 166
    },
    {
      "epoch": 0.014644526320549061,
      "grad_norm": 0.162109375,
      "learning_rate": 0.000501,
      "loss": 1.568,
      "step": 167
    },
    {
      "epoch": 0.014732218094923605,
      "grad_norm": 0.11328125,
      "learning_rate": 0.000504,
      "loss": 1.5792,
      "step": 168
    },
    {
      "epoch": 0.01481990986929815,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.0005070000000000001,
      "loss": 1.5449,
      "step": 169
    },
    {
      "epoch": 0.014907601643672696,
      "grad_norm": 0.1103515625,
      "learning_rate": 0.00051,
      "loss": 1.5593,
      "step": 170
    },
    {
      "epoch": 0.01499529341804724,
      "grad_norm": 0.1064453125,
      "learning_rate": 0.000513,
      "loss": 1.4883,
      "step": 171
    },
    {
      "epoch": 0.015082985192421787,
      "grad_norm": 0.1064453125,
      "learning_rate": 0.000516,
      "loss": 1.529,
      "step": 172
    },
    {
      "epoch": 0.015170676966796331,
      "grad_norm": 0.10302734375,
      "learning_rate": 0.0005189999999999999,
      "loss": 1.466,
      "step": 173
    },
    {
      "epoch": 0.015258368741170878,
      "grad_norm": 0.09033203125,
      "learning_rate": 0.000522,
      "loss": 1.4907,
      "step": 174
    },
    {
      "epoch": 0.015346060515545422,
      "grad_norm": 0.16796875,
      "learning_rate": 0.000525,
      "loss": 1.5739,
      "step": 175
    },
    {
      "epoch": 0.015433752289919967,
      "grad_norm": 0.115234375,
      "learning_rate": 0.0005279999999999999,
      "loss": 1.5142,
      "step": 176
    },
    {
      "epoch": 0.015521444064294513,
      "grad_norm": 0.1904296875,
      "learning_rate": 0.000531,
      "loss": 1.4573,
      "step": 177
    },
    {
      "epoch": 0.015609135838669057,
      "grad_norm": 0.1142578125,
      "learning_rate": 0.000534,
      "loss": 1.5181,
      "step": 178
    },
    {
      "epoch": 0.015696827613043602,
      "grad_norm": 0.1572265625,
      "learning_rate": 0.000537,
      "loss": 1.4548,
      "step": 179
    },
    {
      "epoch": 0.01578451938741815,
      "grad_norm": 0.10498046875,
      "learning_rate": 0.00054,
      "loss": 1.4154,
      "step": 180
    },
    {
      "epoch": 0.015872211161792694,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.000543,
      "loss": 1.4346,
      "step": 181
    },
    {
      "epoch": 0.01595990293616724,
      "grad_norm": 0.103515625,
      "learning_rate": 0.000546,
      "loss": 1.6055,
      "step": 182
    },
    {
      "epoch": 0.016047594710541783,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.000549,
      "loss": 1.4598,
      "step": 183
    },
    {
      "epoch": 0.016135286484916328,
      "grad_norm": 0.11572265625,
      "learning_rate": 0.000552,
      "loss": 1.4608,
      "step": 184
    },
    {
      "epoch": 0.016222978259290876,
      "grad_norm": 0.099609375,
      "learning_rate": 0.000555,
      "loss": 1.5092,
      "step": 185
    },
    {
      "epoch": 0.01631067003366542,
      "grad_norm": 0.099609375,
      "learning_rate": 0.000558,
      "loss": 1.5338,
      "step": 186
    },
    {
      "epoch": 0.016398361808039965,
      "grad_norm": 0.1708984375,
      "learning_rate": 0.000561,
      "loss": 1.4916,
      "step": 187
    },
    {
      "epoch": 0.01648605358241451,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.000564,
      "loss": 1.4815,
      "step": 188
    },
    {
      "epoch": 0.016573745356789057,
      "grad_norm": 0.2470703125,
      "learning_rate": 0.000567,
      "loss": 1.4502,
      "step": 189
    },
    {
      "epoch": 0.016661437131163602,
      "grad_norm": 0.10302734375,
      "learning_rate": 0.00057,
      "loss": 1.5195,
      "step": 190
    },
    {
      "epoch": 0.016749128905538146,
      "grad_norm": 0.15234375,
      "learning_rate": 0.000573,
      "loss": 1.545,
      "step": 191
    },
    {
      "epoch": 0.01683682067991269,
      "grad_norm": 0.146484375,
      "learning_rate": 0.000576,
      "loss": 1.5517,
      "step": 192
    },
    {
      "epoch": 0.016924512454287235,
      "grad_norm": 0.10791015625,
      "learning_rate": 0.000579,
      "loss": 1.5469,
      "step": 193
    },
    {
      "epoch": 0.017012204228661783,
      "grad_norm": 0.1416015625,
      "learning_rate": 0.000582,
      "loss": 1.4311,
      "step": 194
    },
    {
      "epoch": 0.017099896003036328,
      "grad_norm": 0.10888671875,
      "learning_rate": 0.000585,
      "loss": 1.5292,
      "step": 195
    },
    {
      "epoch": 0.017187587777410872,
      "grad_norm": 0.1240234375,
      "learning_rate": 0.000588,
      "loss": 1.4326,
      "step": 196
    },
    {
      "epoch": 0.017275279551785417,
      "grad_norm": 0.10205078125,
      "learning_rate": 0.000591,
      "loss": 1.4556,
      "step": 197
    },
    {
      "epoch": 0.017362971326159965,
      "grad_norm": 0.09521484375,
      "learning_rate": 0.000594,
      "loss": 1.4719,
      "step": 198
    },
    {
      "epoch": 0.01745066310053451,
      "grad_norm": 0.11572265625,
      "learning_rate": 0.0005970000000000001,
      "loss": 1.5113,
      "step": 199
    },
    {
      "epoch": 0.017538354874909054,
      "grad_norm": 0.10498046875,
      "learning_rate": 0.0006000000000000001,
      "loss": 1.4525,
      "step": 200
    },
    {
      "epoch": 0.0176260466492836,
      "grad_norm": 0.0888671875,
      "learning_rate": 0.000603,
      "loss": 1.4616,
      "step": 201
    },
    {
      "epoch": 0.017713738423658143,
      "grad_norm": 0.1474609375,
      "learning_rate": 0.0006060000000000001,
      "loss": 1.4662,
      "step": 202
    },
    {
      "epoch": 0.01780143019803269,
      "grad_norm": 0.1181640625,
      "learning_rate": 0.0006090000000000001,
      "loss": 1.4347,
      "step": 203
    },
    {
      "epoch": 0.017889121972407235,
      "grad_norm": 0.09912109375,
      "learning_rate": 0.000612,
      "loss": 1.5131,
      "step": 204
    },
    {
      "epoch": 0.01797681374678178,
      "grad_norm": 0.1220703125,
      "learning_rate": 0.000615,
      "loss": 1.4848,
      "step": 205
    },
    {
      "epoch": 0.018064505521156324,
      "grad_norm": 0.10498046875,
      "learning_rate": 0.000618,
      "loss": 1.4201,
      "step": 206
    },
    {
      "epoch": 0.018152197295530872,
      "grad_norm": 0.1103515625,
      "learning_rate": 0.000621,
      "loss": 1.4722,
      "step": 207
    },
    {
      "epoch": 0.018239889069905417,
      "grad_norm": 0.0966796875,
      "learning_rate": 0.000624,
      "loss": 1.345,
      "step": 208
    },
    {
      "epoch": 0.01832758084427996,
      "grad_norm": 0.08740234375,
      "learning_rate": 0.000627,
      "loss": 1.3573,
      "step": 209
    },
    {
      "epoch": 0.018415272618654506,
      "grad_norm": 0.1494140625,
      "learning_rate": 0.00063,
      "loss": 1.4277,
      "step": 210
    },
    {
      "epoch": 0.01850296439302905,
      "grad_norm": 0.0849609375,
      "learning_rate": 0.000633,
      "loss": 1.4629,
      "step": 211
    },
    {
      "epoch": 0.0185906561674036,
      "grad_norm": 0.1142578125,
      "learning_rate": 0.000636,
      "loss": 1.4655,
      "step": 212
    },
    {
      "epoch": 0.018678347941778143,
      "grad_norm": 0.1328125,
      "learning_rate": 0.000639,
      "loss": 1.4401,
      "step": 213
    },
    {
      "epoch": 0.018766039716152687,
      "grad_norm": 0.1025390625,
      "learning_rate": 0.000642,
      "loss": 1.4852,
      "step": 214
    },
    {
      "epoch": 0.018853731490527232,
      "grad_norm": 0.099609375,
      "learning_rate": 0.000645,
      "loss": 1.3802,
      "step": 215
    },
    {
      "epoch": 0.01894142326490178,
      "grad_norm": 0.111328125,
      "learning_rate": 0.000648,
      "loss": 1.4288,
      "step": 216
    },
    {
      "epoch": 0.019029115039276324,
      "grad_norm": 0.126953125,
      "learning_rate": 0.000651,
      "loss": 1.467,
      "step": 217
    },
    {
      "epoch": 0.01911680681365087,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.000654,
      "loss": 1.4569,
      "step": 218
    },
    {
      "epoch": 0.019204498588025413,
      "grad_norm": 0.1064453125,
      "learning_rate": 0.000657,
      "loss": 1.4554,
      "step": 219
    },
    {
      "epoch": 0.019292190362399958,
      "grad_norm": 0.154296875,
      "learning_rate": 0.00066,
      "loss": 1.4055,
      "step": 220
    },
    {
      "epoch": 0.019379882136774506,
      "grad_norm": 0.08837890625,
      "learning_rate": 0.0006630000000000001,
      "loss": 1.4994,
      "step": 221
    },
    {
      "epoch": 0.01946757391114905,
      "grad_norm": 0.11767578125,
      "learning_rate": 0.000666,
      "loss": 1.4716,
      "step": 222
    },
    {
      "epoch": 0.019555265685523595,
      "grad_norm": 0.10498046875,
      "learning_rate": 0.000669,
      "loss": 1.3835,
      "step": 223
    },
    {
      "epoch": 0.01964295745989814,
      "grad_norm": 0.11474609375,
      "learning_rate": 0.0006720000000000001,
      "loss": 1.3807,
      "step": 224
    },
    {
      "epoch": 0.019730649234272687,
      "grad_norm": 0.1123046875,
      "learning_rate": 0.000675,
      "loss": 1.4361,
      "step": 225
    },
    {
      "epoch": 0.019818341008647232,
      "grad_norm": 0.109375,
      "learning_rate": 0.000678,
      "loss": 1.4103,
      "step": 226
    },
    {
      "epoch": 0.019906032783021776,
      "grad_norm": 0.1201171875,
      "learning_rate": 0.0006810000000000001,
      "loss": 1.4801,
      "step": 227
    },
    {
      "epoch": 0.01999372455739632,
      "grad_norm": 0.10107421875,
      "learning_rate": 0.000684,
      "loss": 1.4361,
      "step": 228
    },
    {
      "epoch": 0.020081416331770865,
      "grad_norm": 0.11328125,
      "learning_rate": 0.000687,
      "loss": 1.4731,
      "step": 229
    },
    {
      "epoch": 0.020169108106145413,
      "grad_norm": 0.11962890625,
      "learning_rate": 0.0006900000000000001,
      "loss": 1.4505,
      "step": 230
    },
    {
      "epoch": 0.020256799880519958,
      "grad_norm": 0.1484375,
      "learning_rate": 0.000693,
      "loss": 1.3999,
      "step": 231
    },
    {
      "epoch": 0.020344491654894502,
      "grad_norm": 0.0859375,
      "learning_rate": 0.000696,
      "loss": 1.4207,
      "step": 232
    },
    {
      "epoch": 0.020432183429269047,
      "grad_norm": 0.1025390625,
      "learning_rate": 0.0006990000000000001,
      "loss": 1.4174,
      "step": 233
    },
    {
      "epoch": 0.02051987520364359,
      "grad_norm": 0.1279296875,
      "learning_rate": 0.000702,
      "loss": 1.4266,
      "step": 234
    },
    {
      "epoch": 0.02060756697801814,
      "grad_norm": 0.10595703125,
      "learning_rate": 0.000705,
      "loss": 1.3834,
      "step": 235
    },
    {
      "epoch": 0.020695258752392684,
      "grad_norm": 0.09326171875,
      "learning_rate": 0.000708,
      "loss": 1.3796,
      "step": 236
    },
    {
      "epoch": 0.02078295052676723,
      "grad_norm": 0.1455078125,
      "learning_rate": 0.0007109999999999999,
      "loss": 1.4036,
      "step": 237
    },
    {
      "epoch": 0.020870642301141773,
      "grad_norm": 0.1220703125,
      "learning_rate": 0.000714,
      "loss": 1.3561,
      "step": 238
    },
    {
      "epoch": 0.02095833407551632,
      "grad_norm": 0.0849609375,
      "learning_rate": 0.000717,
      "loss": 1.4045,
      "step": 239
    },
    {
      "epoch": 0.021046025849890865,
      "grad_norm": 0.1435546875,
      "learning_rate": 0.0007199999999999999,
      "loss": 1.41,
      "step": 240
    },
    {
      "epoch": 0.02113371762426541,
      "grad_norm": 0.1396484375,
      "learning_rate": 0.000723,
      "loss": 1.4192,
      "step": 241
    },
    {
      "epoch": 0.021221409398639954,
      "grad_norm": 0.11328125,
      "learning_rate": 0.000726,
      "loss": 1.3841,
      "step": 242
    },
    {
      "epoch": 0.0213091011730145,
      "grad_norm": 0.1240234375,
      "learning_rate": 0.000729,
      "loss": 1.3593,
      "step": 243
    },
    {
      "epoch": 0.021396792947389047,
      "grad_norm": 0.1181640625,
      "learning_rate": 0.000732,
      "loss": 1.3946,
      "step": 244
    },
    {
      "epoch": 0.02148448472176359,
      "grad_norm": 0.10400390625,
      "learning_rate": 0.000735,
      "loss": 1.4397,
      "step": 245
    },
    {
      "epoch": 0.021572176496138136,
      "grad_norm": 0.10595703125,
      "learning_rate": 0.000738,
      "loss": 1.3494,
      "step": 246
    },
    {
      "epoch": 0.02165986827051268,
      "grad_norm": 0.1396484375,
      "learning_rate": 0.000741,
      "loss": 1.4703,
      "step": 247
    },
    {
      "epoch": 0.02174756004488723,
      "grad_norm": 0.11181640625,
      "learning_rate": 0.000744,
      "loss": 1.3589,
      "step": 248
    },
    {
      "epoch": 0.021835251819261773,
      "grad_norm": 0.1220703125,
      "learning_rate": 0.000747,
      "loss": 1.3954,
      "step": 249
    },
    {
      "epoch": 0.021922943593636317,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.00075,
      "loss": 1.3917,
      "step": 250
    },
    {
      "epoch": 0.022010635368010862,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.000753,
      "loss": 1.3834,
      "step": 251
    },
    {
      "epoch": 0.022098327142385406,
      "grad_norm": 0.0986328125,
      "learning_rate": 0.000756,
      "loss": 1.4159,
      "step": 252
    },
    {
      "epoch": 0.022186018916759954,
      "grad_norm": 0.103515625,
      "learning_rate": 0.000759,
      "loss": 1.4357,
      "step": 253
    },
    {
      "epoch": 0.0222737106911345,
      "grad_norm": 0.10888671875,
      "learning_rate": 0.000762,
      "loss": 1.4194,
      "step": 254
    },
    {
      "epoch": 0.022361402465509043,
      "grad_norm": 0.09375,
      "learning_rate": 0.0007650000000000001,
      "loss": 1.3669,
      "step": 255
    },
    {
      "epoch": 0.022449094239883588,
      "grad_norm": 0.095703125,
      "learning_rate": 0.000768,
      "loss": 1.4522,
      "step": 256
    },
    {
      "epoch": 0.022536786014258136,
      "grad_norm": 0.1103515625,
      "learning_rate": 0.000771,
      "loss": 1.3672,
      "step": 257
    },
    {
      "epoch": 0.02262447778863268,
      "grad_norm": 0.08984375,
      "learning_rate": 0.0007740000000000001,
      "loss": 1.3274,
      "step": 258
    },
    {
      "epoch": 0.022712169563007225,
      "grad_norm": 0.1298828125,
      "learning_rate": 0.000777,
      "loss": 1.3584,
      "step": 259
    },
    {
      "epoch": 0.02279986133738177,
      "grad_norm": 0.12158203125,
      "learning_rate": 0.0007800000000000001,
      "loss": 1.3933,
      "step": 260
    },
    {
      "epoch": 0.022887553111756314,
      "grad_norm": 0.10888671875,
      "learning_rate": 0.0007830000000000001,
      "loss": 1.4353,
      "step": 261
    },
    {
      "epoch": 0.022975244886130862,
      "grad_norm": 0.10205078125,
      "learning_rate": 0.000786,
      "loss": 1.3784,
      "step": 262
    },
    {
      "epoch": 0.023062936660505406,
      "grad_norm": 0.1708984375,
      "learning_rate": 0.0007890000000000001,
      "loss": 1.4389,
      "step": 263
    },
    {
      "epoch": 0.02315062843487995,
      "grad_norm": 0.10791015625,
      "learning_rate": 0.0007920000000000001,
      "loss": 1.3353,
      "step": 264
    },
    {
      "epoch": 0.023238320209254495,
      "grad_norm": 0.208984375,
      "learning_rate": 0.000795,
      "loss": 1.375,
      "step": 265
    },
    {
      "epoch": 0.023326011983629043,
      "grad_norm": 0.0849609375,
      "learning_rate": 0.0007980000000000001,
      "loss": 1.3568,
      "step": 266
    },
    {
      "epoch": 0.023413703758003588,
      "grad_norm": 0.15625,
      "learning_rate": 0.0008010000000000001,
      "loss": 1.3975,
      "step": 267
    },
    {
      "epoch": 0.023501395532378132,
      "grad_norm": 0.09716796875,
      "learning_rate": 0.000804,
      "loss": 1.4104,
      "step": 268
    },
    {
      "epoch": 0.023589087306752677,
      "grad_norm": 0.1201171875,
      "learning_rate": 0.0008070000000000001,
      "loss": 1.4314,
      "step": 269
    },
    {
      "epoch": 0.02367677908112722,
      "grad_norm": 0.08935546875,
      "learning_rate": 0.0008100000000000001,
      "loss": 1.3977,
      "step": 270
    },
    {
      "epoch": 0.02376447085550177,
      "grad_norm": 0.087890625,
      "learning_rate": 0.000813,
      "loss": 1.4509,
      "step": 271
    },
    {
      "epoch": 0.023852162629876314,
      "grad_norm": 0.154296875,
      "learning_rate": 0.0008160000000000001,
      "loss": 1.4231,
      "step": 272
    },
    {
      "epoch": 0.02393985440425086,
      "grad_norm": 0.099609375,
      "learning_rate": 0.0008190000000000001,
      "loss": 1.381,
      "step": 273
    },
    {
      "epoch": 0.024027546178625403,
      "grad_norm": 0.224609375,
      "learning_rate": 0.000822,
      "loss": 1.423,
      "step": 274
    },
    {
      "epoch": 0.02411523795299995,
      "grad_norm": 0.16015625,
      "learning_rate": 0.0008250000000000001,
      "loss": 1.4066,
      "step": 275
    },
    {
      "epoch": 0.024202929727374495,
      "grad_norm": 0.189453125,
      "learning_rate": 0.0008280000000000001,
      "loss": 1.374,
      "step": 276
    },
    {
      "epoch": 0.02429062150174904,
      "grad_norm": 0.341796875,
      "learning_rate": 0.0008310000000000001,
      "loss": 1.3806,
      "step": 277
    },
    {
      "epoch": 0.024378313276123584,
      "grad_norm": 0.1298828125,
      "learning_rate": 0.0008340000000000001,
      "loss": 1.4127,
      "step": 278
    },
    {
      "epoch": 0.02446600505049813,
      "grad_norm": 0.220703125,
      "learning_rate": 0.0008370000000000001,
      "loss": 1.3966,
      "step": 279
    },
    {
      "epoch": 0.024553696824872677,
      "grad_norm": 0.2138671875,
      "learning_rate": 0.0008400000000000001,
      "loss": 1.401,
      "step": 280
    },
    {
      "epoch": 0.02464138859924722,
      "grad_norm": 0.1513671875,
      "learning_rate": 0.0008430000000000001,
      "loss": 1.4582,
      "step": 281
    },
    {
      "epoch": 0.024729080373621766,
      "grad_norm": 0.138671875,
      "learning_rate": 0.000846,
      "loss": 1.3452,
      "step": 282
    },
    {
      "epoch": 0.02481677214799631,
      "grad_norm": 0.09814453125,
      "learning_rate": 0.0008489999999999999,
      "loss": 1.3579,
      "step": 283
    },
    {
      "epoch": 0.02490446392237086,
      "grad_norm": 0.162109375,
      "learning_rate": 0.0008519999999999999,
      "loss": 1.3803,
      "step": 284
    },
    {
      "epoch": 0.024992155696745403,
      "grad_norm": 0.1279296875,
      "learning_rate": 0.000855,
      "loss": 1.3182,
      "step": 285
    },
    {
      "epoch": 0.025079847471119947,
      "grad_norm": 0.1884765625,
      "learning_rate": 0.0008579999999999999,
      "loss": 1.4524,
      "step": 286
    },
    {
      "epoch": 0.025167539245494492,
      "grad_norm": 0.126953125,
      "learning_rate": 0.000861,
      "loss": 1.3664,
      "step": 287
    },
    {
      "epoch": 0.025255231019869036,
      "grad_norm": 0.1640625,
      "learning_rate": 0.000864,
      "loss": 1.4199,
      "step": 288
    },
    {
      "epoch": 0.025342922794243584,
      "grad_norm": 0.1591796875,
      "learning_rate": 0.0008669999999999999,
      "loss": 1.3351,
      "step": 289
    },
    {
      "epoch": 0.02543061456861813,
      "grad_norm": 0.111328125,
      "learning_rate": 0.00087,
      "loss": 1.3552,
      "step": 290
    },
    {
      "epoch": 0.025518306342992673,
      "grad_norm": 0.193359375,
      "learning_rate": 0.000873,
      "loss": 1.3583,
      "step": 291
    },
    {
      "epoch": 0.025605998117367218,
      "grad_norm": 0.10205078125,
      "learning_rate": 0.0008759999999999999,
      "loss": 1.3731,
      "step": 292
    },
    {
      "epoch": 0.025693689891741762,
      "grad_norm": 0.181640625,
      "learning_rate": 0.000879,
      "loss": 1.3517,
      "step": 293
    },
    {
      "epoch": 0.02578138166611631,
      "grad_norm": 0.09521484375,
      "learning_rate": 0.000882,
      "loss": 1.4209,
      "step": 294
    },
    {
      "epoch": 0.025869073440490855,
      "grad_norm": 0.1357421875,
      "learning_rate": 0.0008849999999999999,
      "loss": 1.3834,
      "step": 295
    },
    {
      "epoch": 0.0259567652148654,
      "grad_norm": 0.12353515625,
      "learning_rate": 0.000888,
      "loss": 1.4558,
      "step": 296
    },
    {
      "epoch": 0.026044456989239944,
      "grad_norm": 0.10791015625,
      "learning_rate": 0.000891,
      "loss": 1.4401,
      "step": 297
    },
    {
      "epoch": 0.026132148763614492,
      "grad_norm": 0.1630859375,
      "learning_rate": 0.0008939999999999999,
      "loss": 1.3981,
      "step": 298
    },
    {
      "epoch": 0.026219840537989036,
      "grad_norm": 0.08740234375,
      "learning_rate": 0.000897,
      "loss": 1.3309,
      "step": 299
    },
    {
      "epoch": 0.02630753231236358,
      "grad_norm": 0.240234375,
      "learning_rate": 0.0009,
      "loss": 1.34,
      "step": 300
    },
    {
      "epoch": 0.026395224086738125,
      "grad_norm": 0.083984375,
      "learning_rate": 0.0009029999999999999,
      "loss": 1.3813,
      "step": 301
    },
    {
      "epoch": 0.02648291586111267,
      "grad_norm": 0.2041015625,
      "learning_rate": 0.000906,
      "loss": 1.4285,
      "step": 302
    },
    {
      "epoch": 0.026570607635487218,
      "grad_norm": 0.11572265625,
      "learning_rate": 0.000909,
      "loss": 1.3756,
      "step": 303
    },
    {
      "epoch": 0.026658299409861762,
      "grad_norm": 0.119140625,
      "learning_rate": 0.000912,
      "loss": 1.394,
      "step": 304
    },
    {
      "epoch": 0.026745991184236307,
      "grad_norm": 0.1552734375,
      "learning_rate": 0.000915,
      "loss": 1.4289,
      "step": 305
    },
    {
      "epoch": 0.02683368295861085,
      "grad_norm": 0.09228515625,
      "learning_rate": 0.000918,
      "loss": 1.3585,
      "step": 306
    },
    {
      "epoch": 0.0269213747329854,
      "grad_norm": 0.193359375,
      "learning_rate": 0.000921,
      "loss": 1.3685,
      "step": 307
    },
    {
      "epoch": 0.027009066507359944,
      "grad_norm": 0.08984375,
      "learning_rate": 0.000924,
      "loss": 1.3597,
      "step": 308
    },
    {
      "epoch": 0.02709675828173449,
      "grad_norm": 0.12451171875,
      "learning_rate": 0.000927,
      "loss": 1.3464,
      "step": 309
    },
    {
      "epoch": 0.027184450056109033,
      "grad_norm": 0.11865234375,
      "learning_rate": 0.00093,
      "loss": 1.4275,
      "step": 310
    },
    {
      "epoch": 0.027272141830483577,
      "grad_norm": 0.10693359375,
      "learning_rate": 0.000933,
      "loss": 1.3538,
      "step": 311
    },
    {
      "epoch": 0.027359833604858125,
      "grad_norm": 0.08740234375,
      "learning_rate": 0.000936,
      "loss": 1.3189,
      "step": 312
    },
    {
      "epoch": 0.02744752537923267,
      "grad_norm": 0.0908203125,
      "learning_rate": 0.0009390000000000001,
      "loss": 1.337,
      "step": 313
    },
    {
      "epoch": 0.027535217153607214,
      "grad_norm": 0.099609375,
      "learning_rate": 0.000942,
      "loss": 1.3509,
      "step": 314
    },
    {
      "epoch": 0.02762290892798176,
      "grad_norm": 0.10302734375,
      "learning_rate": 0.000945,
      "loss": 1.4155,
      "step": 315
    },
    {
      "epoch": 0.027710600702356307,
      "grad_norm": 0.12060546875,
      "learning_rate": 0.0009480000000000001,
      "loss": 1.3987,
      "step": 316
    },
    {
      "epoch": 0.02779829247673085,
      "grad_norm": 0.09765625,
      "learning_rate": 0.000951,
      "loss": 1.3961,
      "step": 317
    },
    {
      "epoch": 0.027885984251105396,
      "grad_norm": 0.10888671875,
      "learning_rate": 0.000954,
      "loss": 1.3766,
      "step": 318
    },
    {
      "epoch": 0.02797367602547994,
      "grad_norm": 0.1171875,
      "learning_rate": 0.0009570000000000001,
      "loss": 1.3738,
      "step": 319
    },
    {
      "epoch": 0.028061367799854485,
      "grad_norm": 0.18359375,
      "learning_rate": 0.00096,
      "loss": 1.4111,
      "step": 320
    },
    {
      "epoch": 0.028149059574229033,
      "grad_norm": 0.111328125,
      "learning_rate": 0.000963,
      "loss": 1.344,
      "step": 321
    },
    {
      "epoch": 0.028236751348603577,
      "grad_norm": 0.2333984375,
      "learning_rate": 0.0009660000000000001,
      "loss": 1.3448,
      "step": 322
    },
    {
      "epoch": 0.028324443122978122,
      "grad_norm": 0.09033203125,
      "learning_rate": 0.000969,
      "loss": 1.4107,
      "step": 323
    },
    {
      "epoch": 0.028412134897352666,
      "grad_norm": 0.1708984375,
      "learning_rate": 0.0009720000000000001,
      "loss": 1.4008,
      "step": 324
    },
    {
      "epoch": 0.028499826671727214,
      "grad_norm": 0.09228515625,
      "learning_rate": 0.0009750000000000001,
      "loss": 1.3442,
      "step": 325
    },
    {
      "epoch": 0.02858751844610176,
      "grad_norm": 0.0966796875,
      "learning_rate": 0.0009780000000000001,
      "loss": 1.4229,
      "step": 326
    },
    {
      "epoch": 0.028675210220476303,
      "grad_norm": 0.1123046875,
      "learning_rate": 0.000981,
      "loss": 1.3201,
      "step": 327
    },
    {
      "epoch": 0.028762901994850848,
      "grad_norm": 0.10595703125,
      "learning_rate": 0.000984,
      "loss": 1.4149,
      "step": 328
    },
    {
      "epoch": 0.028850593769225392,
      "grad_norm": 0.162109375,
      "learning_rate": 0.000987,
      "loss": 1.3379,
      "step": 329
    },
    {
      "epoch": 0.02893828554359994,
      "grad_norm": 0.126953125,
      "learning_rate": 0.00099,
      "loss": 1.3845,
      "step": 330
    },
    {
      "epoch": 0.029025977317974485,
      "grad_norm": 0.126953125,
      "learning_rate": 0.0009930000000000002,
      "loss": 1.3366,
      "step": 331
    },
    {
      "epoch": 0.02911366909234903,
      "grad_norm": 0.130859375,
      "learning_rate": 0.0009960000000000001,
      "loss": 1.4185,
      "step": 332
    },
    {
      "epoch": 0.029201360866723574,
      "grad_norm": 0.126953125,
      "learning_rate": 0.000999,
      "loss": 1.3812,
      "step": 333
    },
    {
      "epoch": 0.029289052641098122,
      "grad_norm": 0.10400390625,
      "learning_rate": 0.001002,
      "loss": 1.3699,
      "step": 334
    },
    {
      "epoch": 0.029376744415472666,
      "grad_norm": 0.1171875,
      "learning_rate": 0.001005,
      "loss": 1.3294,
      "step": 335
    },
    {
      "epoch": 0.02946443618984721,
      "grad_norm": 0.1552734375,
      "learning_rate": 0.001008,
      "loss": 1.3399,
      "step": 336
    },
    {
      "epoch": 0.029552127964221755,
      "grad_norm": 0.09765625,
      "learning_rate": 0.0010110000000000002,
      "loss": 1.3641,
      "step": 337
    },
    {
      "epoch": 0.0296398197385963,
      "grad_norm": 0.1171875,
      "learning_rate": 0.0010140000000000001,
      "loss": 1.3034,
      "step": 338
    },
    {
      "epoch": 0.029727511512970848,
      "grad_norm": 0.09375,
      "learning_rate": 0.0010170000000000001,
      "loss": 1.4665,
      "step": 339
    },
    {
      "epoch": 0.029815203287345392,
      "grad_norm": 0.095703125,
      "learning_rate": 0.00102,
      "loss": 1.362,
      "step": 340
    },
    {
      "epoch": 0.029902895061719937,
      "grad_norm": 0.1328125,
      "learning_rate": 0.001023,
      "loss": 1.4033,
      "step": 341
    },
    {
      "epoch": 0.02999058683609448,
      "grad_norm": 0.091796875,
      "learning_rate": 0.001026,
      "loss": 1.3324,
      "step": 342
    },
    {
      "epoch": 0.030078278610469026,
      "grad_norm": 0.130859375,
      "learning_rate": 0.0010290000000000002,
      "loss": 1.3564,
      "step": 343
    },
    {
      "epoch": 0.030165970384843574,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.001032,
      "loss": 1.4015,
      "step": 344
    },
    {
      "epoch": 0.03025366215921812,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.001035,
      "loss": 1.4183,
      "step": 345
    },
    {
      "epoch": 0.030341353933592663,
      "grad_norm": 0.09716796875,
      "learning_rate": 0.0010379999999999999,
      "loss": 1.402,
      "step": 346
    },
    {
      "epoch": 0.030429045707967207,
      "grad_norm": 0.1279296875,
      "learning_rate": 0.001041,
      "loss": 1.4356,
      "step": 347
    },
    {
      "epoch": 0.030516737482341755,
      "grad_norm": 0.10888671875,
      "learning_rate": 0.001044,
      "loss": 1.3444,
      "step": 348
    },
    {
      "epoch": 0.0306044292567163,
      "grad_norm": 0.1005859375,
      "learning_rate": 0.001047,
      "loss": 1.2408,
      "step": 349
    },
    {
      "epoch": 0.030692121031090844,
      "grad_norm": 0.1181640625,
      "learning_rate": 0.00105,
      "loss": 1.3412,
      "step": 350
    },
    {
      "epoch": 0.03077981280546539,
      "grad_norm": 0.107421875,
      "learning_rate": 0.001053,
      "loss": 1.3695,
      "step": 351
    },
    {
      "epoch": 0.030867504579839933,
      "grad_norm": 0.10498046875,
      "learning_rate": 0.0010559999999999999,
      "loss": 1.2612,
      "step": 352
    },
    {
      "epoch": 0.03095519635421448,
      "grad_norm": 0.126953125,
      "learning_rate": 0.001059,
      "loss": 1.3209,
      "step": 353
    },
    {
      "epoch": 0.031042888128589026,
      "grad_norm": 0.11181640625,
      "learning_rate": 0.001062,
      "loss": 1.3619,
      "step": 354
    },
    {
      "epoch": 0.03113057990296357,
      "grad_norm": 0.11669921875,
      "learning_rate": 0.001065,
      "loss": 1.3898,
      "step": 355
    },
    {
      "epoch": 0.031218271677338115,
      "grad_norm": 0.1083984375,
      "learning_rate": 0.001068,
      "loss": 1.3756,
      "step": 356
    },
    {
      "epoch": 0.03130596345171266,
      "grad_norm": 0.09814453125,
      "learning_rate": 0.001071,
      "loss": 1.3883,
      "step": 357
    },
    {
      "epoch": 0.031393655226087204,
      "grad_norm": 0.1015625,
      "learning_rate": 0.001074,
      "loss": 1.3457,
      "step": 358
    },
    {
      "epoch": 0.031481347000461755,
      "grad_norm": 0.1318359375,
      "learning_rate": 0.001077,
      "loss": 1.3499,
      "step": 359
    },
    {
      "epoch": 0.0315690387748363,
      "grad_norm": 0.125,
      "learning_rate": 0.00108,
      "loss": 1.385,
      "step": 360
    },
    {
      "epoch": 0.031656730549210844,
      "grad_norm": 0.0966796875,
      "learning_rate": 0.001083,
      "loss": 1.3921,
      "step": 361
    },
    {
      "epoch": 0.03174442232358539,
      "grad_norm": 0.10888671875,
      "learning_rate": 0.001086,
      "loss": 1.3615,
      "step": 362
    },
    {
      "epoch": 0.03183211409795993,
      "grad_norm": 0.1376953125,
      "learning_rate": 0.001089,
      "loss": 1.3125,
      "step": 363
    },
    {
      "epoch": 0.03191980587233448,
      "grad_norm": 0.2080078125,
      "learning_rate": 0.001092,
      "loss": 1.3186,
      "step": 364
    },
    {
      "epoch": 0.03200749764670902,
      "grad_norm": 0.09375,
      "learning_rate": 0.001095,
      "loss": 1.364,
      "step": 365
    },
    {
      "epoch": 0.03209518942108357,
      "grad_norm": 0.205078125,
      "learning_rate": 0.001098,
      "loss": 1.4273,
      "step": 366
    },
    {
      "epoch": 0.03218288119545811,
      "grad_norm": 0.09375,
      "learning_rate": 0.001101,
      "loss": 1.3585,
      "step": 367
    },
    {
      "epoch": 0.032270572969832656,
      "grad_norm": 0.111328125,
      "learning_rate": 0.001104,
      "loss": 1.3809,
      "step": 368
    },
    {
      "epoch": 0.03235826474420721,
      "grad_norm": 0.11572265625,
      "learning_rate": 0.001107,
      "loss": 1.4134,
      "step": 369
    },
    {
      "epoch": 0.03244595651858175,
      "grad_norm": 0.154296875,
      "learning_rate": 0.00111,
      "loss": 1.3046,
      "step": 370
    },
    {
      "epoch": 0.032533648292956296,
      "grad_norm": 0.1455078125,
      "learning_rate": 0.001113,
      "loss": 1.44,
      "step": 371
    },
    {
      "epoch": 0.03262134006733084,
      "grad_norm": 0.10400390625,
      "learning_rate": 0.001116,
      "loss": 1.3767,
      "step": 372
    },
    {
      "epoch": 0.032709031841705385,
      "grad_norm": 0.10400390625,
      "learning_rate": 0.001119,
      "loss": 1.3568,
      "step": 373
    },
    {
      "epoch": 0.03279672361607993,
      "grad_norm": 0.10302734375,
      "learning_rate": 0.001122,
      "loss": 1.2876,
      "step": 374
    },
    {
      "epoch": 0.032884415390454474,
      "grad_norm": 0.09326171875,
      "learning_rate": 0.0011250000000000001,
      "loss": 1.3378,
      "step": 375
    },
    {
      "epoch": 0.03297210716482902,
      "grad_norm": 0.12255859375,
      "learning_rate": 0.001128,
      "loss": 1.2907,
      "step": 376
    },
    {
      "epoch": 0.033059798939203563,
      "grad_norm": 0.091796875,
      "learning_rate": 0.001131,
      "loss": 1.3901,
      "step": 377
    },
    {
      "epoch": 0.033147490713578115,
      "grad_norm": 0.12158203125,
      "learning_rate": 0.001134,
      "loss": 1.3821,
      "step": 378
    },
    {
      "epoch": 0.03323518248795266,
      "grad_norm": 0.083984375,
      "learning_rate": 0.001137,
      "loss": 1.3682,
      "step": 379
    },
    {
      "epoch": 0.033322874262327204,
      "grad_norm": 0.1123046875,
      "learning_rate": 0.00114,
      "loss": 1.3609,
      "step": 380
    },
    {
      "epoch": 0.03341056603670175,
      "grad_norm": 0.0966796875,
      "learning_rate": 0.0011430000000000001,
      "loss": 1.3734,
      "step": 381
    },
    {
      "epoch": 0.03349825781107629,
      "grad_norm": 0.1591796875,
      "learning_rate": 0.001146,
      "loss": 1.426,
      "step": 382
    },
    {
      "epoch": 0.03358594958545084,
      "grad_norm": 0.1376953125,
      "learning_rate": 0.001149,
      "loss": 1.3942,
      "step": 383
    },
    {
      "epoch": 0.03367364135982538,
      "grad_norm": 0.0908203125,
      "learning_rate": 0.001152,
      "loss": 1.3952,
      "step": 384
    },
    {
      "epoch": 0.033761333134199926,
      "grad_norm": 0.09033203125,
      "learning_rate": 0.001155,
      "loss": 1.4302,
      "step": 385
    },
    {
      "epoch": 0.03384902490857447,
      "grad_norm": 0.10009765625,
      "learning_rate": 0.001158,
      "loss": 1.3961,
      "step": 386
    },
    {
      "epoch": 0.03393671668294902,
      "grad_norm": 0.10888671875,
      "learning_rate": 0.0011610000000000001,
      "loss": 1.3277,
      "step": 387
    },
    {
      "epoch": 0.03402440845732357,
      "grad_norm": 0.09765625,
      "learning_rate": 0.001164,
      "loss": 1.2633,
      "step": 388
    },
    {
      "epoch": 0.03411210023169811,
      "grad_norm": 0.1220703125,
      "learning_rate": 0.001167,
      "loss": 1.3971,
      "step": 389
    },
    {
      "epoch": 0.034199792006072656,
      "grad_norm": 0.09619140625,
      "learning_rate": 0.00117,
      "loss": 1.3659,
      "step": 390
    },
    {
      "epoch": 0.0342874837804472,
      "grad_norm": 0.1640625,
      "learning_rate": 0.001173,
      "loss": 1.4147,
      "step": 391
    },
    {
      "epoch": 0.034375175554821745,
      "grad_norm": 0.10595703125,
      "learning_rate": 0.001176,
      "loss": 1.3504,
      "step": 392
    },
    {
      "epoch": 0.03446286732919629,
      "grad_norm": 0.15625,
      "learning_rate": 0.0011790000000000001,
      "loss": 1.3842,
      "step": 393
    },
    {
      "epoch": 0.034550559103570834,
      "grad_norm": 0.11279296875,
      "learning_rate": 0.001182,
      "loss": 1.4277,
      "step": 394
    },
    {
      "epoch": 0.03463825087794538,
      "grad_norm": 0.142578125,
      "learning_rate": 0.001185,
      "loss": 1.3052,
      "step": 395
    },
    {
      "epoch": 0.03472594265231993,
      "grad_norm": 0.140625,
      "learning_rate": 0.001188,
      "loss": 1.41,
      "step": 396
    },
    {
      "epoch": 0.034813634426694474,
      "grad_norm": 0.154296875,
      "learning_rate": 0.001191,
      "loss": 1.3605,
      "step": 397
    },
    {
      "epoch": 0.03490132620106902,
      "grad_norm": 0.1572265625,
      "learning_rate": 0.0011940000000000002,
      "loss": 1.3762,
      "step": 398
    },
    {
      "epoch": 0.03498901797544356,
      "grad_norm": 0.1845703125,
      "learning_rate": 0.0011970000000000001,
      "loss": 1.2598,
      "step": 399
    },
    {
      "epoch": 0.03507670974981811,
      "grad_norm": 0.2236328125,
      "learning_rate": 0.0012000000000000001,
      "loss": 1.352,
      "step": 400
    },
    {
      "epoch": 0.03516440152419265,
      "grad_norm": 0.1259765625,
      "learning_rate": 0.001203,
      "loss": 1.3672,
      "step": 401
    },
    {
      "epoch": 0.0352520932985672,
      "grad_norm": 0.1767578125,
      "learning_rate": 0.001206,
      "loss": 1.4493,
      "step": 402
    },
    {
      "epoch": 0.03533978507294174,
      "grad_norm": 0.2236328125,
      "learning_rate": 0.001209,
      "loss": 1.3753,
      "step": 403
    },
    {
      "epoch": 0.035427476847316286,
      "grad_norm": 0.15234375,
      "learning_rate": 0.0012120000000000002,
      "loss": 1.3092,
      "step": 404
    },
    {
      "epoch": 0.03551516862169084,
      "grad_norm": 0.12890625,
      "learning_rate": 0.0012150000000000002,
      "loss": 1.3083,
      "step": 405
    },
    {
      "epoch": 0.03560286039606538,
      "grad_norm": 0.1025390625,
      "learning_rate": 0.0012180000000000001,
      "loss": 1.2666,
      "step": 406
    },
    {
      "epoch": 0.035690552170439926,
      "grad_norm": 0.18359375,
      "learning_rate": 0.0012209999999999999,
      "loss": 1.31,
      "step": 407
    },
    {
      "epoch": 0.03577824394481447,
      "grad_norm": 0.130859375,
      "learning_rate": 0.001224,
      "loss": 1.3723,
      "step": 408
    },
    {
      "epoch": 0.035865935719189015,
      "grad_norm": 0.10400390625,
      "learning_rate": 0.001227,
      "loss": 1.296,
      "step": 409
    },
    {
      "epoch": 0.03595362749356356,
      "grad_norm": 0.1689453125,
      "learning_rate": 0.00123,
      "loss": 1.3209,
      "step": 410
    },
    {
      "epoch": 0.036041319267938104,
      "grad_norm": 0.0869140625,
      "learning_rate": 0.001233,
      "loss": 1.3439,
      "step": 411
    },
    {
      "epoch": 0.03612901104231265,
      "grad_norm": 0.1640625,
      "learning_rate": 0.001236,
      "loss": 1.361,
      "step": 412
    },
    {
      "epoch": 0.036216702816687193,
      "grad_norm": 0.150390625,
      "learning_rate": 0.0012389999999999999,
      "loss": 1.3845,
      "step": 413
    },
    {
      "epoch": 0.036304394591061745,
      "grad_norm": 0.12255859375,
      "learning_rate": 0.001242,
      "loss": 1.3486,
      "step": 414
    },
    {
      "epoch": 0.03639208636543629,
      "grad_norm": 0.173828125,
      "learning_rate": 0.001245,
      "loss": 1.3611,
      "step": 415
    },
    {
      "epoch": 0.036479778139810834,
      "grad_norm": 0.158203125,
      "learning_rate": 0.001248,
      "loss": 1.3968,
      "step": 416
    },
    {
      "epoch": 0.03656746991418538,
      "grad_norm": 0.1904296875,
      "learning_rate": 0.001251,
      "loss": 1.3516,
      "step": 417
    },
    {
      "epoch": 0.03665516168855992,
      "grad_norm": 0.216796875,
      "learning_rate": 0.001254,
      "loss": 1.3711,
      "step": 418
    },
    {
      "epoch": 0.03674285346293447,
      "grad_norm": 0.1689453125,
      "learning_rate": 0.0012569999999999999,
      "loss": 1.3854,
      "step": 419
    },
    {
      "epoch": 0.03683054523730901,
      "grad_norm": 0.1650390625,
      "learning_rate": 0.00126,
      "loss": 1.3558,
      "step": 420
    },
    {
      "epoch": 0.036918237011683556,
      "grad_norm": 0.1728515625,
      "learning_rate": 0.001263,
      "loss": 1.2976,
      "step": 421
    },
    {
      "epoch": 0.0370059287860581,
      "grad_norm": 0.103515625,
      "learning_rate": 0.001266,
      "loss": 1.3582,
      "step": 422
    },
    {
      "epoch": 0.03709362056043265,
      "grad_norm": 0.1689453125,
      "learning_rate": 0.001269,
      "loss": 1.3893,
      "step": 423
    },
    {
      "epoch": 0.0371813123348072,
      "grad_norm": 0.10986328125,
      "learning_rate": 0.001272,
      "loss": 1.2884,
      "step": 424
    },
    {
      "epoch": 0.03726900410918174,
      "grad_norm": 0.1181640625,
      "learning_rate": 0.001275,
      "loss": 1.3434,
      "step": 425
    },
    {
      "epoch": 0.037356695883556286,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.001278,
      "loss": 1.3381,
      "step": 426
    },
    {
      "epoch": 0.03744438765793083,
      "grad_norm": 0.140625,
      "learning_rate": 0.001281,
      "loss": 1.3903,
      "step": 427
    },
    {
      "epoch": 0.037532079432305375,
      "grad_norm": 0.08935546875,
      "learning_rate": 0.001284,
      "loss": 1.3506,
      "step": 428
    },
    {
      "epoch": 0.03761977120667992,
      "grad_norm": 0.09814453125,
      "learning_rate": 0.001287,
      "loss": 1.2621,
      "step": 429
    },
    {
      "epoch": 0.037707462981054464,
      "grad_norm": 0.09814453125,
      "learning_rate": 0.00129,
      "loss": 1.3424,
      "step": 430
    },
    {
      "epoch": 0.03779515475542901,
      "grad_norm": 0.1162109375,
      "learning_rate": 0.001293,
      "loss": 1.4801,
      "step": 431
    },
    {
      "epoch": 0.03788284652980356,
      "grad_norm": 0.09814453125,
      "learning_rate": 0.001296,
      "loss": 1.3269,
      "step": 432
    },
    {
      "epoch": 0.037970538304178104,
      "grad_norm": 0.138671875,
      "learning_rate": 0.001299,
      "loss": 1.3656,
      "step": 433
    },
    {
      "epoch": 0.03805823007855265,
      "grad_norm": 0.08837890625,
      "learning_rate": 0.001302,
      "loss": 1.3883,
      "step": 434
    },
    {
      "epoch": 0.03814592185292719,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.001305,
      "loss": 1.3636,
      "step": 435
    },
    {
      "epoch": 0.03823361362730174,
      "grad_norm": 0.09130859375,
      "learning_rate": 0.001308,
      "loss": 1.3846,
      "step": 436
    },
    {
      "epoch": 0.03832130540167628,
      "grad_norm": 0.1201171875,
      "learning_rate": 0.001311,
      "loss": 1.3506,
      "step": 437
    },
    {
      "epoch": 0.03840899717605083,
      "grad_norm": 0.0859375,
      "learning_rate": 0.001314,
      "loss": 1.3023,
      "step": 438
    },
    {
      "epoch": 0.03849668895042537,
      "grad_norm": 0.109375,
      "learning_rate": 0.001317,
      "loss": 1.3817,
      "step": 439
    },
    {
      "epoch": 0.038584380724799916,
      "grad_norm": 0.1435546875,
      "learning_rate": 0.00132,
      "loss": 1.3687,
      "step": 440
    },
    {
      "epoch": 0.03867207249917447,
      "grad_norm": 0.13671875,
      "learning_rate": 0.001323,
      "loss": 1.3563,
      "step": 441
    },
    {
      "epoch": 0.03875976427354901,
      "grad_norm": 0.08837890625,
      "learning_rate": 0.0013260000000000001,
      "loss": 1.3338,
      "step": 442
    },
    {
      "epoch": 0.038847456047923556,
      "grad_norm": 0.1103515625,
      "learning_rate": 0.001329,
      "loss": 1.3922,
      "step": 443
    },
    {
      "epoch": 0.0389351478222981,
      "grad_norm": 0.12158203125,
      "learning_rate": 0.001332,
      "loss": 1.2568,
      "step": 444
    },
    {
      "epoch": 0.039022839596672645,
      "grad_norm": 0.1337890625,
      "learning_rate": 0.001335,
      "loss": 1.3628,
      "step": 445
    },
    {
      "epoch": 0.03911053137104719,
      "grad_norm": 0.1064453125,
      "learning_rate": 0.001338,
      "loss": 1.3751,
      "step": 446
    },
    {
      "epoch": 0.039198223145421734,
      "grad_norm": 0.126953125,
      "learning_rate": 0.001341,
      "loss": 1.3495,
      "step": 447
    },
    {
      "epoch": 0.03928591491979628,
      "grad_norm": 0.1474609375,
      "learning_rate": 0.0013440000000000001,
      "loss": 1.317,
      "step": 448
    },
    {
      "epoch": 0.039373606694170823,
      "grad_norm": 0.09765625,
      "learning_rate": 0.001347,
      "loss": 1.3535,
      "step": 449
    },
    {
      "epoch": 0.039461298468545375,
      "grad_norm": 0.12890625,
      "learning_rate": 0.00135,
      "loss": 1.2891,
      "step": 450
    },
    {
      "epoch": 0.03954899024291992,
      "grad_norm": 0.1376953125,
      "learning_rate": 0.001353,
      "loss": 1.3959,
      "step": 451
    },
    {
      "epoch": 0.039636682017294464,
      "grad_norm": 0.154296875,
      "learning_rate": 0.001356,
      "loss": 1.3192,
      "step": 452
    },
    {
      "epoch": 0.03972437379166901,
      "grad_norm": 0.1572265625,
      "learning_rate": 0.001359,
      "loss": 1.4324,
      "step": 453
    },
    {
      "epoch": 0.03981206556604355,
      "grad_norm": 0.0986328125,
      "learning_rate": 0.0013620000000000001,
      "loss": 1.3246,
      "step": 454
    },
    {
      "epoch": 0.0398997573404181,
      "grad_norm": 0.138671875,
      "learning_rate": 0.0013650000000000001,
      "loss": 1.2848,
      "step": 455
    },
    {
      "epoch": 0.03998744911479264,
      "grad_norm": 0.142578125,
      "learning_rate": 0.001368,
      "loss": 1.3973,
      "step": 456
    },
    {
      "epoch": 0.040075140889167186,
      "grad_norm": 0.1572265625,
      "learning_rate": 0.001371,
      "loss": 1.3794,
      "step": 457
    },
    {
      "epoch": 0.04016283266354173,
      "grad_norm": 0.099609375,
      "learning_rate": 0.001374,
      "loss": 1.2708,
      "step": 458
    },
    {
      "epoch": 0.04025052443791628,
      "grad_norm": 0.171875,
      "learning_rate": 0.0013770000000000002,
      "loss": 1.3587,
      "step": 459
    },
    {
      "epoch": 0.04033821621229083,
      "grad_norm": 0.1611328125,
      "learning_rate": 0.0013800000000000002,
      "loss": 1.2727,
      "step": 460
    },
    {
      "epoch": 0.04042590798666537,
      "grad_norm": 0.171875,
      "learning_rate": 0.0013830000000000001,
      "loss": 1.3551,
      "step": 461
    },
    {
      "epoch": 0.040513599761039916,
      "grad_norm": 0.1328125,
      "learning_rate": 0.001386,
      "loss": 1.3897,
      "step": 462
    },
    {
      "epoch": 0.04060129153541446,
      "grad_norm": 0.14453125,
      "learning_rate": 0.001389,
      "loss": 1.3319,
      "step": 463
    },
    {
      "epoch": 0.040688983309789005,
      "grad_norm": 0.16796875,
      "learning_rate": 0.001392,
      "loss": 1.2967,
      "step": 464
    },
    {
      "epoch": 0.04077667508416355,
      "grad_norm": 0.138671875,
      "learning_rate": 0.0013950000000000002,
      "loss": 1.3376,
      "step": 465
    },
    {
      "epoch": 0.040864366858538094,
      "grad_norm": 0.146484375,
      "learning_rate": 0.0013980000000000002,
      "loss": 1.3628,
      "step": 466
    },
    {
      "epoch": 0.04095205863291264,
      "grad_norm": 0.1982421875,
      "learning_rate": 0.0014010000000000001,
      "loss": 1.2868,
      "step": 467
    },
    {
      "epoch": 0.04103975040728718,
      "grad_norm": 0.0947265625,
      "learning_rate": 0.001404,
      "loss": 1.3618,
      "step": 468
    },
    {
      "epoch": 0.041127442181661734,
      "grad_norm": 0.2197265625,
      "learning_rate": 0.001407,
      "loss": 1.3419,
      "step": 469
    },
    {
      "epoch": 0.04121513395603628,
      "grad_norm": 0.09765625,
      "learning_rate": 0.00141,
      "loss": 1.3019,
      "step": 470
    },
    {
      "epoch": 0.04130282573041082,
      "grad_norm": 0.2470703125,
      "learning_rate": 0.001413,
      "loss": 1.4217,
      "step": 471
    },
    {
      "epoch": 0.04139051750478537,
      "grad_norm": 0.0986328125,
      "learning_rate": 0.001416,
      "loss": 1.3689,
      "step": 472
    },
    {
      "epoch": 0.04147820927915991,
      "grad_norm": 0.173828125,
      "learning_rate": 0.001419,
      "loss": 1.3036,
      "step": 473
    },
    {
      "epoch": 0.04156590105353446,
      "grad_norm": 0.0986328125,
      "learning_rate": 0.0014219999999999999,
      "loss": 1.3835,
      "step": 474
    },
    {
      "epoch": 0.041653592827909,
      "grad_norm": 0.1611328125,
      "learning_rate": 0.001425,
      "loss": 1.2816,
      "step": 475
    },
    {
      "epoch": 0.041741284602283546,
      "grad_norm": 0.107421875,
      "learning_rate": 0.001428,
      "loss": 1.3424,
      "step": 476
    },
    {
      "epoch": 0.04182897637665809,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.001431,
      "loss": 1.3255,
      "step": 477
    },
    {
      "epoch": 0.04191666815103264,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.001434,
      "loss": 1.358,
      "step": 478
    },
    {
      "epoch": 0.042004359925407186,
      "grad_norm": 0.103515625,
      "learning_rate": 0.001437,
      "loss": 1.3379,
      "step": 479
    },
    {
      "epoch": 0.04209205169978173,
      "grad_norm": 0.1162109375,
      "learning_rate": 0.0014399999999999999,
      "loss": 1.2963,
      "step": 480
    },
    {
      "epoch": 0.042179743474156275,
      "grad_norm": 0.08984375,
      "learning_rate": 0.001443,
      "loss": 1.2964,
      "step": 481
    },
    {
      "epoch": 0.04226743524853082,
      "grad_norm": 0.08837890625,
      "learning_rate": 0.001446,
      "loss": 1.2804,
      "step": 482
    },
    {
      "epoch": 0.042355127022905364,
      "grad_norm": 0.1181640625,
      "learning_rate": 0.001449,
      "loss": 1.353,
      "step": 483
    },
    {
      "epoch": 0.04244281879727991,
      "grad_norm": 0.140625,
      "learning_rate": 0.001452,
      "loss": 1.2805,
      "step": 484
    },
    {
      "epoch": 0.042530510571654453,
      "grad_norm": 0.130859375,
      "learning_rate": 0.001455,
      "loss": 1.3353,
      "step": 485
    },
    {
      "epoch": 0.042618202346029,
      "grad_norm": 0.1845703125,
      "learning_rate": 0.001458,
      "loss": 1.4368,
      "step": 486
    },
    {
      "epoch": 0.04270589412040355,
      "grad_norm": 0.1416015625,
      "learning_rate": 0.001461,
      "loss": 1.355,
      "step": 487
    },
    {
      "epoch": 0.042793585894778094,
      "grad_norm": 0.1318359375,
      "learning_rate": 0.001464,
      "loss": 1.3571,
      "step": 488
    },
    {
      "epoch": 0.04288127766915264,
      "grad_norm": 0.12890625,
      "learning_rate": 0.001467,
      "loss": 1.3144,
      "step": 489
    },
    {
      "epoch": 0.04296896944352718,
      "grad_norm": 0.09716796875,
      "learning_rate": 0.00147,
      "loss": 1.3431,
      "step": 490
    },
    {
      "epoch": 0.04305666121790173,
      "grad_norm": 0.119140625,
      "learning_rate": 0.001473,
      "loss": 1.3331,
      "step": 491
    },
    {
      "epoch": 0.04314435299227627,
      "grad_norm": 0.10205078125,
      "learning_rate": 0.001476,
      "loss": 1.3873,
      "step": 492
    },
    {
      "epoch": 0.043232044766650816,
      "grad_norm": 0.1181640625,
      "learning_rate": 0.001479,
      "loss": 1.3456,
      "step": 493
    },
    {
      "epoch": 0.04331973654102536,
      "grad_norm": 0.10107421875,
      "learning_rate": 0.001482,
      "loss": 1.354,
      "step": 494
    },
    {
      "epoch": 0.043407428315399905,
      "grad_norm": 0.1513671875,
      "learning_rate": 0.001485,
      "loss": 1.3155,
      "step": 495
    },
    {
      "epoch": 0.04349512008977446,
      "grad_norm": 0.1318359375,
      "learning_rate": 0.001488,
      "loss": 1.3318,
      "step": 496
    },
    {
      "epoch": 0.043582811864149,
      "grad_norm": 0.1318359375,
      "learning_rate": 0.001491,
      "loss": 1.3467,
      "step": 497
    },
    {
      "epoch": 0.043670503638523546,
      "grad_norm": 0.1005859375,
      "learning_rate": 0.001494,
      "loss": 1.33,
      "step": 498
    },
    {
      "epoch": 0.04375819541289809,
      "grad_norm": 0.09765625,
      "learning_rate": 0.001497,
      "loss": 1.3274,
      "step": 499
    },
    {
      "epoch": 0.043845887187272635,
      "grad_norm": 0.138671875,
      "learning_rate": 0.0015,
      "loss": 1.3365,
      "step": 500
    },
    {
      "epoch": 0.043845887187272635,
      "eval_loss": 1.3519667387008667,
      "eval_runtime": 427.9384,
      "eval_samples_per_second": 33.76,
      "eval_steps_per_second": 8.44,
      "step": 500
    },
    {
      "epoch": 0.04393357896164718,
      "grad_norm": 0.1767578125,
      "learning_rate": 0.001503,
      "loss": 1.3827,
      "step": 501
    },
    {
      "epoch": 0.044021270736021724,
      "grad_norm": 0.08984375,
      "learning_rate": 0.001506,
      "loss": 1.3601,
      "step": 502
    },
    {
      "epoch": 0.04410896251039627,
      "grad_norm": 0.1298828125,
      "learning_rate": 0.0015090000000000001,
      "loss": 1.3613,
      "step": 503
    },
    {
      "epoch": 0.04419665428477081,
      "grad_norm": 0.15234375,
      "learning_rate": 0.001512,
      "loss": 1.3476,
      "step": 504
    },
    {
      "epoch": 0.044284346059145364,
      "grad_norm": 0.09619140625,
      "learning_rate": 0.001515,
      "loss": 1.382,
      "step": 505
    },
    {
      "epoch": 0.04437203783351991,
      "grad_norm": 0.134765625,
      "learning_rate": 0.001518,
      "loss": 1.3764,
      "step": 506
    },
    {
      "epoch": 0.04445972960789445,
      "grad_norm": 0.09130859375,
      "learning_rate": 0.001521,
      "loss": 1.265,
      "step": 507
    },
    {
      "epoch": 0.044547421382269,
      "grad_norm": 0.0908203125,
      "learning_rate": 0.001524,
      "loss": 1.3309,
      "step": 508
    },
    {
      "epoch": 0.04463511315664354,
      "grad_norm": 0.095703125,
      "learning_rate": 0.0015270000000000001,
      "loss": 1.3333,
      "step": 509
    },
    {
      "epoch": 0.04472280493101809,
      "grad_norm": 0.09765625,
      "learning_rate": 0.0015300000000000001,
      "loss": 1.3741,
      "step": 510
    },
    {
      "epoch": 0.04481049670539263,
      "grad_norm": 0.1279296875,
      "learning_rate": 0.001533,
      "loss": 1.3363,
      "step": 511
    },
    {
      "epoch": 0.044898188479767176,
      "grad_norm": 0.12890625,
      "learning_rate": 0.001536,
      "loss": 1.3045,
      "step": 512
    },
    {
      "epoch": 0.04498588025414172,
      "grad_norm": 0.15625,
      "learning_rate": 0.001539,
      "loss": 1.4171,
      "step": 513
    },
    {
      "epoch": 0.04507357202851627,
      "grad_norm": 0.0849609375,
      "learning_rate": 0.001542,
      "loss": 1.3343,
      "step": 514
    },
    {
      "epoch": 0.045161263802890816,
      "grad_norm": 0.1640625,
      "learning_rate": 0.0015450000000000001,
      "loss": 1.3433,
      "step": 515
    },
    {
      "epoch": 0.04524895557726536,
      "grad_norm": 0.09912109375,
      "learning_rate": 0.0015480000000000001,
      "loss": 1.281,
      "step": 516
    },
    {
      "epoch": 0.045336647351639905,
      "grad_norm": 0.1171875,
      "learning_rate": 0.001551,
      "loss": 1.3229,
      "step": 517
    },
    {
      "epoch": 0.04542433912601445,
      "grad_norm": 0.0869140625,
      "learning_rate": 0.001554,
      "loss": 1.3284,
      "step": 518
    },
    {
      "epoch": 0.045512030900388994,
      "grad_norm": 0.11767578125,
      "learning_rate": 0.001557,
      "loss": 1.3185,
      "step": 519
    },
    {
      "epoch": 0.04559972267476354,
      "grad_norm": 0.12255859375,
      "learning_rate": 0.0015600000000000002,
      "loss": 1.4157,
      "step": 520
    },
    {
      "epoch": 0.045687414449138083,
      "grad_norm": 0.1171875,
      "learning_rate": 0.0015630000000000002,
      "loss": 1.3225,
      "step": 521
    },
    {
      "epoch": 0.04577510622351263,
      "grad_norm": 0.1591796875,
      "learning_rate": 0.0015660000000000001,
      "loss": 1.2768,
      "step": 522
    },
    {
      "epoch": 0.04586279799788718,
      "grad_norm": 0.1083984375,
      "learning_rate": 0.001569,
      "loss": 1.3378,
      "step": 523
    },
    {
      "epoch": 0.045950489772261724,
      "grad_norm": 0.2158203125,
      "learning_rate": 0.001572,
      "loss": 1.3312,
      "step": 524
    },
    {
      "epoch": 0.04603818154663627,
      "grad_norm": 0.146484375,
      "learning_rate": 0.001575,
      "loss": 1.3473,
      "step": 525
    },
    {
      "epoch": 0.04612587332101081,
      "grad_norm": 0.1533203125,
      "learning_rate": 0.0015780000000000002,
      "loss": 1.3193,
      "step": 526
    },
    {
      "epoch": 0.04621356509538536,
      "grad_norm": 0.166015625,
      "learning_rate": 0.0015810000000000002,
      "loss": 1.3617,
      "step": 527
    },
    {
      "epoch": 0.0463012568697599,
      "grad_norm": 0.146484375,
      "learning_rate": 0.0015840000000000001,
      "loss": 1.4205,
      "step": 528
    },
    {
      "epoch": 0.046388948644134446,
      "grad_norm": 0.1943359375,
      "learning_rate": 0.001587,
      "loss": 1.3412,
      "step": 529
    },
    {
      "epoch": 0.04647664041850899,
      "grad_norm": 0.138671875,
      "learning_rate": 0.00159,
      "loss": 1.3731,
      "step": 530
    },
    {
      "epoch": 0.046564332192883535,
      "grad_norm": 0.158203125,
      "learning_rate": 0.001593,
      "loss": 1.3481,
      "step": 531
    },
    {
      "epoch": 0.04665202396725809,
      "grad_norm": 0.11865234375,
      "learning_rate": 0.0015960000000000002,
      "loss": 1.3438,
      "step": 532
    },
    {
      "epoch": 0.04673971574163263,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.0015990000000000002,
      "loss": 1.3893,
      "step": 533
    },
    {
      "epoch": 0.046827407516007176,
      "grad_norm": 0.12060546875,
      "learning_rate": 0.0016020000000000001,
      "loss": 1.3365,
      "step": 534
    },
    {
      "epoch": 0.04691509929038172,
      "grad_norm": 0.177734375,
      "learning_rate": 0.001605,
      "loss": 1.3429,
      "step": 535
    },
    {
      "epoch": 0.047002791064756265,
      "grad_norm": 0.109375,
      "learning_rate": 0.001608,
      "loss": 1.3629,
      "step": 536
    },
    {
      "epoch": 0.04709048283913081,
      "grad_norm": 0.177734375,
      "learning_rate": 0.0016110000000000002,
      "loss": 1.2756,
      "step": 537
    },
    {
      "epoch": 0.047178174613505354,
      "grad_norm": 0.1103515625,
      "learning_rate": 0.0016140000000000002,
      "loss": 1.3064,
      "step": 538
    },
    {
      "epoch": 0.0472658663878799,
      "grad_norm": 0.1455078125,
      "learning_rate": 0.0016170000000000002,
      "loss": 1.3871,
      "step": 539
    },
    {
      "epoch": 0.04735355816225444,
      "grad_norm": 0.1220703125,
      "learning_rate": 0.0016200000000000001,
      "loss": 1.2368,
      "step": 540
    },
    {
      "epoch": 0.047441249936628994,
      "grad_norm": 0.1572265625,
      "learning_rate": 0.001623,
      "loss": 1.376,
      "step": 541
    },
    {
      "epoch": 0.04752894171100354,
      "grad_norm": 0.16015625,
      "learning_rate": 0.001626,
      "loss": 1.3061,
      "step": 542
    },
    {
      "epoch": 0.04761663348537808,
      "grad_norm": 0.130859375,
      "learning_rate": 0.0016290000000000002,
      "loss": 1.2628,
      "step": 543
    },
    {
      "epoch": 0.04770432525975263,
      "grad_norm": 0.1337890625,
      "learning_rate": 0.0016320000000000002,
      "loss": 1.2792,
      "step": 544
    },
    {
      "epoch": 0.04779201703412717,
      "grad_norm": 0.10107421875,
      "learning_rate": 0.0016350000000000002,
      "loss": 1.285,
      "step": 545
    },
    {
      "epoch": 0.04787970880850172,
      "grad_norm": 0.193359375,
      "learning_rate": 0.0016380000000000001,
      "loss": 1.3308,
      "step": 546
    },
    {
      "epoch": 0.04796740058287626,
      "grad_norm": 0.10302734375,
      "learning_rate": 0.001641,
      "loss": 1.3833,
      "step": 547
    },
    {
      "epoch": 0.048055092357250806,
      "grad_norm": 0.2080078125,
      "learning_rate": 0.001644,
      "loss": 1.3428,
      "step": 548
    },
    {
      "epoch": 0.04814278413162535,
      "grad_norm": 0.083984375,
      "learning_rate": 0.0016470000000000002,
      "loss": 1.3376,
      "step": 549
    },
    {
      "epoch": 0.0482304759059999,
      "grad_norm": 0.099609375,
      "learning_rate": 0.0016500000000000002,
      "loss": 1.4305,
      "step": 550
    },
    {
      "epoch": 0.048318167680374446,
      "grad_norm": 0.140625,
      "learning_rate": 0.0016530000000000002,
      "loss": 1.3828,
      "step": 551
    },
    {
      "epoch": 0.04840585945474899,
      "grad_norm": 0.09716796875,
      "learning_rate": 0.0016560000000000001,
      "loss": 1.323,
      "step": 552
    },
    {
      "epoch": 0.048493551229123535,
      "grad_norm": 0.12890625,
      "learning_rate": 0.001659,
      "loss": 1.3733,
      "step": 553
    },
    {
      "epoch": 0.04858124300349808,
      "grad_norm": 0.08837890625,
      "learning_rate": 0.0016620000000000003,
      "loss": 1.3101,
      "step": 554
    },
    {
      "epoch": 0.048668934777872624,
      "grad_norm": 0.1064453125,
      "learning_rate": 0.0016650000000000002,
      "loss": 1.3806,
      "step": 555
    },
    {
      "epoch": 0.04875662655224717,
      "grad_norm": 0.119140625,
      "learning_rate": 0.0016680000000000002,
      "loss": 1.321,
      "step": 556
    },
    {
      "epoch": 0.04884431832662171,
      "grad_norm": 0.1416015625,
      "learning_rate": 0.0016710000000000002,
      "loss": 1.3756,
      "step": 557
    },
    {
      "epoch": 0.04893201010099626,
      "grad_norm": 0.1572265625,
      "learning_rate": 0.0016740000000000001,
      "loss": 1.3487,
      "step": 558
    },
    {
      "epoch": 0.04901970187537081,
      "grad_norm": 0.11669921875,
      "learning_rate": 0.001677,
      "loss": 1.2843,
      "step": 559
    },
    {
      "epoch": 0.049107393649745354,
      "grad_norm": 0.1220703125,
      "learning_rate": 0.0016800000000000003,
      "loss": 1.3796,
      "step": 560
    },
    {
      "epoch": 0.0491950854241199,
      "grad_norm": 0.08056640625,
      "learning_rate": 0.0016830000000000003,
      "loss": 1.3371,
      "step": 561
    },
    {
      "epoch": 0.04928277719849444,
      "grad_norm": 0.10107421875,
      "learning_rate": 0.0016860000000000002,
      "loss": 1.3571,
      "step": 562
    },
    {
      "epoch": 0.04937046897286899,
      "grad_norm": 0.140625,
      "learning_rate": 0.001689,
      "loss": 1.3039,
      "step": 563
    },
    {
      "epoch": 0.04945816074724353,
      "grad_norm": 0.0888671875,
      "learning_rate": 0.001692,
      "loss": 1.3606,
      "step": 564
    },
    {
      "epoch": 0.049545852521618076,
      "grad_norm": 0.1259765625,
      "learning_rate": 0.001695,
      "loss": 1.348,
      "step": 565
    },
    {
      "epoch": 0.04963354429599262,
      "grad_norm": 0.10400390625,
      "learning_rate": 0.0016979999999999999,
      "loss": 1.291,
      "step": 566
    },
    {
      "epoch": 0.049721236070367165,
      "grad_norm": 0.142578125,
      "learning_rate": 0.0017009999999999998,
      "loss": 1.3581,
      "step": 567
    },
    {
      "epoch": 0.04980892784474172,
      "grad_norm": 0.0947265625,
      "learning_rate": 0.0017039999999999998,
      "loss": 1.311,
      "step": 568
    },
    {
      "epoch": 0.04989661961911626,
      "grad_norm": 0.1416015625,
      "learning_rate": 0.001707,
      "loss": 1.3596,
      "step": 569
    },
    {
      "epoch": 0.049984311393490806,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.00171,
      "loss": 1.3427,
      "step": 570
    },
    {
      "epoch": 0.05007200316786535,
      "grad_norm": 0.1396484375,
      "learning_rate": 0.001713,
      "loss": 1.3852,
      "step": 571
    },
    {
      "epoch": 0.050159694942239895,
      "grad_norm": 0.1767578125,
      "learning_rate": 0.0017159999999999999,
      "loss": 1.3147,
      "step": 572
    },
    {
      "epoch": 0.05024738671661444,
      "grad_norm": 0.11181640625,
      "learning_rate": 0.0017189999999999998,
      "loss": 1.3285,
      "step": 573
    },
    {
      "epoch": 0.050335078490988984,
      "grad_norm": 0.1865234375,
      "learning_rate": 0.001722,
      "loss": 1.3213,
      "step": 574
    },
    {
      "epoch": 0.05042277026536353,
      "grad_norm": 0.09619140625,
      "learning_rate": 0.001725,
      "loss": 1.2918,
      "step": 575
    },
    {
      "epoch": 0.05051046203973807,
      "grad_norm": 0.09912109375,
      "learning_rate": 0.001728,
      "loss": 1.3342,
      "step": 576
    },
    {
      "epoch": 0.05059815381411262,
      "grad_norm": 0.11865234375,
      "learning_rate": 0.001731,
      "loss": 1.3351,
      "step": 577
    },
    {
      "epoch": 0.05068584558848717,
      "grad_norm": 0.140625,
      "learning_rate": 0.0017339999999999999,
      "loss": 1.4707,
      "step": 578
    },
    {
      "epoch": 0.05077353736286171,
      "grad_norm": 0.125,
      "learning_rate": 0.0017369999999999998,
      "loss": 1.3766,
      "step": 579
    },
    {
      "epoch": 0.05086122913723626,
      "grad_norm": 0.1630859375,
      "learning_rate": 0.00174,
      "loss": 1.3022,
      "step": 580
    },
    {
      "epoch": 0.0509489209116108,
      "grad_norm": 0.109375,
      "learning_rate": 0.001743,
      "loss": 1.3365,
      "step": 581
    },
    {
      "epoch": 0.05103661268598535,
      "grad_norm": 0.1552734375,
      "learning_rate": 0.001746,
      "loss": 1.2805,
      "step": 582
    },
    {
      "epoch": 0.05112430446035989,
      "grad_norm": 0.1396484375,
      "learning_rate": 0.001749,
      "loss": 1.3641,
      "step": 583
    },
    {
      "epoch": 0.051211996234734436,
      "grad_norm": 0.130859375,
      "learning_rate": 0.0017519999999999999,
      "loss": 1.3591,
      "step": 584
    },
    {
      "epoch": 0.05129968800910898,
      "grad_norm": 0.1123046875,
      "learning_rate": 0.0017549999999999998,
      "loss": 1.2795,
      "step": 585
    },
    {
      "epoch": 0.051387379783483525,
      "grad_norm": 0.09814453125,
      "learning_rate": 0.001758,
      "loss": 1.4044,
      "step": 586
    },
    {
      "epoch": 0.051475071557858076,
      "grad_norm": 0.142578125,
      "learning_rate": 0.001761,
      "loss": 1.2848,
      "step": 587
    },
    {
      "epoch": 0.05156276333223262,
      "grad_norm": 0.203125,
      "learning_rate": 0.001764,
      "loss": 1.3604,
      "step": 588
    },
    {
      "epoch": 0.051650455106607165,
      "grad_norm": 0.1396484375,
      "learning_rate": 0.001767,
      "loss": 1.265,
      "step": 589
    },
    {
      "epoch": 0.05173814688098171,
      "grad_norm": 0.1708984375,
      "learning_rate": 0.0017699999999999999,
      "loss": 1.3369,
      "step": 590
    },
    {
      "epoch": 0.051825838655356254,
      "grad_norm": 0.1123046875,
      "learning_rate": 0.001773,
      "loss": 1.3202,
      "step": 591
    },
    {
      "epoch": 0.0519135304297308,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.001776,
      "loss": 1.3488,
      "step": 592
    },
    {
      "epoch": 0.05200122220410534,
      "grad_norm": 0.1435546875,
      "learning_rate": 0.001779,
      "loss": 1.3653,
      "step": 593
    },
    {
      "epoch": 0.05208891397847989,
      "grad_norm": 0.111328125,
      "learning_rate": 0.001782,
      "loss": 1.3817,
      "step": 594
    },
    {
      "epoch": 0.05217660575285443,
      "grad_norm": 0.1259765625,
      "learning_rate": 0.001785,
      "loss": 1.3422,
      "step": 595
    },
    {
      "epoch": 0.052264297527228984,
      "grad_norm": 0.10302734375,
      "learning_rate": 0.0017879999999999999,
      "loss": 1.3877,
      "step": 596
    },
    {
      "epoch": 0.05235198930160353,
      "grad_norm": 0.1318359375,
      "learning_rate": 0.001791,
      "loss": 1.3137,
      "step": 597
    },
    {
      "epoch": 0.05243968107597807,
      "grad_norm": 0.10986328125,
      "learning_rate": 0.001794,
      "loss": 1.3798,
      "step": 598
    },
    {
      "epoch": 0.05252737285035262,
      "grad_norm": 0.11474609375,
      "learning_rate": 0.001797,
      "loss": 1.3497,
      "step": 599
    },
    {
      "epoch": 0.05261506462472716,
      "grad_norm": 0.1142578125,
      "learning_rate": 0.0018,
      "loss": 1.3417,
      "step": 600
    },
    {
      "epoch": 0.052702756399101706,
      "grad_norm": 0.125,
      "learning_rate": 0.001803,
      "loss": 1.3298,
      "step": 601
    },
    {
      "epoch": 0.05279044817347625,
      "grad_norm": 0.10595703125,
      "learning_rate": 0.0018059999999999999,
      "loss": 1.3585,
      "step": 602
    },
    {
      "epoch": 0.052878139947850795,
      "grad_norm": 0.11279296875,
      "learning_rate": 0.001809,
      "loss": 1.331,
      "step": 603
    },
    {
      "epoch": 0.05296583172222534,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.001812,
      "loss": 1.2992,
      "step": 604
    },
    {
      "epoch": 0.05305352349659989,
      "grad_norm": 0.09619140625,
      "learning_rate": 0.001815,
      "loss": 1.2915,
      "step": 605
    },
    {
      "epoch": 0.053141215270974436,
      "grad_norm": 0.09033203125,
      "learning_rate": 0.001818,
      "loss": 1.32,
      "step": 606
    },
    {
      "epoch": 0.05322890704534898,
      "grad_norm": 0.0771484375,
      "learning_rate": 0.001821,
      "loss": 1.3601,
      "step": 607
    },
    {
      "epoch": 0.053316598819723525,
      "grad_norm": 0.0927734375,
      "learning_rate": 0.001824,
      "loss": 1.3336,
      "step": 608
    },
    {
      "epoch": 0.05340429059409807,
      "grad_norm": 0.09765625,
      "learning_rate": 0.001827,
      "loss": 1.347,
      "step": 609
    },
    {
      "epoch": 0.053491982368472614,
      "grad_norm": 0.1650390625,
      "learning_rate": 0.00183,
      "loss": 1.3347,
      "step": 610
    },
    {
      "epoch": 0.05357967414284716,
      "grad_norm": 0.10546875,
      "learning_rate": 0.001833,
      "loss": 1.3217,
      "step": 611
    },
    {
      "epoch": 0.0536673659172217,
      "grad_norm": 0.09716796875,
      "learning_rate": 0.001836,
      "loss": 1.2653,
      "step": 612
    },
    {
      "epoch": 0.05375505769159625,
      "grad_norm": 0.11083984375,
      "learning_rate": 0.001839,
      "loss": 1.3131,
      "step": 613
    },
    {
      "epoch": 0.0538427494659708,
      "grad_norm": 0.166015625,
      "learning_rate": 0.001842,
      "loss": 1.3646,
      "step": 614
    },
    {
      "epoch": 0.05393044124034534,
      "grad_norm": 0.091796875,
      "learning_rate": 0.001845,
      "loss": 1.3012,
      "step": 615
    },
    {
      "epoch": 0.05401813301471989,
      "grad_norm": 0.11181640625,
      "learning_rate": 0.001848,
      "loss": 1.2882,
      "step": 616
    },
    {
      "epoch": 0.05410582478909443,
      "grad_norm": 0.0927734375,
      "learning_rate": 0.001851,
      "loss": 1.2802,
      "step": 617
    },
    {
      "epoch": 0.05419351656346898,
      "grad_norm": 0.107421875,
      "learning_rate": 0.001854,
      "loss": 1.3446,
      "step": 618
    },
    {
      "epoch": 0.05428120833784352,
      "grad_norm": 0.138671875,
      "learning_rate": 0.001857,
      "loss": 1.3352,
      "step": 619
    },
    {
      "epoch": 0.054368900112218066,
      "grad_norm": 0.14453125,
      "learning_rate": 0.00186,
      "loss": 1.3795,
      "step": 620
    },
    {
      "epoch": 0.05445659188659261,
      "grad_norm": 0.142578125,
      "learning_rate": 0.001863,
      "loss": 1.3184,
      "step": 621
    },
    {
      "epoch": 0.054544283660967155,
      "grad_norm": 0.12890625,
      "learning_rate": 0.001866,
      "loss": 1.306,
      "step": 622
    },
    {
      "epoch": 0.054631975435341706,
      "grad_norm": 0.1357421875,
      "learning_rate": 0.001869,
      "loss": 1.3334,
      "step": 623
    },
    {
      "epoch": 0.05471966720971625,
      "grad_norm": 0.09765625,
      "learning_rate": 0.001872,
      "loss": 1.3268,
      "step": 624
    },
    {
      "epoch": 0.054807358984090795,
      "grad_norm": 0.11181640625,
      "learning_rate": 0.001875,
      "loss": 1.3805,
      "step": 625
    },
    {
      "epoch": 0.05489505075846534,
      "grad_norm": 0.083984375,
      "learning_rate": 0.0018780000000000001,
      "loss": 1.4231,
      "step": 626
    },
    {
      "epoch": 0.054982742532839884,
      "grad_norm": 0.1455078125,
      "learning_rate": 0.001881,
      "loss": 1.3484,
      "step": 627
    },
    {
      "epoch": 0.05507043430721443,
      "grad_norm": 0.1064453125,
      "learning_rate": 0.001884,
      "loss": 1.3057,
      "step": 628
    },
    {
      "epoch": 0.05515812608158897,
      "grad_norm": 0.1181640625,
      "learning_rate": 0.001887,
      "loss": 1.3342,
      "step": 629
    },
    {
      "epoch": 0.05524581785596352,
      "grad_norm": 0.1396484375,
      "learning_rate": 0.00189,
      "loss": 1.3539,
      "step": 630
    },
    {
      "epoch": 0.05533350963033806,
      "grad_norm": 0.1279296875,
      "learning_rate": 0.0018930000000000002,
      "loss": 1.3593,
      "step": 631
    },
    {
      "epoch": 0.055421201404712614,
      "grad_norm": 0.1279296875,
      "learning_rate": 0.0018960000000000001,
      "loss": 1.3262,
      "step": 632
    },
    {
      "epoch": 0.05550889317908716,
      "grad_norm": 0.126953125,
      "learning_rate": 0.001899,
      "loss": 1.3629,
      "step": 633
    },
    {
      "epoch": 0.0555965849534617,
      "grad_norm": 0.107421875,
      "learning_rate": 0.001902,
      "loss": 1.313,
      "step": 634
    },
    {
      "epoch": 0.05568427672783625,
      "grad_norm": 0.1376953125,
      "learning_rate": 0.001905,
      "loss": 1.3779,
      "step": 635
    },
    {
      "epoch": 0.05577196850221079,
      "grad_norm": 0.07958984375,
      "learning_rate": 0.001908,
      "loss": 1.3722,
      "step": 636
    },
    {
      "epoch": 0.055859660276585336,
      "grad_norm": 0.1494140625,
      "learning_rate": 0.0019110000000000002,
      "loss": 1.2903,
      "step": 637
    },
    {
      "epoch": 0.05594735205095988,
      "grad_norm": 0.09619140625,
      "learning_rate": 0.0019140000000000001,
      "loss": 1.2875,
      "step": 638
    },
    {
      "epoch": 0.056035043825334425,
      "grad_norm": 0.1337890625,
      "learning_rate": 0.001917,
      "loss": 1.3276,
      "step": 639
    },
    {
      "epoch": 0.05612273559970897,
      "grad_norm": 0.1142578125,
      "learning_rate": 0.00192,
      "loss": 1.2509,
      "step": 640
    },
    {
      "epoch": 0.05621042737408352,
      "grad_norm": 0.083984375,
      "learning_rate": 0.001923,
      "loss": 1.3149,
      "step": 641
    },
    {
      "epoch": 0.056298119148458066,
      "grad_norm": 0.1279296875,
      "learning_rate": 0.001926,
      "loss": 1.3123,
      "step": 642
    },
    {
      "epoch": 0.05638581092283261,
      "grad_norm": 0.09765625,
      "learning_rate": 0.0019290000000000002,
      "loss": 1.3152,
      "step": 643
    },
    {
      "epoch": 0.056473502697207155,
      "grad_norm": 0.1669921875,
      "learning_rate": 0.0019320000000000001,
      "loss": 1.337,
      "step": 644
    },
    {
      "epoch": 0.0565611944715817,
      "grad_norm": 0.11767578125,
      "learning_rate": 0.001935,
      "loss": 1.3222,
      "step": 645
    },
    {
      "epoch": 0.056648886245956244,
      "grad_norm": 0.1484375,
      "learning_rate": 0.001938,
      "loss": 1.2985,
      "step": 646
    },
    {
      "epoch": 0.05673657802033079,
      "grad_norm": 0.107421875,
      "learning_rate": 0.001941,
      "loss": 1.4031,
      "step": 647
    },
    {
      "epoch": 0.05682426979470533,
      "grad_norm": 0.1123046875,
      "learning_rate": 0.0019440000000000002,
      "loss": 1.2429,
      "step": 648
    },
    {
      "epoch": 0.05691196156907988,
      "grad_norm": 0.08154296875,
      "learning_rate": 0.0019470000000000002,
      "loss": 1.3375,
      "step": 649
    },
    {
      "epoch": 0.05699965334345443,
      "grad_norm": 0.12451171875,
      "learning_rate": 0.0019500000000000001,
      "loss": 1.3173,
      "step": 650
    },
    {
      "epoch": 0.05708734511782897,
      "grad_norm": 0.1171875,
      "learning_rate": 0.001953,
      "loss": 1.3144,
      "step": 651
    },
    {
      "epoch": 0.05717503689220352,
      "grad_norm": 0.12353515625,
      "learning_rate": 0.0019560000000000003,
      "loss": 1.4079,
      "step": 652
    },
    {
      "epoch": 0.05726272866657806,
      "grad_norm": 0.134765625,
      "learning_rate": 0.0019590000000000002,
      "loss": 1.4005,
      "step": 653
    },
    {
      "epoch": 0.05735042044095261,
      "grad_norm": 0.142578125,
      "learning_rate": 0.001962,
      "loss": 1.4134,
      "step": 654
    },
    {
      "epoch": 0.05743811221532715,
      "grad_norm": 0.169921875,
      "learning_rate": 0.001965,
      "loss": 1.3191,
      "step": 655
    },
    {
      "epoch": 0.057525803989701696,
      "grad_norm": 0.1220703125,
      "learning_rate": 0.001968,
      "loss": 1.3546,
      "step": 656
    },
    {
      "epoch": 0.05761349576407624,
      "grad_norm": 0.1513671875,
      "learning_rate": 0.001971,
      "loss": 1.3522,
      "step": 657
    },
    {
      "epoch": 0.057701187538450785,
      "grad_norm": 0.1484375,
      "learning_rate": 0.001974,
      "loss": 1.318,
      "step": 658
    },
    {
      "epoch": 0.057788879312825336,
      "grad_norm": 0.1103515625,
      "learning_rate": 0.001977,
      "loss": 1.3126,
      "step": 659
    },
    {
      "epoch": 0.05787657108719988,
      "grad_norm": 0.0947265625,
      "learning_rate": 0.00198,
      "loss": 1.3112,
      "step": 660
    },
    {
      "epoch": 0.057964262861574425,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.001983,
      "loss": 1.3347,
      "step": 661
    },
    {
      "epoch": 0.05805195463594897,
      "grad_norm": 0.125,
      "learning_rate": 0.0019860000000000004,
      "loss": 1.384,
      "step": 662
    },
    {
      "epoch": 0.058139646410323514,
      "grad_norm": 0.11279296875,
      "learning_rate": 0.0019890000000000003,
      "loss": 1.3456,
      "step": 663
    },
    {
      "epoch": 0.05822733818469806,
      "grad_norm": 0.1220703125,
      "learning_rate": 0.0019920000000000003,
      "loss": 1.3589,
      "step": 664
    },
    {
      "epoch": 0.0583150299590726,
      "grad_norm": 0.1025390625,
      "learning_rate": 0.0019950000000000002,
      "loss": 1.3744,
      "step": 665
    },
    {
      "epoch": 0.05840272173344715,
      "grad_norm": 0.1474609375,
      "learning_rate": 0.001998,
      "loss": 1.3614,
      "step": 666
    },
    {
      "epoch": 0.05849041350782169,
      "grad_norm": 0.10400390625,
      "learning_rate": 0.002001,
      "loss": 1.3101,
      "step": 667
    },
    {
      "epoch": 0.058578105282196244,
      "grad_norm": 0.11767578125,
      "learning_rate": 0.002004,
      "loss": 1.3496,
      "step": 668
    },
    {
      "epoch": 0.05866579705657079,
      "grad_norm": 0.09716796875,
      "learning_rate": 0.002007,
      "loss": 1.3225,
      "step": 669
    },
    {
      "epoch": 0.05875348883094533,
      "grad_norm": 0.1318359375,
      "learning_rate": 0.00201,
      "loss": 1.3426,
      "step": 670
    },
    {
      "epoch": 0.05884118060531988,
      "grad_norm": 0.09521484375,
      "learning_rate": 0.002013,
      "loss": 1.2791,
      "step": 671
    },
    {
      "epoch": 0.05892887237969442,
      "grad_norm": 0.11279296875,
      "learning_rate": 0.002016,
      "loss": 1.2956,
      "step": 672
    },
    {
      "epoch": 0.059016564154068966,
      "grad_norm": 0.10400390625,
      "learning_rate": 0.002019,
      "loss": 1.2956,
      "step": 673
    },
    {
      "epoch": 0.05910425592844351,
      "grad_norm": 0.09228515625,
      "learning_rate": 0.0020220000000000004,
      "loss": 1.3212,
      "step": 674
    },
    {
      "epoch": 0.059191947702818055,
      "grad_norm": 0.10498046875,
      "learning_rate": 0.0020250000000000003,
      "loss": 1.3192,
      "step": 675
    },
    {
      "epoch": 0.0592796394771926,
      "grad_norm": 0.07958984375,
      "learning_rate": 0.0020280000000000003,
      "loss": 1.373,
      "step": 676
    },
    {
      "epoch": 0.059367331251567144,
      "grad_norm": 0.09423828125,
      "learning_rate": 0.0020310000000000003,
      "loss": 1.4266,
      "step": 677
    },
    {
      "epoch": 0.059455023025941696,
      "grad_norm": 0.08447265625,
      "learning_rate": 0.0020340000000000002,
      "loss": 1.3394,
      "step": 678
    },
    {
      "epoch": 0.05954271480031624,
      "grad_norm": 0.10400390625,
      "learning_rate": 0.002037,
      "loss": 1.3182,
      "step": 679
    },
    {
      "epoch": 0.059630406574690785,
      "grad_norm": 0.11328125,
      "learning_rate": 0.00204,
      "loss": 1.3112,
      "step": 680
    },
    {
      "epoch": 0.05971809834906533,
      "grad_norm": 0.08740234375,
      "learning_rate": 0.002043,
      "loss": 1.2881,
      "step": 681
    },
    {
      "epoch": 0.059805790123439874,
      "grad_norm": 0.1279296875,
      "learning_rate": 0.002046,
      "loss": 1.3234,
      "step": 682
    },
    {
      "epoch": 0.05989348189781442,
      "grad_norm": 0.1220703125,
      "learning_rate": 0.002049,
      "loss": 1.3557,
      "step": 683
    },
    {
      "epoch": 0.05998117367218896,
      "grad_norm": 0.138671875,
      "learning_rate": 0.002052,
      "loss": 1.3126,
      "step": 684
    },
    {
      "epoch": 0.06006886544656351,
      "grad_norm": 0.1259765625,
      "learning_rate": 0.0020550000000000004,
      "loss": 1.3457,
      "step": 685
    },
    {
      "epoch": 0.06015655722093805,
      "grad_norm": 0.0986328125,
      "learning_rate": 0.0020580000000000004,
      "loss": 1.295,
      "step": 686
    },
    {
      "epoch": 0.0602442489953126,
      "grad_norm": 0.115234375,
      "learning_rate": 0.0020610000000000003,
      "loss": 1.3832,
      "step": 687
    },
    {
      "epoch": 0.06033194076968715,
      "grad_norm": 0.130859375,
      "learning_rate": 0.002064,
      "loss": 1.338,
      "step": 688
    },
    {
      "epoch": 0.06041963254406169,
      "grad_norm": 0.08984375,
      "learning_rate": 0.002067,
      "loss": 1.3339,
      "step": 689
    },
    {
      "epoch": 0.06050732431843624,
      "grad_norm": 0.138671875,
      "learning_rate": 0.00207,
      "loss": 1.3436,
      "step": 690
    },
    {
      "epoch": 0.06059501609281078,
      "grad_norm": 0.09423828125,
      "learning_rate": 0.0020729999999999998,
      "loss": 1.3324,
      "step": 691
    },
    {
      "epoch": 0.060682707867185326,
      "grad_norm": 0.1435546875,
      "learning_rate": 0.0020759999999999997,
      "loss": 1.3587,
      "step": 692
    },
    {
      "epoch": 0.06077039964155987,
      "grad_norm": 0.15234375,
      "learning_rate": 0.0020789999999999997,
      "loss": 1.3409,
      "step": 693
    },
    {
      "epoch": 0.060858091415934415,
      "grad_norm": 0.09716796875,
      "learning_rate": 0.002082,
      "loss": 1.358,
      "step": 694
    },
    {
      "epoch": 0.06094578319030896,
      "grad_norm": 0.11767578125,
      "learning_rate": 0.002085,
      "loss": 1.2721,
      "step": 695
    },
    {
      "epoch": 0.06103347496468351,
      "grad_norm": 0.1533203125,
      "learning_rate": 0.002088,
      "loss": 1.3178,
      "step": 696
    },
    {
      "epoch": 0.061121166739058055,
      "grad_norm": 0.1494140625,
      "learning_rate": 0.002091,
      "loss": 1.2921,
      "step": 697
    },
    {
      "epoch": 0.0612088585134326,
      "grad_norm": 0.07958984375,
      "learning_rate": 0.002094,
      "loss": 1.2661,
      "step": 698
    },
    {
      "epoch": 0.061296550287807144,
      "grad_norm": 0.1259765625,
      "learning_rate": 0.002097,
      "loss": 1.3789,
      "step": 699
    },
    {
      "epoch": 0.06138424206218169,
      "grad_norm": 0.0927734375,
      "learning_rate": 0.0021,
      "loss": 1.3957,
      "step": 700
    },
    {
      "epoch": 0.06147193383655623,
      "grad_norm": 0.125,
      "learning_rate": 0.002103,
      "loss": 1.348,
      "step": 701
    },
    {
      "epoch": 0.06155962561093078,
      "grad_norm": 0.103515625,
      "learning_rate": 0.002106,
      "loss": 1.3111,
      "step": 702
    },
    {
      "epoch": 0.06164731738530532,
      "grad_norm": 0.08251953125,
      "learning_rate": 0.0021089999999999998,
      "loss": 1.321,
      "step": 703
    },
    {
      "epoch": 0.06173500915967987,
      "grad_norm": 0.09814453125,
      "learning_rate": 0.0021119999999999997,
      "loss": 1.4042,
      "step": 704
    },
    {
      "epoch": 0.06182270093405442,
      "grad_norm": 0.12890625,
      "learning_rate": 0.002115,
      "loss": 1.2891,
      "step": 705
    },
    {
      "epoch": 0.06191039270842896,
      "grad_norm": 0.125,
      "learning_rate": 0.002118,
      "loss": 1.2994,
      "step": 706
    },
    {
      "epoch": 0.06199808448280351,
      "grad_norm": 0.1640625,
      "learning_rate": 0.002121,
      "loss": 1.4005,
      "step": 707
    },
    {
      "epoch": 0.06208577625717805,
      "grad_norm": 0.109375,
      "learning_rate": 0.002124,
      "loss": 1.3539,
      "step": 708
    },
    {
      "epoch": 0.062173468031552596,
      "grad_norm": 0.1513671875,
      "learning_rate": 0.002127,
      "loss": 1.3263,
      "step": 709
    },
    {
      "epoch": 0.06226115980592714,
      "grad_norm": 0.083984375,
      "learning_rate": 0.00213,
      "loss": 1.3272,
      "step": 710
    },
    {
      "epoch": 0.062348851580301685,
      "grad_norm": 0.11865234375,
      "learning_rate": 0.002133,
      "loss": 1.3914,
      "step": 711
    },
    {
      "epoch": 0.06243654335467623,
      "grad_norm": 0.0966796875,
      "learning_rate": 0.002136,
      "loss": 1.3741,
      "step": 712
    },
    {
      "epoch": 0.06252423512905078,
      "grad_norm": 0.1298828125,
      "learning_rate": 0.002139,
      "loss": 1.3691,
      "step": 713
    },
    {
      "epoch": 0.06261192690342532,
      "grad_norm": 0.08984375,
      "learning_rate": 0.002142,
      "loss": 1.3387,
      "step": 714
    },
    {
      "epoch": 0.06269961867779987,
      "grad_norm": 0.095703125,
      "learning_rate": 0.0021449999999999998,
      "loss": 1.3177,
      "step": 715
    },
    {
      "epoch": 0.06278731045217441,
      "grad_norm": 0.08447265625,
      "learning_rate": 0.002148,
      "loss": 1.3291,
      "step": 716
    },
    {
      "epoch": 0.06287500222654896,
      "grad_norm": 0.115234375,
      "learning_rate": 0.002151,
      "loss": 1.3442,
      "step": 717
    },
    {
      "epoch": 0.06296269400092351,
      "grad_norm": 0.08154296875,
      "learning_rate": 0.002154,
      "loss": 1.352,
      "step": 718
    },
    {
      "epoch": 0.06305038577529805,
      "grad_norm": 0.12255859375,
      "learning_rate": 0.002157,
      "loss": 1.3209,
      "step": 719
    },
    {
      "epoch": 0.0631380775496726,
      "grad_norm": 0.080078125,
      "learning_rate": 0.00216,
      "loss": 1.3168,
      "step": 720
    },
    {
      "epoch": 0.06322576932404714,
      "grad_norm": 0.12060546875,
      "learning_rate": 0.002163,
      "loss": 1.3624,
      "step": 721
    },
    {
      "epoch": 0.06331346109842169,
      "grad_norm": 0.09619140625,
      "learning_rate": 0.002166,
      "loss": 1.3289,
      "step": 722
    },
    {
      "epoch": 0.06340115287279623,
      "grad_norm": 0.1416015625,
      "learning_rate": 0.002169,
      "loss": 1.3062,
      "step": 723
    },
    {
      "epoch": 0.06348884464717078,
      "grad_norm": 0.09912109375,
      "learning_rate": 0.002172,
      "loss": 1.2927,
      "step": 724
    },
    {
      "epoch": 0.06357653642154532,
      "grad_norm": 0.126953125,
      "learning_rate": 0.002175,
      "loss": 1.3349,
      "step": 725
    },
    {
      "epoch": 0.06366422819591987,
      "grad_norm": 0.162109375,
      "learning_rate": 0.002178,
      "loss": 1.4218,
      "step": 726
    },
    {
      "epoch": 0.0637519199702944,
      "grad_norm": 0.23828125,
      "learning_rate": 0.0021809999999999998,
      "loss": 1.4004,
      "step": 727
    },
    {
      "epoch": 0.06383961174466896,
      "grad_norm": 0.1533203125,
      "learning_rate": 0.002184,
      "loss": 1.3236,
      "step": 728
    },
    {
      "epoch": 0.06392730351904351,
      "grad_norm": 0.1640625,
      "learning_rate": 0.002187,
      "loss": 1.2873,
      "step": 729
    },
    {
      "epoch": 0.06401499529341804,
      "grad_norm": 0.10400390625,
      "learning_rate": 0.00219,
      "loss": 1.3388,
      "step": 730
    },
    {
      "epoch": 0.0641026870677926,
      "grad_norm": 0.16796875,
      "learning_rate": 0.002193,
      "loss": 1.3179,
      "step": 731
    },
    {
      "epoch": 0.06419037884216713,
      "grad_norm": 0.1318359375,
      "learning_rate": 0.002196,
      "loss": 1.3483,
      "step": 732
    },
    {
      "epoch": 0.06427807061654169,
      "grad_norm": 0.1357421875,
      "learning_rate": 0.002199,
      "loss": 1.3194,
      "step": 733
    },
    {
      "epoch": 0.06436576239091622,
      "grad_norm": 0.1318359375,
      "learning_rate": 0.002202,
      "loss": 1.3404,
      "step": 734
    },
    {
      "epoch": 0.06445345416529077,
      "grad_norm": 0.125,
      "learning_rate": 0.002205,
      "loss": 1.2977,
      "step": 735
    },
    {
      "epoch": 0.06454114593966531,
      "grad_norm": 0.1083984375,
      "learning_rate": 0.002208,
      "loss": 1.362,
      "step": 736
    },
    {
      "epoch": 0.06462883771403986,
      "grad_norm": 0.09423828125,
      "learning_rate": 0.002211,
      "loss": 1.3174,
      "step": 737
    },
    {
      "epoch": 0.06471652948841441,
      "grad_norm": 0.1435546875,
      "learning_rate": 0.002214,
      "loss": 1.3242,
      "step": 738
    },
    {
      "epoch": 0.06480422126278895,
      "grad_norm": 0.10888671875,
      "learning_rate": 0.0022170000000000002,
      "loss": 1.296,
      "step": 739
    },
    {
      "epoch": 0.0648919130371635,
      "grad_norm": 0.08837890625,
      "learning_rate": 0.00222,
      "loss": 1.3848,
      "step": 740
    },
    {
      "epoch": 0.06497960481153804,
      "grad_norm": 0.10693359375,
      "learning_rate": 0.002223,
      "loss": 1.3586,
      "step": 741
    },
    {
      "epoch": 0.06506729658591259,
      "grad_norm": 0.09912109375,
      "learning_rate": 0.002226,
      "loss": 1.2866,
      "step": 742
    },
    {
      "epoch": 0.06515498836028713,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.002229,
      "loss": 1.2886,
      "step": 743
    },
    {
      "epoch": 0.06524268013466168,
      "grad_norm": 0.10009765625,
      "learning_rate": 0.002232,
      "loss": 1.3764,
      "step": 744
    },
    {
      "epoch": 0.06533037190903622,
      "grad_norm": 0.08837890625,
      "learning_rate": 0.002235,
      "loss": 1.3943,
      "step": 745
    },
    {
      "epoch": 0.06541806368341077,
      "grad_norm": 0.12890625,
      "learning_rate": 0.002238,
      "loss": 1.3645,
      "step": 746
    },
    {
      "epoch": 0.06550575545778532,
      "grad_norm": 0.0869140625,
      "learning_rate": 0.002241,
      "loss": 1.2752,
      "step": 747
    },
    {
      "epoch": 0.06559344723215986,
      "grad_norm": 0.11474609375,
      "learning_rate": 0.002244,
      "loss": 1.2695,
      "step": 748
    },
    {
      "epoch": 0.06568113900653441,
      "grad_norm": 0.1357421875,
      "learning_rate": 0.002247,
      "loss": 1.3008,
      "step": 749
    },
    {
      "epoch": 0.06576883078090895,
      "grad_norm": 0.126953125,
      "learning_rate": 0.0022500000000000003,
      "loss": 1.2855,
      "step": 750
    },
    {
      "epoch": 0.0658565225552835,
      "grad_norm": 0.09033203125,
      "learning_rate": 0.0022530000000000002,
      "loss": 1.3534,
      "step": 751
    },
    {
      "epoch": 0.06594421432965804,
      "grad_norm": 0.08642578125,
      "learning_rate": 0.002256,
      "loss": 1.3676,
      "step": 752
    },
    {
      "epoch": 0.06603190610403259,
      "grad_norm": 0.1279296875,
      "learning_rate": 0.002259,
      "loss": 1.2932,
      "step": 753
    },
    {
      "epoch": 0.06611959787840713,
      "grad_norm": 0.1298828125,
      "learning_rate": 0.002262,
      "loss": 1.3035,
      "step": 754
    },
    {
      "epoch": 0.06620728965278168,
      "grad_norm": 0.103515625,
      "learning_rate": 0.002265,
      "loss": 1.2848,
      "step": 755
    },
    {
      "epoch": 0.06629498142715623,
      "grad_norm": 0.126953125,
      "learning_rate": 0.002268,
      "loss": 1.2856,
      "step": 756
    },
    {
      "epoch": 0.06638267320153077,
      "grad_norm": 0.109375,
      "learning_rate": 0.002271,
      "loss": 1.3616,
      "step": 757
    },
    {
      "epoch": 0.06647036497590532,
      "grad_norm": 0.1259765625,
      "learning_rate": 0.002274,
      "loss": 1.3047,
      "step": 758
    },
    {
      "epoch": 0.06655805675027986,
      "grad_norm": 0.1689453125,
      "learning_rate": 0.002277,
      "loss": 1.4104,
      "step": 759
    },
    {
      "epoch": 0.06664574852465441,
      "grad_norm": 0.11865234375,
      "learning_rate": 0.00228,
      "loss": 1.2926,
      "step": 760
    },
    {
      "epoch": 0.06673344029902895,
      "grad_norm": 0.0986328125,
      "learning_rate": 0.002283,
      "loss": 1.3173,
      "step": 761
    },
    {
      "epoch": 0.0668211320734035,
      "grad_norm": 0.1796875,
      "learning_rate": 0.0022860000000000003,
      "loss": 1.3172,
      "step": 762
    },
    {
      "epoch": 0.06690882384777803,
      "grad_norm": 0.1201171875,
      "learning_rate": 0.0022890000000000002,
      "loss": 1.3202,
      "step": 763
    },
    {
      "epoch": 0.06699651562215259,
      "grad_norm": 0.1435546875,
      "learning_rate": 0.002292,
      "loss": 1.3341,
      "step": 764
    },
    {
      "epoch": 0.06708420739652714,
      "grad_norm": 0.1767578125,
      "learning_rate": 0.002295,
      "loss": 1.3316,
      "step": 765
    },
    {
      "epoch": 0.06717189917090167,
      "grad_norm": 0.09716796875,
      "learning_rate": 0.002298,
      "loss": 1.3074,
      "step": 766
    },
    {
      "epoch": 0.06725959094527623,
      "grad_norm": 0.26171875,
      "learning_rate": 0.002301,
      "loss": 1.3714,
      "step": 767
    },
    {
      "epoch": 0.06734728271965076,
      "grad_norm": 0.21484375,
      "learning_rate": 0.002304,
      "loss": 1.3505,
      "step": 768
    },
    {
      "epoch": 0.06743497449402532,
      "grad_norm": 0.091796875,
      "learning_rate": 0.002307,
      "loss": 1.2795,
      "step": 769
    },
    {
      "epoch": 0.06752266626839985,
      "grad_norm": 0.146484375,
      "learning_rate": 0.00231,
      "loss": 1.2834,
      "step": 770
    },
    {
      "epoch": 0.0676103580427744,
      "grad_norm": 0.09619140625,
      "learning_rate": 0.002313,
      "loss": 1.2901,
      "step": 771
    },
    {
      "epoch": 0.06769804981714894,
      "grad_norm": 0.11083984375,
      "learning_rate": 0.002316,
      "loss": 1.2928,
      "step": 772
    },
    {
      "epoch": 0.0677857415915235,
      "grad_norm": 0.0869140625,
      "learning_rate": 0.0023190000000000003,
      "loss": 1.3493,
      "step": 773
    },
    {
      "epoch": 0.06787343336589804,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.0023220000000000003,
      "loss": 1.3889,
      "step": 774
    },
    {
      "epoch": 0.06796112514027258,
      "grad_norm": 0.10009765625,
      "learning_rate": 0.0023250000000000002,
      "loss": 1.2773,
      "step": 775
    },
    {
      "epoch": 0.06804881691464713,
      "grad_norm": 0.08740234375,
      "learning_rate": 0.002328,
      "loss": 1.3119,
      "step": 776
    },
    {
      "epoch": 0.06813650868902167,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.002331,
      "loss": 1.3411,
      "step": 777
    },
    {
      "epoch": 0.06822420046339622,
      "grad_norm": 0.0771484375,
      "learning_rate": 0.002334,
      "loss": 1.3557,
      "step": 778
    },
    {
      "epoch": 0.06831189223777076,
      "grad_norm": 0.11669921875,
      "learning_rate": 0.002337,
      "loss": 1.3282,
      "step": 779
    },
    {
      "epoch": 0.06839958401214531,
      "grad_norm": 0.07666015625,
      "learning_rate": 0.00234,
      "loss": 1.3127,
      "step": 780
    },
    {
      "epoch": 0.06848727578651985,
      "grad_norm": 0.119140625,
      "learning_rate": 0.002343,
      "loss": 1.3505,
      "step": 781
    },
    {
      "epoch": 0.0685749675608944,
      "grad_norm": 0.08447265625,
      "learning_rate": 0.002346,
      "loss": 1.368,
      "step": 782
    },
    {
      "epoch": 0.06866265933526895,
      "grad_norm": 0.1572265625,
      "learning_rate": 0.002349,
      "loss": 1.3141,
      "step": 783
    },
    {
      "epoch": 0.06875035110964349,
      "grad_norm": 0.0830078125,
      "learning_rate": 0.002352,
      "loss": 1.3129,
      "step": 784
    },
    {
      "epoch": 0.06883804288401804,
      "grad_norm": 0.1240234375,
      "learning_rate": 0.0023550000000000003,
      "loss": 1.3506,
      "step": 785
    },
    {
      "epoch": 0.06892573465839258,
      "grad_norm": 0.0888671875,
      "learning_rate": 0.0023580000000000003,
      "loss": 1.3198,
      "step": 786
    },
    {
      "epoch": 0.06901342643276713,
      "grad_norm": 0.1162109375,
      "learning_rate": 0.0023610000000000003,
      "loss": 1.3936,
      "step": 787
    },
    {
      "epoch": 0.06910111820714167,
      "grad_norm": 0.09326171875,
      "learning_rate": 0.002364,
      "loss": 1.3044,
      "step": 788
    },
    {
      "epoch": 0.06918880998151622,
      "grad_norm": 0.13671875,
      "learning_rate": 0.002367,
      "loss": 1.3279,
      "step": 789
    },
    {
      "epoch": 0.06927650175589076,
      "grad_norm": 0.134765625,
      "learning_rate": 0.00237,
      "loss": 1.3941,
      "step": 790
    },
    {
      "epoch": 0.06936419353026531,
      "grad_norm": 0.08349609375,
      "learning_rate": 0.002373,
      "loss": 1.3437,
      "step": 791
    },
    {
      "epoch": 0.06945188530463986,
      "grad_norm": 0.0859375,
      "learning_rate": 0.002376,
      "loss": 1.4024,
      "step": 792
    },
    {
      "epoch": 0.0695395770790144,
      "grad_norm": 0.08642578125,
      "learning_rate": 0.002379,
      "loss": 1.3566,
      "step": 793
    },
    {
      "epoch": 0.06962726885338895,
      "grad_norm": 0.09814453125,
      "learning_rate": 0.002382,
      "loss": 1.2291,
      "step": 794
    },
    {
      "epoch": 0.06971496062776349,
      "grad_norm": 0.0849609375,
      "learning_rate": 0.002385,
      "loss": 1.3785,
      "step": 795
    },
    {
      "epoch": 0.06980265240213804,
      "grad_norm": 0.1103515625,
      "learning_rate": 0.0023880000000000004,
      "loss": 1.3573,
      "step": 796
    },
    {
      "epoch": 0.06989034417651258,
      "grad_norm": 0.140625,
      "learning_rate": 0.0023910000000000003,
      "loss": 1.3114,
      "step": 797
    },
    {
      "epoch": 0.06997803595088713,
      "grad_norm": 0.09228515625,
      "learning_rate": 0.0023940000000000003,
      "loss": 1.3111,
      "step": 798
    },
    {
      "epoch": 0.07006572772526166,
      "grad_norm": 0.1728515625,
      "learning_rate": 0.0023970000000000003,
      "loss": 1.416,
      "step": 799
    },
    {
      "epoch": 0.07015341949963622,
      "grad_norm": 0.181640625,
      "learning_rate": 0.0024000000000000002,
      "loss": 1.2953,
      "step": 800
    },
    {
      "epoch": 0.07024111127401077,
      "grad_norm": 0.150390625,
      "learning_rate": 0.002403,
      "loss": 1.3164,
      "step": 801
    },
    {
      "epoch": 0.0703288030483853,
      "grad_norm": 0.205078125,
      "learning_rate": 0.002406,
      "loss": 1.44,
      "step": 802
    },
    {
      "epoch": 0.07041649482275986,
      "grad_norm": 0.12158203125,
      "learning_rate": 0.002409,
      "loss": 1.3285,
      "step": 803
    },
    {
      "epoch": 0.0705041865971344,
      "grad_norm": 0.11962890625,
      "learning_rate": 0.002412,
      "loss": 1.2978,
      "step": 804
    },
    {
      "epoch": 0.07059187837150895,
      "grad_norm": 0.10791015625,
      "learning_rate": 0.002415,
      "loss": 1.3385,
      "step": 805
    },
    {
      "epoch": 0.07067957014588348,
      "grad_norm": 0.16796875,
      "learning_rate": 0.002418,
      "loss": 1.2502,
      "step": 806
    },
    {
      "epoch": 0.07076726192025803,
      "grad_norm": 0.09619140625,
      "learning_rate": 0.0024210000000000004,
      "loss": 1.2764,
      "step": 807
    },
    {
      "epoch": 0.07085495369463257,
      "grad_norm": 0.1884765625,
      "learning_rate": 0.0024240000000000004,
      "loss": 1.306,
      "step": 808
    },
    {
      "epoch": 0.07094264546900712,
      "grad_norm": 0.103515625,
      "learning_rate": 0.0024270000000000003,
      "loss": 1.2523,
      "step": 809
    },
    {
      "epoch": 0.07103033724338167,
      "grad_norm": 0.251953125,
      "learning_rate": 0.0024300000000000003,
      "loss": 1.3436,
      "step": 810
    },
    {
      "epoch": 0.07111802901775621,
      "grad_norm": 0.076171875,
      "learning_rate": 0.0024330000000000003,
      "loss": 1.329,
      "step": 811
    },
    {
      "epoch": 0.07120572079213076,
      "grad_norm": 0.2578125,
      "learning_rate": 0.0024360000000000002,
      "loss": 1.3534,
      "step": 812
    },
    {
      "epoch": 0.0712934125665053,
      "grad_norm": 0.0869140625,
      "learning_rate": 0.0024389999999999998,
      "loss": 1.2848,
      "step": 813
    },
    {
      "epoch": 0.07138110434087985,
      "grad_norm": 0.1787109375,
      "learning_rate": 0.0024419999999999997,
      "loss": 1.3431,
      "step": 814
    },
    {
      "epoch": 0.07146879611525439,
      "grad_norm": 0.09716796875,
      "learning_rate": 0.0024449999999999997,
      "loss": 1.2798,
      "step": 815
    },
    {
      "epoch": 0.07155648788962894,
      "grad_norm": 0.1640625,
      "learning_rate": 0.002448,
      "loss": 1.3048,
      "step": 816
    },
    {
      "epoch": 0.07164417966400348,
      "grad_norm": 0.11865234375,
      "learning_rate": 0.002451,
      "loss": 1.2944,
      "step": 817
    },
    {
      "epoch": 0.07173187143837803,
      "grad_norm": 0.296875,
      "learning_rate": 0.002454,
      "loss": 1.4416,
      "step": 818
    },
    {
      "epoch": 0.07181956321275258,
      "grad_norm": 0.1767578125,
      "learning_rate": 0.002457,
      "loss": 1.3192,
      "step": 819
    },
    {
      "epoch": 0.07190725498712712,
      "grad_norm": 0.2138671875,
      "learning_rate": 0.00246,
      "loss": 1.3287,
      "step": 820
    },
    {
      "epoch": 0.07199494676150167,
      "grad_norm": 0.1591796875,
      "learning_rate": 0.002463,
      "loss": 1.3128,
      "step": 821
    },
    {
      "epoch": 0.07208263853587621,
      "grad_norm": 0.162109375,
      "learning_rate": 0.002466,
      "loss": 1.3772,
      "step": 822
    },
    {
      "epoch": 0.07217033031025076,
      "grad_norm": 0.140625,
      "learning_rate": 0.002469,
      "loss": 1.3326,
      "step": 823
    },
    {
      "epoch": 0.0722580220846253,
      "grad_norm": 0.16796875,
      "learning_rate": 0.002472,
      "loss": 1.3595,
      "step": 824
    },
    {
      "epoch": 0.07234571385899985,
      "grad_norm": 0.12060546875,
      "learning_rate": 0.0024749999999999998,
      "loss": 1.3271,
      "step": 825
    },
    {
      "epoch": 0.07243340563337439,
      "grad_norm": 0.169921875,
      "learning_rate": 0.0024779999999999997,
      "loss": 1.3399,
      "step": 826
    },
    {
      "epoch": 0.07252109740774894,
      "grad_norm": 0.142578125,
      "learning_rate": 0.002481,
      "loss": 1.3322,
      "step": 827
    },
    {
      "epoch": 0.07260878918212349,
      "grad_norm": 0.20703125,
      "learning_rate": 0.002484,
      "loss": 1.3868,
      "step": 828
    },
    {
      "epoch": 0.07269648095649803,
      "grad_norm": 0.1767578125,
      "learning_rate": 0.002487,
      "loss": 1.3419,
      "step": 829
    },
    {
      "epoch": 0.07278417273087258,
      "grad_norm": 0.1865234375,
      "learning_rate": 0.00249,
      "loss": 1.3186,
      "step": 830
    },
    {
      "epoch": 0.07287186450524712,
      "grad_norm": 0.1640625,
      "learning_rate": 0.002493,
      "loss": 1.2891,
      "step": 831
    },
    {
      "epoch": 0.07295955627962167,
      "grad_norm": 0.154296875,
      "learning_rate": 0.002496,
      "loss": 1.3262,
      "step": 832
    },
    {
      "epoch": 0.0730472480539962,
      "grad_norm": 0.12451171875,
      "learning_rate": 0.002499,
      "loss": 1.3283,
      "step": 833
    },
    {
      "epoch": 0.07313493982837076,
      "grad_norm": 0.10009765625,
      "learning_rate": 0.002502,
      "loss": 1.3316,
      "step": 834
    },
    {
      "epoch": 0.0732226316027453,
      "grad_norm": 0.0888671875,
      "learning_rate": 0.002505,
      "loss": 1.324,
      "step": 835
    },
    {
      "epoch": 0.07331032337711985,
      "grad_norm": 0.08984375,
      "learning_rate": 0.002508,
      "loss": 1.3424,
      "step": 836
    },
    {
      "epoch": 0.0733980151514944,
      "grad_norm": 0.09765625,
      "learning_rate": 0.0025109999999999998,
      "loss": 1.3264,
      "step": 837
    },
    {
      "epoch": 0.07348570692586893,
      "grad_norm": 0.08740234375,
      "learning_rate": 0.0025139999999999997,
      "loss": 1.3647,
      "step": 838
    },
    {
      "epoch": 0.07357339870024349,
      "grad_norm": 0.10693359375,
      "learning_rate": 0.002517,
      "loss": 1.3417,
      "step": 839
    },
    {
      "epoch": 0.07366109047461802,
      "grad_norm": 0.10888671875,
      "learning_rate": 0.00252,
      "loss": 1.2623,
      "step": 840
    },
    {
      "epoch": 0.07374878224899258,
      "grad_norm": 0.142578125,
      "learning_rate": 0.002523,
      "loss": 1.3707,
      "step": 841
    },
    {
      "epoch": 0.07383647402336711,
      "grad_norm": 0.10400390625,
      "learning_rate": 0.002526,
      "loss": 1.398,
      "step": 842
    },
    {
      "epoch": 0.07392416579774166,
      "grad_norm": 0.09912109375,
      "learning_rate": 0.002529,
      "loss": 1.319,
      "step": 843
    },
    {
      "epoch": 0.0740118575721162,
      "grad_norm": 0.11962890625,
      "learning_rate": 0.002532,
      "loss": 1.324,
      "step": 844
    },
    {
      "epoch": 0.07409954934649075,
      "grad_norm": 0.16015625,
      "learning_rate": 0.002535,
      "loss": 1.2863,
      "step": 845
    },
    {
      "epoch": 0.0741872411208653,
      "grad_norm": 0.1396484375,
      "learning_rate": 0.002538,
      "loss": 1.3849,
      "step": 846
    },
    {
      "epoch": 0.07427493289523984,
      "grad_norm": 0.12353515625,
      "learning_rate": 0.002541,
      "loss": 1.3583,
      "step": 847
    },
    {
      "epoch": 0.0743626246696144,
      "grad_norm": 0.1396484375,
      "learning_rate": 0.002544,
      "loss": 1.383,
      "step": 848
    },
    {
      "epoch": 0.07445031644398893,
      "grad_norm": 0.1298828125,
      "learning_rate": 0.002547,
      "loss": 1.2858,
      "step": 849
    },
    {
      "epoch": 0.07453800821836348,
      "grad_norm": 0.1904296875,
      "learning_rate": 0.00255,
      "loss": 1.3304,
      "step": 850
    },
    {
      "epoch": 0.07462569999273802,
      "grad_norm": 0.1796875,
      "learning_rate": 0.002553,
      "loss": 1.4514,
      "step": 851
    },
    {
      "epoch": 0.07471339176711257,
      "grad_norm": 0.177734375,
      "learning_rate": 0.002556,
      "loss": 1.3373,
      "step": 852
    },
    {
      "epoch": 0.07480108354148711,
      "grad_norm": 0.1875,
      "learning_rate": 0.002559,
      "loss": 1.4162,
      "step": 853
    },
    {
      "epoch": 0.07488877531586166,
      "grad_norm": 0.1484375,
      "learning_rate": 0.002562,
      "loss": 1.3607,
      "step": 854
    },
    {
      "epoch": 0.07497646709023621,
      "grad_norm": 0.1435546875,
      "learning_rate": 0.002565,
      "loss": 1.3378,
      "step": 855
    },
    {
      "epoch": 0.07506415886461075,
      "grad_norm": 0.1435546875,
      "learning_rate": 0.002568,
      "loss": 1.3712,
      "step": 856
    },
    {
      "epoch": 0.0751518506389853,
      "grad_norm": 0.111328125,
      "learning_rate": 0.002571,
      "loss": 1.3485,
      "step": 857
    },
    {
      "epoch": 0.07523954241335984,
      "grad_norm": 0.0869140625,
      "learning_rate": 0.002574,
      "loss": 1.3551,
      "step": 858
    },
    {
      "epoch": 0.07532723418773439,
      "grad_norm": 0.072265625,
      "learning_rate": 0.002577,
      "loss": 1.3312,
      "step": 859
    },
    {
      "epoch": 0.07541492596210893,
      "grad_norm": 0.09912109375,
      "learning_rate": 0.00258,
      "loss": 1.3454,
      "step": 860
    },
    {
      "epoch": 0.07550261773648348,
      "grad_norm": 0.103515625,
      "learning_rate": 0.0025830000000000002,
      "loss": 1.3705,
      "step": 861
    },
    {
      "epoch": 0.07559030951085802,
      "grad_norm": 0.11181640625,
      "learning_rate": 0.002586,
      "loss": 1.2941,
      "step": 862
    },
    {
      "epoch": 0.07567800128523257,
      "grad_norm": 0.0947265625,
      "learning_rate": 0.002589,
      "loss": 1.3392,
      "step": 863
    },
    {
      "epoch": 0.07576569305960712,
      "grad_norm": 0.1357421875,
      "learning_rate": 0.002592,
      "loss": 1.3258,
      "step": 864
    },
    {
      "epoch": 0.07585338483398166,
      "grad_norm": 0.07568359375,
      "learning_rate": 0.002595,
      "loss": 1.288,
      "step": 865
    },
    {
      "epoch": 0.07594107660835621,
      "grad_norm": 0.17578125,
      "learning_rate": 0.002598,
      "loss": 1.4316,
      "step": 866
    },
    {
      "epoch": 0.07602876838273075,
      "grad_norm": 0.130859375,
      "learning_rate": 0.002601,
      "loss": 1.3078,
      "step": 867
    },
    {
      "epoch": 0.0761164601571053,
      "grad_norm": 0.1552734375,
      "learning_rate": 0.002604,
      "loss": 1.2932,
      "step": 868
    },
    {
      "epoch": 0.07620415193147984,
      "grad_norm": 0.1669921875,
      "learning_rate": 0.002607,
      "loss": 1.3165,
      "step": 869
    },
    {
      "epoch": 0.07629184370585439,
      "grad_norm": 0.123046875,
      "learning_rate": 0.00261,
      "loss": 1.3193,
      "step": 870
    },
    {
      "epoch": 0.07637953548022892,
      "grad_norm": 0.09765625,
      "learning_rate": 0.002613,
      "loss": 1.3277,
      "step": 871
    },
    {
      "epoch": 0.07646722725460348,
      "grad_norm": 0.111328125,
      "learning_rate": 0.002616,
      "loss": 1.2847,
      "step": 872
    },
    {
      "epoch": 0.07655491902897803,
      "grad_norm": 0.12158203125,
      "learning_rate": 0.0026190000000000002,
      "loss": 1.335,
      "step": 873
    },
    {
      "epoch": 0.07664261080335256,
      "grad_norm": 0.09130859375,
      "learning_rate": 0.002622,
      "loss": 1.2934,
      "step": 874
    },
    {
      "epoch": 0.07673030257772712,
      "grad_norm": 0.11181640625,
      "learning_rate": 0.002625,
      "loss": 1.3118,
      "step": 875
    },
    {
      "epoch": 0.07681799435210165,
      "grad_norm": 0.08837890625,
      "learning_rate": 0.002628,
      "loss": 1.268,
      "step": 876
    },
    {
      "epoch": 0.0769056861264762,
      "grad_norm": 0.1630859375,
      "learning_rate": 0.002631,
      "loss": 1.3718,
      "step": 877
    },
    {
      "epoch": 0.07699337790085074,
      "grad_norm": 0.087890625,
      "learning_rate": 0.002634,
      "loss": 1.347,
      "step": 878
    },
    {
      "epoch": 0.0770810696752253,
      "grad_norm": 0.1025390625,
      "learning_rate": 0.002637,
      "loss": 1.321,
      "step": 879
    },
    {
      "epoch": 0.07716876144959983,
      "grad_norm": 0.10498046875,
      "learning_rate": 0.00264,
      "loss": 1.3698,
      "step": 880
    },
    {
      "epoch": 0.07725645322397438,
      "grad_norm": 0.1083984375,
      "learning_rate": 0.002643,
      "loss": 1.3034,
      "step": 881
    },
    {
      "epoch": 0.07734414499834893,
      "grad_norm": 0.08251953125,
      "learning_rate": 0.002646,
      "loss": 1.3008,
      "step": 882
    },
    {
      "epoch": 0.07743183677272347,
      "grad_norm": 0.107421875,
      "learning_rate": 0.002649,
      "loss": 1.3692,
      "step": 883
    },
    {
      "epoch": 0.07751952854709802,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.0026520000000000003,
      "loss": 1.3591,
      "step": 884
    },
    {
      "epoch": 0.07760722032147256,
      "grad_norm": 0.0732421875,
      "learning_rate": 0.0026550000000000002,
      "loss": 1.2839,
      "step": 885
    },
    {
      "epoch": 0.07769491209584711,
      "grad_norm": 0.10107421875,
      "learning_rate": 0.002658,
      "loss": 1.3315,
      "step": 886
    },
    {
      "epoch": 0.07778260387022165,
      "grad_norm": 0.10595703125,
      "learning_rate": 0.002661,
      "loss": 1.306,
      "step": 887
    },
    {
      "epoch": 0.0778702956445962,
      "grad_norm": 0.103515625,
      "learning_rate": 0.002664,
      "loss": 1.3655,
      "step": 888
    },
    {
      "epoch": 0.07795798741897074,
      "grad_norm": 0.0908203125,
      "learning_rate": 0.002667,
      "loss": 1.3113,
      "step": 889
    },
    {
      "epoch": 0.07804567919334529,
      "grad_norm": 0.1884765625,
      "learning_rate": 0.00267,
      "loss": 1.3352,
      "step": 890
    },
    {
      "epoch": 0.07813337096771984,
      "grad_norm": 0.1689453125,
      "learning_rate": 0.002673,
      "loss": 1.2859,
      "step": 891
    },
    {
      "epoch": 0.07822106274209438,
      "grad_norm": 0.10205078125,
      "learning_rate": 0.002676,
      "loss": 1.3539,
      "step": 892
    },
    {
      "epoch": 0.07830875451646893,
      "grad_norm": 0.234375,
      "learning_rate": 0.002679,
      "loss": 1.3543,
      "step": 893
    },
    {
      "epoch": 0.07839644629084347,
      "grad_norm": 0.2080078125,
      "learning_rate": 0.002682,
      "loss": 1.3276,
      "step": 894
    },
    {
      "epoch": 0.07848413806521802,
      "grad_norm": 0.10302734375,
      "learning_rate": 0.0026850000000000003,
      "loss": 1.3294,
      "step": 895
    },
    {
      "epoch": 0.07857182983959256,
      "grad_norm": 0.111328125,
      "learning_rate": 0.0026880000000000003,
      "loss": 1.2468,
      "step": 896
    },
    {
      "epoch": 0.07865952161396711,
      "grad_norm": 0.1845703125,
      "learning_rate": 0.0026910000000000002,
      "loss": 1.3238,
      "step": 897
    },
    {
      "epoch": 0.07874721338834165,
      "grad_norm": 0.1767578125,
      "learning_rate": 0.002694,
      "loss": 1.3838,
      "step": 898
    },
    {
      "epoch": 0.0788349051627162,
      "grad_norm": 0.205078125,
      "learning_rate": 0.002697,
      "loss": 1.3141,
      "step": 899
    },
    {
      "epoch": 0.07892259693709075,
      "grad_norm": 0.1455078125,
      "learning_rate": 0.0027,
      "loss": 1.2963,
      "step": 900
    },
    {
      "epoch": 0.07901028871146529,
      "grad_norm": 0.15234375,
      "learning_rate": 0.002703,
      "loss": 1.256,
      "step": 901
    },
    {
      "epoch": 0.07909798048583984,
      "grad_norm": 0.1259765625,
      "learning_rate": 0.002706,
      "loss": 1.3802,
      "step": 902
    },
    {
      "epoch": 0.07918567226021438,
      "grad_norm": 0.10693359375,
      "learning_rate": 0.002709,
      "loss": 1.2523,
      "step": 903
    },
    {
      "epoch": 0.07927336403458893,
      "grad_norm": 0.11083984375,
      "learning_rate": 0.002712,
      "loss": 1.2966,
      "step": 904
    },
    {
      "epoch": 0.07936105580896347,
      "grad_norm": 0.1103515625,
      "learning_rate": 0.002715,
      "loss": 1.2996,
      "step": 905
    },
    {
      "epoch": 0.07944874758333802,
      "grad_norm": 0.10986328125,
      "learning_rate": 0.002718,
      "loss": 1.3653,
      "step": 906
    },
    {
      "epoch": 0.07953643935771255,
      "grad_norm": 0.1396484375,
      "learning_rate": 0.0027210000000000003,
      "loss": 1.3168,
      "step": 907
    },
    {
      "epoch": 0.0796241311320871,
      "grad_norm": 0.095703125,
      "learning_rate": 0.0027240000000000003,
      "loss": 1.4055,
      "step": 908
    },
    {
      "epoch": 0.07971182290646166,
      "grad_norm": 0.1357421875,
      "learning_rate": 0.0027270000000000003,
      "loss": 1.3193,
      "step": 909
    },
    {
      "epoch": 0.0797995146808362,
      "grad_norm": 0.09912109375,
      "learning_rate": 0.0027300000000000002,
      "loss": 1.3323,
      "step": 910
    },
    {
      "epoch": 0.07988720645521075,
      "grad_norm": 0.193359375,
      "learning_rate": 0.002733,
      "loss": 1.361,
      "step": 911
    },
    {
      "epoch": 0.07997489822958528,
      "grad_norm": 0.255859375,
      "learning_rate": 0.002736,
      "loss": 1.3238,
      "step": 912
    },
    {
      "epoch": 0.08006259000395984,
      "grad_norm": 0.1005859375,
      "learning_rate": 0.002739,
      "loss": 1.3205,
      "step": 913
    },
    {
      "epoch": 0.08015028177833437,
      "grad_norm": 0.111328125,
      "learning_rate": 0.002742,
      "loss": 1.2774,
      "step": 914
    },
    {
      "epoch": 0.08023797355270892,
      "grad_norm": 0.09521484375,
      "learning_rate": 0.002745,
      "loss": 1.3301,
      "step": 915
    },
    {
      "epoch": 0.08032566532708346,
      "grad_norm": 0.09423828125,
      "learning_rate": 0.002748,
      "loss": 1.3214,
      "step": 916
    },
    {
      "epoch": 0.08041335710145801,
      "grad_norm": 0.1416015625,
      "learning_rate": 0.002751,
      "loss": 1.3142,
      "step": 917
    },
    {
      "epoch": 0.08050104887583256,
      "grad_norm": 0.07470703125,
      "learning_rate": 0.0027540000000000004,
      "loss": 1.3766,
      "step": 918
    },
    {
      "epoch": 0.0805887406502071,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.0027570000000000003,
      "loss": 1.2802,
      "step": 919
    },
    {
      "epoch": 0.08067643242458165,
      "grad_norm": 0.087890625,
      "learning_rate": 0.0027600000000000003,
      "loss": 1.3042,
      "step": 920
    },
    {
      "epoch": 0.08076412419895619,
      "grad_norm": 0.10986328125,
      "learning_rate": 0.0027630000000000003,
      "loss": 1.3302,
      "step": 921
    },
    {
      "epoch": 0.08085181597333074,
      "grad_norm": 0.1044921875,
      "learning_rate": 0.0027660000000000002,
      "loss": 1.3524,
      "step": 922
    },
    {
      "epoch": 0.08093950774770528,
      "grad_norm": 0.10888671875,
      "learning_rate": 0.002769,
      "loss": 1.229,
      "step": 923
    },
    {
      "epoch": 0.08102719952207983,
      "grad_norm": 0.10107421875,
      "learning_rate": 0.002772,
      "loss": 1.334,
      "step": 924
    },
    {
      "epoch": 0.08111489129645437,
      "grad_norm": 0.1318359375,
      "learning_rate": 0.002775,
      "loss": 1.3263,
      "step": 925
    },
    {
      "epoch": 0.08120258307082892,
      "grad_norm": 0.201171875,
      "learning_rate": 0.002778,
      "loss": 1.3455,
      "step": 926
    },
    {
      "epoch": 0.08129027484520347,
      "grad_norm": 0.0849609375,
      "learning_rate": 0.002781,
      "loss": 1.3536,
      "step": 927
    },
    {
      "epoch": 0.08137796661957801,
      "grad_norm": 0.11865234375,
      "learning_rate": 0.002784,
      "loss": 1.3306,
      "step": 928
    },
    {
      "epoch": 0.08146565839395256,
      "grad_norm": 0.0771484375,
      "learning_rate": 0.0027870000000000004,
      "loss": 1.3462,
      "step": 929
    },
    {
      "epoch": 0.0815533501683271,
      "grad_norm": 0.12890625,
      "learning_rate": 0.0027900000000000004,
      "loss": 1.2984,
      "step": 930
    },
    {
      "epoch": 0.08164104194270165,
      "grad_norm": 0.10205078125,
      "learning_rate": 0.0027930000000000003,
      "loss": 1.3302,
      "step": 931
    },
    {
      "epoch": 0.08172873371707619,
      "grad_norm": 0.1005859375,
      "learning_rate": 0.0027960000000000003,
      "loss": 1.356,
      "step": 932
    },
    {
      "epoch": 0.08181642549145074,
      "grad_norm": 0.091796875,
      "learning_rate": 0.0027990000000000003,
      "loss": 1.3102,
      "step": 933
    },
    {
      "epoch": 0.08190411726582528,
      "grad_norm": 0.10205078125,
      "learning_rate": 0.0028020000000000002,
      "loss": 1.4226,
      "step": 934
    },
    {
      "epoch": 0.08199180904019983,
      "grad_norm": 0.0869140625,
      "learning_rate": 0.002805,
      "loss": 1.3052,
      "step": 935
    },
    {
      "epoch": 0.08207950081457437,
      "grad_norm": 0.095703125,
      "learning_rate": 0.002808,
      "loss": 1.3552,
      "step": 936
    },
    {
      "epoch": 0.08216719258894892,
      "grad_norm": 0.07958984375,
      "learning_rate": 0.002811,
      "loss": 1.3334,
      "step": 937
    },
    {
      "epoch": 0.08225488436332347,
      "grad_norm": 0.15234375,
      "learning_rate": 0.002814,
      "loss": 1.3238,
      "step": 938
    },
    {
      "epoch": 0.082342576137698,
      "grad_norm": 0.08154296875,
      "learning_rate": 0.002817,
      "loss": 1.2994,
      "step": 939
    },
    {
      "epoch": 0.08243026791207256,
      "grad_norm": 0.12158203125,
      "learning_rate": 0.00282,
      "loss": 1.2673,
      "step": 940
    },
    {
      "epoch": 0.0825179596864471,
      "grad_norm": 0.126953125,
      "learning_rate": 0.002823,
      "loss": 1.2656,
      "step": 941
    },
    {
      "epoch": 0.08260565146082165,
      "grad_norm": 0.10009765625,
      "learning_rate": 0.002826,
      "loss": 1.3713,
      "step": 942
    },
    {
      "epoch": 0.08269334323519618,
      "grad_norm": 0.27734375,
      "learning_rate": 0.002829,
      "loss": 1.3689,
      "step": 943
    },
    {
      "epoch": 0.08278103500957074,
      "grad_norm": 0.1533203125,
      "learning_rate": 0.002832,
      "loss": 1.3078,
      "step": 944
    },
    {
      "epoch": 0.08286872678394527,
      "grad_norm": 0.201171875,
      "learning_rate": 0.002835,
      "loss": 1.3463,
      "step": 945
    },
    {
      "epoch": 0.08295641855831982,
      "grad_norm": 0.2109375,
      "learning_rate": 0.002838,
      "loss": 1.2728,
      "step": 946
    },
    {
      "epoch": 0.08304411033269438,
      "grad_norm": 0.10205078125,
      "learning_rate": 0.0028409999999999998,
      "loss": 1.2866,
      "step": 947
    },
    {
      "epoch": 0.08313180210706891,
      "grad_norm": 0.09375,
      "learning_rate": 0.0028439999999999997,
      "loss": 1.3144,
      "step": 948
    },
    {
      "epoch": 0.08321949388144347,
      "grad_norm": 0.10986328125,
      "learning_rate": 0.002847,
      "loss": 1.2829,
      "step": 949
    },
    {
      "epoch": 0.083307185655818,
      "grad_norm": 0.07763671875,
      "learning_rate": 0.00285,
      "loss": 1.3705,
      "step": 950
    },
    {
      "epoch": 0.08339487743019255,
      "grad_norm": 0.107421875,
      "learning_rate": 0.002853,
      "loss": 1.3597,
      "step": 951
    },
    {
      "epoch": 0.08348256920456709,
      "grad_norm": 0.10791015625,
      "learning_rate": 0.002856,
      "loss": 1.3528,
      "step": 952
    },
    {
      "epoch": 0.08357026097894164,
      "grad_norm": 0.126953125,
      "learning_rate": 0.002859,
      "loss": 1.3411,
      "step": 953
    },
    {
      "epoch": 0.08365795275331618,
      "grad_norm": 0.1376953125,
      "learning_rate": 0.002862,
      "loss": 1.3656,
      "step": 954
    },
    {
      "epoch": 0.08374564452769073,
      "grad_norm": 0.259765625,
      "learning_rate": 0.002865,
      "loss": 1.3907,
      "step": 955
    },
    {
      "epoch": 0.08383333630206528,
      "grad_norm": 0.10986328125,
      "learning_rate": 0.002868,
      "loss": 1.3499,
      "step": 956
    },
    {
      "epoch": 0.08392102807643982,
      "grad_norm": 0.13671875,
      "learning_rate": 0.002871,
      "loss": 1.326,
      "step": 957
    },
    {
      "epoch": 0.08400871985081437,
      "grad_norm": 0.11083984375,
      "learning_rate": 0.002874,
      "loss": 1.2958,
      "step": 958
    },
    {
      "epoch": 0.08409641162518891,
      "grad_norm": 0.087890625,
      "learning_rate": 0.002877,
      "loss": 1.2943,
      "step": 959
    },
    {
      "epoch": 0.08418410339956346,
      "grad_norm": 0.15625,
      "learning_rate": 0.0028799999999999997,
      "loss": 1.2968,
      "step": 960
    },
    {
      "epoch": 0.084271795173938,
      "grad_norm": 0.119140625,
      "learning_rate": 0.002883,
      "loss": 1.3424,
      "step": 961
    },
    {
      "epoch": 0.08435948694831255,
      "grad_norm": 0.130859375,
      "learning_rate": 0.002886,
      "loss": 1.326,
      "step": 962
    },
    {
      "epoch": 0.08444717872268709,
      "grad_norm": 0.1845703125,
      "learning_rate": 0.002889,
      "loss": 1.3498,
      "step": 963
    },
    {
      "epoch": 0.08453487049706164,
      "grad_norm": 0.1787109375,
      "learning_rate": 0.002892,
      "loss": 1.2791,
      "step": 964
    },
    {
      "epoch": 0.08462256227143619,
      "grad_norm": 0.0849609375,
      "learning_rate": 0.002895,
      "loss": 1.3582,
      "step": 965
    },
    {
      "epoch": 0.08471025404581073,
      "grad_norm": 0.1337890625,
      "learning_rate": 0.002898,
      "loss": 1.3123,
      "step": 966
    },
    {
      "epoch": 0.08479794582018528,
      "grad_norm": 0.1689453125,
      "learning_rate": 0.002901,
      "loss": 1.3389,
      "step": 967
    },
    {
      "epoch": 0.08488563759455982,
      "grad_norm": 0.111328125,
      "learning_rate": 0.002904,
      "loss": 1.3073,
      "step": 968
    },
    {
      "epoch": 0.08497332936893437,
      "grad_norm": 0.388671875,
      "learning_rate": 0.002907,
      "loss": 1.3919,
      "step": 969
    },
    {
      "epoch": 0.08506102114330891,
      "grad_norm": 0.255859375,
      "learning_rate": 0.00291,
      "loss": 1.3619,
      "step": 970
    },
    {
      "epoch": 0.08514871291768346,
      "grad_norm": 0.1748046875,
      "learning_rate": 0.002913,
      "loss": 1.2457,
      "step": 971
    },
    {
      "epoch": 0.085236404692058,
      "grad_norm": 0.140625,
      "learning_rate": 0.002916,
      "loss": 1.3531,
      "step": 972
    },
    {
      "epoch": 0.08532409646643255,
      "grad_norm": 0.19921875,
      "learning_rate": 0.002919,
      "loss": 1.3493,
      "step": 973
    },
    {
      "epoch": 0.0854117882408071,
      "grad_norm": 0.1865234375,
      "learning_rate": 0.002922,
      "loss": 1.3564,
      "step": 974
    },
    {
      "epoch": 0.08549948001518164,
      "grad_norm": 0.1572265625,
      "learning_rate": 0.002925,
      "loss": 1.3556,
      "step": 975
    },
    {
      "epoch": 0.08558717178955619,
      "grad_norm": 0.10791015625,
      "learning_rate": 0.002928,
      "loss": 1.3064,
      "step": 976
    },
    {
      "epoch": 0.08567486356393073,
      "grad_norm": 0.181640625,
      "learning_rate": 0.002931,
      "loss": 1.3248,
      "step": 977
    },
    {
      "epoch": 0.08576255533830528,
      "grad_norm": 0.09619140625,
      "learning_rate": 0.002934,
      "loss": 1.2805,
      "step": 978
    },
    {
      "epoch": 0.08585024711267981,
      "grad_norm": 0.138671875,
      "learning_rate": 0.002937,
      "loss": 1.3346,
      "step": 979
    },
    {
      "epoch": 0.08593793888705437,
      "grad_norm": 0.08740234375,
      "learning_rate": 0.00294,
      "loss": 1.2495,
      "step": 980
    },
    {
      "epoch": 0.0860256306614289,
      "grad_norm": 0.10693359375,
      "learning_rate": 0.002943,
      "loss": 1.3576,
      "step": 981
    },
    {
      "epoch": 0.08611332243580345,
      "grad_norm": 0.0751953125,
      "learning_rate": 0.002946,
      "loss": 1.2991,
      "step": 982
    },
    {
      "epoch": 0.086201014210178,
      "grad_norm": 0.08935546875,
      "learning_rate": 0.0029490000000000002,
      "loss": 1.4072,
      "step": 983
    },
    {
      "epoch": 0.08628870598455254,
      "grad_norm": 0.10546875,
      "learning_rate": 0.002952,
      "loss": 1.3707,
      "step": 984
    },
    {
      "epoch": 0.0863763977589271,
      "grad_norm": 0.095703125,
      "learning_rate": 0.002955,
      "loss": 1.3455,
      "step": 985
    },
    {
      "epoch": 0.08646408953330163,
      "grad_norm": 0.099609375,
      "learning_rate": 0.002958,
      "loss": 1.3252,
      "step": 986
    },
    {
      "epoch": 0.08655178130767618,
      "grad_norm": 0.08935546875,
      "learning_rate": 0.002961,
      "loss": 1.3285,
      "step": 987
    },
    {
      "epoch": 0.08663947308205072,
      "grad_norm": 0.1328125,
      "learning_rate": 0.002964,
      "loss": 1.4067,
      "step": 988
    },
    {
      "epoch": 0.08672716485642527,
      "grad_norm": 0.10009765625,
      "learning_rate": 0.002967,
      "loss": 1.321,
      "step": 989
    },
    {
      "epoch": 0.08681485663079981,
      "grad_norm": 0.111328125,
      "learning_rate": 0.00297,
      "loss": 1.3814,
      "step": 990
    },
    {
      "epoch": 0.08690254840517436,
      "grad_norm": 0.1953125,
      "learning_rate": 0.002973,
      "loss": 1.3265,
      "step": 991
    },
    {
      "epoch": 0.08699024017954891,
      "grad_norm": 0.1865234375,
      "learning_rate": 0.002976,
      "loss": 1.4072,
      "step": 992
    },
    {
      "epoch": 0.08707793195392345,
      "grad_norm": 0.10302734375,
      "learning_rate": 0.002979,
      "loss": 1.3561,
      "step": 993
    },
    {
      "epoch": 0.087165623728298,
      "grad_norm": 0.08349609375,
      "learning_rate": 0.002982,
      "loss": 1.3047,
      "step": 994
    },
    {
      "epoch": 0.08725331550267254,
      "grad_norm": 0.0888671875,
      "learning_rate": 0.0029850000000000002,
      "loss": 1.2992,
      "step": 995
    },
    {
      "epoch": 0.08734100727704709,
      "grad_norm": 0.11572265625,
      "learning_rate": 0.002988,
      "loss": 1.3093,
      "step": 996
    },
    {
      "epoch": 0.08742869905142163,
      "grad_norm": 0.08984375,
      "learning_rate": 0.002991,
      "loss": 1.3291,
      "step": 997
    },
    {
      "epoch": 0.08751639082579618,
      "grad_norm": 0.1484375,
      "learning_rate": 0.002994,
      "loss": 1.4177,
      "step": 998
    },
    {
      "epoch": 0.08760408260017072,
      "grad_norm": 0.1513671875,
      "learning_rate": 0.002997,
      "loss": 1.2768,
      "step": 999
    },
    {
      "epoch": 0.08769177437454527,
      "grad_norm": 0.10205078125,
      "learning_rate": 0.003,
      "loss": 1.3108,
      "step": 1000
    },
    {
      "epoch": 0.08769177437454527,
      "eval_loss": 1.3424164056777954,
      "eval_runtime": 429.1223,
      "eval_samples_per_second": 33.666,
      "eval_steps_per_second": 8.417,
      "step": 1000
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 11403,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.948783792128e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}