{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9994810586403736,
  "eval_steps": 500,
  "global_step": 963,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 0.1219773218035698,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 1.3759,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.1272401511669159,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.3538,
      "step": 2
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.13286344707012177,
      "learning_rate": 3e-06,
      "loss": 1.3886,
      "step": 3
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.1291491985321045,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.3143,
      "step": 4
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.12544086575508118,
      "learning_rate": 5e-06,
      "loss": 1.317,
      "step": 5
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.15034854412078857,
      "learning_rate": 6e-06,
      "loss": 1.3537,
      "step": 6
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.14342203736305237,
      "learning_rate": 7.000000000000001e-06,
      "loss": 1.3083,
      "step": 7
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.1392456442117691,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.3501,
      "step": 8
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.13647185266017914,
      "learning_rate": 9e-06,
      "loss": 1.1859,
      "step": 9
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.12714740633964539,
      "learning_rate": 1e-05,
      "loss": 1.3746,
      "step": 10
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.12918396294116974,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 1.3672,
      "step": 11
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.2055857628583908,
      "learning_rate": 1.2e-05,
      "loss": 1.3692,
      "step": 12
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.1577647179365158,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 1.2316,
      "step": 13
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.1306232064962387,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 1.2453,
      "step": 14
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.1246161013841629,
      "learning_rate": 1.5e-05,
      "loss": 1.3613,
      "step": 15
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.1408628672361374,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.3905,
      "step": 16
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.1561778485774994,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 1.2617,
      "step": 17
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.13680391013622284,
      "learning_rate": 1.8e-05,
      "loss": 1.3604,
      "step": 18
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.12442106008529663,
      "learning_rate": 1.9e-05,
      "loss": 1.3812,
      "step": 19
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.13730080425739288,
      "learning_rate": 2e-05,
      "loss": 1.2252,
      "step": 20
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.11723917722702026,
      "learning_rate": 2.1e-05,
      "loss": 1.1512,
      "step": 21
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.13034303486347198,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 1.2698,
      "step": 22
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.15671290457248688,
      "learning_rate": 2.3000000000000003e-05,
      "loss": 1.2823,
      "step": 23
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.11263088881969452,
      "learning_rate": 2.4e-05,
      "loss": 1.375,
      "step": 24
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.13272984325885773,
      "learning_rate": 2.5e-05,
      "loss": 1.2721,
      "step": 25
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.13560321927070618,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 1.3223,
      "step": 26
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.13235026597976685,
      "learning_rate": 2.7000000000000002e-05,
      "loss": 1.2181,
      "step": 27
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.12842555344104767,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 1.3783,
      "step": 28
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.13284699618816376,
      "learning_rate": 2.9e-05,
      "loss": 1.3668,
      "step": 29
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.12673886120319366,
      "learning_rate": 3e-05,
      "loss": 1.3657,
      "step": 30
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.1453860104084015,
      "learning_rate": 3.1e-05,
      "loss": 1.2646,
      "step": 31
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.17890802025794983,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 1.1986,
      "step": 32
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.1185929998755455,
      "learning_rate": 3.3e-05,
      "loss": 1.1962,
      "step": 33
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.12994253635406494,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 1.2893,
      "step": 34
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.1479964405298233,
      "learning_rate": 3.5e-05,
      "loss": 1.1221,
      "step": 35
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.13566361367702484,
      "learning_rate": 3.6e-05,
      "loss": 1.2425,
      "step": 36
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.1559300720691681,
      "learning_rate": 3.7e-05,
      "loss": 1.2995,
      "step": 37
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.14187201857566833,
      "learning_rate": 3.8e-05,
      "loss": 1.3666,
      "step": 38
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.13634175062179565,
      "learning_rate": 3.9000000000000006e-05,
      "loss": 1.3111,
      "step": 39
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.1550116240978241,
      "learning_rate": 4e-05,
      "loss": 1.3729,
      "step": 40
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.14702185988426208,
      "learning_rate": 4.1e-05,
      "loss": 1.2447,
      "step": 41
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.1325961798429489,
      "learning_rate": 4.2e-05,
      "loss": 1.174,
      "step": 42
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.1384323686361313,
      "learning_rate": 4.3e-05,
      "loss": 1.2333,
      "step": 43
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.13662421703338623,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 1.1815,
      "step": 44
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.13229218125343323,
      "learning_rate": 4.5e-05,
      "loss": 1.2125,
      "step": 45
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.14347675442695618,
      "learning_rate": 4.600000000000001e-05,
      "loss": 1.1498,
      "step": 46
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.18536897003650665,
      "learning_rate": 4.7e-05,
      "loss": 1.2404,
      "step": 47
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.13089048862457275,
      "learning_rate": 4.8e-05,
      "loss": 1.2904,
      "step": 48
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.12752604484558105,
      "learning_rate": 4.9e-05,
      "loss": 1.2702,
      "step": 49
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.15005581080913544,
      "learning_rate": 5e-05,
      "loss": 1.2866,
      "step": 50
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.16374410688877106,
      "learning_rate": 5.1000000000000006e-05,
      "loss": 1.333,
      "step": 51
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.1684991866350174,
      "learning_rate": 5.2000000000000004e-05,
      "loss": 1.4278,
      "step": 52
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.1903846263885498,
      "learning_rate": 5.300000000000001e-05,
      "loss": 1.3035,
      "step": 53
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.16056384146213531,
      "learning_rate": 5.4000000000000005e-05,
      "loss": 1.2323,
      "step": 54
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.14987166225910187,
      "learning_rate": 5.500000000000001e-05,
      "loss": 1.0263,
      "step": 55
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.13468973338603973,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 1.2534,
      "step": 56
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.1467704474925995,
      "learning_rate": 5.6999999999999996e-05,
      "loss": 1.3025,
      "step": 57
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.17560411989688873,
      "learning_rate": 5.8e-05,
      "loss": 1.2307,
      "step": 58
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.1423458606004715,
      "learning_rate": 5.9e-05,
      "loss": 1.1628,
      "step": 59
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.12863807380199432,
      "learning_rate": 6e-05,
      "loss": 1.1174,
      "step": 60
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.13459883630275726,
      "learning_rate": 6.1e-05,
      "loss": 1.2223,
      "step": 61
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.13762715458869934,
      "learning_rate": 6.2e-05,
      "loss": 1.2393,
      "step": 62
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.1310354322195053,
      "learning_rate": 6.3e-05,
      "loss": 1.2487,
      "step": 63
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.12477163225412369,
      "learning_rate": 6.400000000000001e-05,
      "loss": 1.2632,
      "step": 64
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.15380260348320007,
      "learning_rate": 6.500000000000001e-05,
      "loss": 1.3762,
      "step": 65
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.1248747706413269,
      "learning_rate": 6.6e-05,
      "loss": 1.1336,
      "step": 66
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.12641173601150513,
      "learning_rate": 6.7e-05,
      "loss": 1.2674,
      "step": 67
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.13872478902339935,
      "learning_rate": 6.800000000000001e-05,
      "loss": 1.3014,
      "step": 68
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.1377810686826706,
      "learning_rate": 6.9e-05,
      "loss": 1.2816,
      "step": 69
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.13459351658821106,
      "learning_rate": 7e-05,
      "loss": 1.2872,
      "step": 70
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.13774238526821136,
      "learning_rate": 7.1e-05,
      "loss": 1.1481,
      "step": 71
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.1276445835828781,
      "learning_rate": 7.2e-05,
      "loss": 1.1924,
      "step": 72
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.12415210157632828,
      "learning_rate": 7.3e-05,
      "loss": 1.3093,
      "step": 73
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.11668894439935684,
      "learning_rate": 7.4e-05,
      "loss": 1.1941,
      "step": 74
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.16712161898612976,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.2581,
      "step": 75
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.1783616989850998,
      "learning_rate": 7.6e-05,
      "loss": 1.0885,
      "step": 76
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.1388709843158722,
      "learning_rate": 7.7e-05,
      "loss": 1.1999,
      "step": 77
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.13248670101165771,
      "learning_rate": 7.800000000000001e-05,
      "loss": 1.1312,
      "step": 78
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.8402003645896912,
      "learning_rate": 7.900000000000001e-05,
      "loss": 1.281,
      "step": 79
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.1560419201850891,
      "learning_rate": 8e-05,
      "loss": 1.3314,
      "step": 80
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.13449999690055847,
      "learning_rate": 8.1e-05,
      "loss": 1.1742,
      "step": 81
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.14383117854595184,
      "learning_rate": 8.2e-05,
      "loss": 1.1958,
      "step": 82
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.15058472752571106,
      "learning_rate": 8.3e-05,
      "loss": 1.2992,
      "step": 83
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.14776362478733063,
      "learning_rate": 8.4e-05,
      "loss": 1.3611,
      "step": 84
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.14592403173446655,
      "learning_rate": 8.5e-05,
      "loss": 1.1165,
      "step": 85
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.1357845813035965,
      "learning_rate": 8.6e-05,
      "loss": 1.2509,
      "step": 86
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.15681371092796326,
      "learning_rate": 8.7e-05,
      "loss": 1.0857,
      "step": 87
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.14496199786663055,
      "learning_rate": 8.800000000000001e-05,
      "loss": 1.1111,
      "step": 88
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.1501457244157791,
      "learning_rate": 8.900000000000001e-05,
      "loss": 1.2556,
      "step": 89
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.14578907191753387,
      "learning_rate": 9e-05,
      "loss": 1.1509,
      "step": 90
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.13890138268470764,
      "learning_rate": 9.1e-05,
      "loss": 1.1687,
      "step": 91
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.14948885142803192,
      "learning_rate": 9.200000000000001e-05,
      "loss": 1.0968,
      "step": 92
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.13117392361164093,
      "learning_rate": 9.300000000000001e-05,
      "loss": 1.3073,
      "step": 93
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.13922806084156036,
      "learning_rate": 9.4e-05,
      "loss": 1.2502,
      "step": 94
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.16392332315444946,
      "learning_rate": 9.5e-05,
      "loss": 1.3148,
      "step": 95
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.11756494641304016,
      "learning_rate": 9.6e-05,
      "loss": 1.3215,
      "step": 96
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.11909925937652588,
      "learning_rate": 9.7e-05,
      "loss": 1.2593,
      "step": 97
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.1371874213218689,
      "learning_rate": 9.8e-05,
      "loss": 1.1537,
      "step": 98
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.14141380786895752,
      "learning_rate": 9.900000000000001e-05,
      "loss": 1.221,
      "step": 99
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.13502179086208344,
      "learning_rate": 0.0001,
      "loss": 1.3492,
      "step": 100
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.15468691289424896,
      "learning_rate": 0.000101,
      "loss": 1.2592,
      "step": 101
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.13041648268699646,
      "learning_rate": 0.00010200000000000001,
      "loss": 1.3792,
      "step": 102
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.11658702045679092,
      "learning_rate": 0.00010300000000000001,
      "loss": 1.1845,
      "step": 103
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.13084186613559723,
      "learning_rate": 0.00010400000000000001,
      "loss": 1.1808,
      "step": 104
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.13180844485759735,
      "learning_rate": 0.000105,
      "loss": 1.1721,
      "step": 105
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.1375647485256195,
      "learning_rate": 0.00010600000000000002,
      "loss": 1.3622,
      "step": 106
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.11864122748374939,
      "learning_rate": 0.00010700000000000001,
      "loss": 1.238,
      "step": 107
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.11780542880296707,
      "learning_rate": 0.00010800000000000001,
      "loss": 1.1581,
      "step": 108
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.11449900269508362,
      "learning_rate": 0.000109,
      "loss": 1.3076,
      "step": 109
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.11585865169763565,
      "learning_rate": 0.00011000000000000002,
      "loss": 1.3553,
      "step": 110
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.12288644164800644,
      "learning_rate": 0.00011100000000000001,
      "loss": 1.1448,
      "step": 111
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.12131894379854202,
      "learning_rate": 0.00011200000000000001,
      "loss": 1.3456,
      "step": 112
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.13827624917030334,
      "learning_rate": 0.000113,
      "loss": 1.3183,
      "step": 113
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.13708753883838654,
      "learning_rate": 0.00011399999999999999,
      "loss": 1.3293,
      "step": 114
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.11893680691719055,
      "learning_rate": 0.00011499999999999999,
      "loss": 1.1954,
      "step": 115
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.12022116035223007,
      "learning_rate": 0.000116,
      "loss": 1.1822,
      "step": 116
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.11215274035930634,
      "learning_rate": 0.000117,
      "loss": 1.2453,
      "step": 117
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.11778578907251358,
      "learning_rate": 0.000118,
      "loss": 1.1939,
      "step": 118
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.12714551389217377,
      "learning_rate": 0.000119,
      "loss": 1.1917,
      "step": 119
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.11972431093454361,
      "learning_rate": 0.00012,
      "loss": 1.2769,
      "step": 120
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.12945646047592163,
      "learning_rate": 0.000121,
      "loss": 1.2929,
      "step": 121
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.11721435189247131,
      "learning_rate": 0.000122,
      "loss": 1.3606,
      "step": 122
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.117381751537323,
      "learning_rate": 0.000123,
      "loss": 1.3187,
      "step": 123
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.13110613822937012,
      "learning_rate": 0.000124,
      "loss": 1.3786,
      "step": 124
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.12571324408054352,
      "learning_rate": 0.000125,
      "loss": 1.0165,
      "step": 125
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.12401648610830307,
      "learning_rate": 0.000126,
      "loss": 1.1858,
      "step": 126
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.11960897594690323,
      "learning_rate": 0.000127,
      "loss": 1.2638,
      "step": 127
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.11195061355829239,
      "learning_rate": 0.00012800000000000002,
      "loss": 1.2469,
      "step": 128
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.1292635053396225,
      "learning_rate": 0.00012900000000000002,
      "loss": 1.3642,
      "step": 129
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.11140226572751999,
      "learning_rate": 0.00013000000000000002,
      "loss": 1.2238,
      "step": 130
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.10454066097736359,
      "learning_rate": 0.000131,
      "loss": 1.2418,
      "step": 131
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.11138935387134552,
      "learning_rate": 0.000132,
      "loss": 1.1964,
      "step": 132
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.12519143521785736,
      "learning_rate": 0.000133,
      "loss": 1.1889,
      "step": 133
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.1129814088344574,
      "learning_rate": 0.000134,
      "loss": 1.3425,
      "step": 134
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.11608937382698059,
      "learning_rate": 0.00013500000000000003,
      "loss": 1.0762,
      "step": 135
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.1410205215215683,
      "learning_rate": 0.00013600000000000003,
      "loss": 1.2502,
      "step": 136
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.11246364563703537,
      "learning_rate": 0.00013700000000000002,
      "loss": 1.3605,
      "step": 137
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.1213245838880539,
      "learning_rate": 0.000138,
      "loss": 1.2632,
      "step": 138
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.11144573241472244,
      "learning_rate": 0.000139,
      "loss": 1.1771,
      "step": 139
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.1700059324502945,
      "learning_rate": 0.00014,
      "loss": 1.3396,
      "step": 140
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.11950355768203735,
      "learning_rate": 0.000141,
      "loss": 1.1329,
      "step": 141
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.11367864161729813,
      "learning_rate": 0.000142,
      "loss": 1.3323,
      "step": 142
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.22367842495441437,
      "learning_rate": 0.000143,
      "loss": 1.201,
      "step": 143
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.11639752984046936,
      "learning_rate": 0.000144,
      "loss": 1.2507,
      "step": 144
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.12356352061033249,
      "learning_rate": 0.000145,
      "loss": 1.3437,
      "step": 145
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.10859037935733795,
      "learning_rate": 0.000146,
      "loss": 1.1677,
      "step": 146
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.11901260167360306,
      "learning_rate": 0.000147,
      "loss": 1.1125,
      "step": 147
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.1092861220240593,
      "learning_rate": 0.000148,
      "loss": 1.1939,
      "step": 148
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.10728005319833755,
      "learning_rate": 0.00014900000000000002,
      "loss": 1.1974,
      "step": 149
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.11113261431455612,
      "learning_rate": 0.00015000000000000001,
      "loss": 1.1107,
      "step": 150
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.14674580097198486,
      "learning_rate": 0.000151,
      "loss": 1.2339,
      "step": 151
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.10521555691957474,
      "learning_rate": 0.000152,
      "loss": 1.2917,
      "step": 152
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.12526080012321472,
      "learning_rate": 0.000153,
      "loss": 1.3247,
      "step": 153
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.2366744726896286,
      "learning_rate": 0.000154,
      "loss": 1.3637,
      "step": 154
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.11178407818078995,
      "learning_rate": 0.000155,
      "loss": 1.2649,
      "step": 155
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.10156022757291794,
      "learning_rate": 0.00015600000000000002,
      "loss": 1.1637,
      "step": 156
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.09755547344684601,
      "learning_rate": 0.00015700000000000002,
      "loss": 1.241,
      "step": 157
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.11344686150550842,
      "learning_rate": 0.00015800000000000002,
      "loss": 1.2208,
      "step": 158
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.1021842285990715,
      "learning_rate": 0.00015900000000000002,
      "loss": 1.1936,
      "step": 159
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.11096473783254623,
      "learning_rate": 0.00016,
      "loss": 1.2908,
      "step": 160
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.11365038901567459,
      "learning_rate": 0.000161,
      "loss": 1.1754,
      "step": 161
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.1159452423453331,
      "learning_rate": 0.000162,
      "loss": 1.2411,
      "step": 162
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.1240570917725563,
      "learning_rate": 0.000163,
      "loss": 1.2655,
      "step": 163
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.11343025416135788,
      "learning_rate": 0.000164,
      "loss": 1.3486,
      "step": 164
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.10576006770133972,
      "learning_rate": 0.000165,
      "loss": 1.2787,
      "step": 165
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.182003453373909,
      "learning_rate": 0.000166,
      "loss": 1.2045,
      "step": 166
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.10536637157201767,
      "learning_rate": 0.000167,
      "loss": 1.1388,
      "step": 167
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.10617152601480484,
      "learning_rate": 0.000168,
      "loss": 1.2686,
      "step": 168
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.11899720877408981,
      "learning_rate": 0.00016900000000000002,
      "loss": 1.1732,
      "step": 169
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.11736088991165161,
      "learning_rate": 0.00017,
      "loss": 1.2772,
      "step": 170
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.10500625520944595,
      "learning_rate": 0.000171,
      "loss": 1.2193,
      "step": 171
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.10299772769212723,
      "learning_rate": 0.000172,
      "loss": 1.2021,
      "step": 172
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.10550544410943985,
      "learning_rate": 0.000173,
      "loss": 1.2393,
      "step": 173
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.1132335364818573,
      "learning_rate": 0.000174,
      "loss": 1.2713,
      "step": 174
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.11581739038228989,
      "learning_rate": 0.000175,
      "loss": 1.3282,
      "step": 175
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.11211756616830826,
      "learning_rate": 0.00017600000000000002,
      "loss": 1.2374,
      "step": 176
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.10848426073789597,
      "learning_rate": 0.00017700000000000002,
      "loss": 1.2916,
      "step": 177
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.11354674398899078,
      "learning_rate": 0.00017800000000000002,
      "loss": 1.1914,
      "step": 178
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.12108771502971649,
      "learning_rate": 0.00017900000000000001,
      "loss": 1.3693,
      "step": 179
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.11068333685398102,
      "learning_rate": 0.00018,
      "loss": 1.086,
      "step": 180
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.11752796918153763,
      "learning_rate": 0.000181,
      "loss": 1.1508,
      "step": 181
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.10650038719177246,
      "learning_rate": 0.000182,
      "loss": 1.204,
      "step": 182
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.12248033285140991,
      "learning_rate": 0.000183,
      "loss": 1.2288,
      "step": 183
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.10320232063531876,
      "learning_rate": 0.00018400000000000003,
      "loss": 1.1309,
      "step": 184
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.1450732797384262,
      "learning_rate": 0.00018500000000000002,
      "loss": 1.2696,
      "step": 185
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.10568010061979294,
      "learning_rate": 0.00018600000000000002,
      "loss": 1.2643,
      "step": 186
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.12027282267808914,
      "learning_rate": 0.00018700000000000002,
      "loss": 1.2455,
      "step": 187
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.09627476334571838,
      "learning_rate": 0.000188,
      "loss": 1.1999,
      "step": 188
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.1297677904367447,
      "learning_rate": 0.00018899999999999999,
      "loss": 1.0843,
      "step": 189
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.1251845359802246,
      "learning_rate": 0.00019,
      "loss": 1.2957,
      "step": 190
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.11355076730251312,
      "learning_rate": 0.000191,
      "loss": 1.2705,
      "step": 191
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.11948814988136292,
      "learning_rate": 0.000192,
      "loss": 1.2352,
      "step": 192
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.12006771564483643,
      "learning_rate": 0.000193,
      "loss": 1.2489,
      "step": 193
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.11943700909614563,
      "learning_rate": 0.000194,
      "loss": 1.0768,
      "step": 194
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.12115556001663208,
      "learning_rate": 0.000195,
      "loss": 1.2614,
      "step": 195
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.1361999660730362,
      "learning_rate": 0.000196,
      "loss": 1.1418,
      "step": 196
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.10292702168226242,
      "learning_rate": 0.00019700000000000002,
      "loss": 1.1649,
      "step": 197
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.12569741904735565,
      "learning_rate": 0.00019800000000000002,
      "loss": 1.2054,
      "step": 198
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.09684847295284271,
      "learning_rate": 0.000199,
      "loss": 1.0854,
      "step": 199
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.10697442293167114,
      "learning_rate": 0.0002,
      "loss": 1.1895,
      "step": 200
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.10017522424459457,
      "learning_rate": 0.00019999915234318063,
      "loss": 1.2775,
      "step": 201
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.0991068184375763,
      "learning_rate": 0.0001999966093870929,
      "loss": 1.3151,
      "step": 202
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.10074901580810547,
      "learning_rate": 0.00019999237117484796,
      "loss": 1.2666,
      "step": 203
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.10549337416887283,
      "learning_rate": 0.00019998643777829672,
      "loss": 1.2483,
      "step": 204
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.10364524275064468,
      "learning_rate": 0.00019997880929802894,
      "loss": 1.0562,
      "step": 205
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.21587559580802917,
      "learning_rate": 0.00019996948586337125,
      "loss": 1.3881,
      "step": 206
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.10044362396001816,
      "learning_rate": 0.00019995846763238512,
      "loss": 1.2114,
      "step": 207
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.1037297397851944,
      "learning_rate": 0.0001999457547918641,
      "loss": 1.2611,
      "step": 208
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.11380340903997421,
      "learning_rate": 0.00019993134755733074,
      "loss": 1.3762,
      "step": 209
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.11318037658929825,
      "learning_rate": 0.0001999152461730328,
      "loss": 1.2906,
      "step": 210
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.1097273975610733,
      "learning_rate": 0.0001998974509119393,
      "loss": 1.0855,
      "step": 211
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.11358966678380966,
      "learning_rate": 0.00019987796207573573,
      "loss": 1.2061,
      "step": 212
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.11139972507953644,
      "learning_rate": 0.00019985677999481894,
      "loss": 1.2453,
      "step": 213
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.1275399923324585,
      "learning_rate": 0.00019983390502829166,
      "loss": 1.1917,
      "step": 214
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.1333334743976593,
      "learning_rate": 0.00019980933756395634,
      "loss": 1.2102,
      "step": 215
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.10892786085605621,
      "learning_rate": 0.00019978307801830854,
      "loss": 1.1177,
      "step": 216
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.10664794594049454,
      "learning_rate": 0.00019975512683652986,
      "loss": 1.1484,
      "step": 217
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.20084381103515625,
      "learning_rate": 0.0001997254844924806,
      "loss": 1.0399,
      "step": 218
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.11178454756736755,
      "learning_rate": 0.0001996941514886914,
      "loss": 1.3401,
      "step": 219
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.10734831541776657,
      "learning_rate": 0.00019966112835635493,
      "loss": 1.1964,
      "step": 220
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.12020622938871384,
      "learning_rate": 0.00019962641565531692,
      "loss": 1.1535,
      "step": 221
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.10945051908493042,
      "learning_rate": 0.00019959001397406646,
      "loss": 1.2086,
      "step": 222
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.11263303458690643,
      "learning_rate": 0.00019955192392972628,
      "loss": 1.1488,
      "step": 223
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.119475819170475,
      "learning_rate": 0.00019951214616804202,
      "loss": 1.3071,
      "step": 224
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.10934551060199738,
      "learning_rate": 0.00019947068136337158,
      "loss": 1.2639,
      "step": 225
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.11675988882780075,
      "learning_rate": 0.0001994275302186734,
      "loss": 1.142,
      "step": 226
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.28856417536735535,
      "learning_rate": 0.00019938269346549475,
      "loss": 1.2437,
      "step": 227
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.11922767013311386,
      "learning_rate": 0.00019933617186395917,
      "loss": 1.1668,
      "step": 228
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.12038363516330719,
      "learning_rate": 0.00019928796620275377,
      "loss": 1.0465,
      "step": 229
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.10866128653287888,
      "learning_rate": 0.00019923807729911567,
      "loss": 1.1417,
      "step": 230
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.134980708360672,
      "learning_rate": 0.00019918650599881827,
      "loss": 1.3574,
      "step": 231
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.11622115224599838,
      "learning_rate": 0.00019913325317615684,
      "loss": 1.2457,
      "step": 232
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.11134631186723709,
      "learning_rate": 0.0001990783197339338,
      "loss": 1.2194,
      "step": 233
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.11312945932149887,
      "learning_rate": 0.0001990217066034432,
      "loss": 1.179,
      "step": 234
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.10070719569921494,
      "learning_rate": 0.00019896341474445525,
      "loss": 1.299,
      "step": 235
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.11468905210494995,
      "learning_rate": 0.00019890344514519973,
      "loss": 1.2141,
      "step": 236
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.12275680154561996,
      "learning_rate": 0.00019884179882234944,
      "loss": 1.2581,
      "step": 237
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.1063394546508789,
      "learning_rate": 0.00019877847682100294,
      "loss": 1.3027,
      "step": 238
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.10849407315254211,
      "learning_rate": 0.00019871348021466672,
      "loss": 1.1914,
      "step": 239
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.1102900579571724,
      "learning_rate": 0.0001986468101052371,
      "loss": 1.2589,
      "step": 240
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.11001666635274887,
      "learning_rate": 0.00019857846762298155,
      "loss": 1.141,
      "step": 241
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.10907820612192154,
      "learning_rate": 0.0001985084539265195,
      "loss": 1.1139,
      "step": 242
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.1067519262433052,
      "learning_rate": 0.0001984367702028027,
      "loss": 1.2389,
      "step": 243
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.11710456013679504,
      "learning_rate": 0.0001983634176670951,
      "loss": 1.2421,
      "step": 244
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.10886801034212112,
      "learning_rate": 0.00019828839756295223,
      "loss": 1.1696,
      "step": 245
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.1404796689748764,
      "learning_rate": 0.00019821171116220012,
      "loss": 1.29,
      "step": 246
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.1096196323633194,
      "learning_rate": 0.00019813335976491385,
      "loss": 1.3536,
      "step": 247
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.11487864702939987,
      "learning_rate": 0.00019805334469939528,
      "loss": 1.1371,
      "step": 248
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.1555541455745697,
      "learning_rate": 0.00019797166732215076,
      "loss": 1.1812,
      "step": 249
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.10709488391876221,
      "learning_rate": 0.000197888329017868,
      "loss": 1.101,
      "step": 250
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.10238959640264511,
      "learning_rate": 0.00019780333119939262,
      "loss": 1.2363,
      "step": 251
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.12807561457157135,
      "learning_rate": 0.00019771667530770425,
      "loss": 1.217,
      "step": 252
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.10418405383825302,
      "learning_rate": 0.00019762836281189206,
      "loss": 1.36,
      "step": 253
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.10400766879320145,
      "learning_rate": 0.00019753839520912983,
      "loss": 1.2368,
      "step": 254
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.1629195362329483,
      "learning_rate": 0.00019744677402465052,
      "loss": 1.2236,
      "step": 255
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.11335594952106476,
      "learning_rate": 0.00019735350081172067,
      "loss": 1.2062,
      "step": 256
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.18272048234939575,
      "learning_rate": 0.00019725857715161373,
      "loss": 1.3138,
      "step": 257
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.10922593623399734,
      "learning_rate": 0.0001971620046535835,
      "loss": 1.1697,
      "step": 258
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.10762567073106766,
      "learning_rate": 0.00019706378495483665,
      "loss": 1.1303,
      "step": 259
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.11802787333726883,
      "learning_rate": 0.00019696391972050514,
      "loss": 1.2909,
      "step": 260
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.11715908348560333,
      "learning_rate": 0.00019686241064361792,
      "loss": 1.1456,
      "step": 261
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.11491627991199493,
      "learning_rate": 0.00019675925944507224,
      "loss": 1.2083,
      "step": 262
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.11131855100393295,
      "learning_rate": 0.0001966544678736044,
      "loss": 1.2579,
      "step": 263
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.12063883244991302,
      "learning_rate": 0.00019654803770576026,
      "loss": 1.3887,
      "step": 264
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.14915111660957336,
      "learning_rate": 0.0001964399707458649,
      "loss": 1.2351,
      "step": 265
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.11454611271619797,
      "learning_rate": 0.00019633026882599227,
      "loss": 1.3325,
      "step": 266
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.10960078984498978,
      "learning_rate": 0.00019621893380593396,
      "loss": 1.3,
      "step": 267
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.11405951529741287,
      "learning_rate": 0.0001961059675731678,
      "loss": 1.0641,
      "step": 268
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.10570972412824631,
      "learning_rate": 0.00019599137204282565,
      "loss": 1.1543,
      "step": 269
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.11385346204042435,
      "learning_rate": 0.00019587514915766124,
      "loss": 1.1546,
      "step": 270
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.10122133791446686,
      "learning_rate": 0.00019575730088801697,
      "loss": 1.1946,
      "step": 271
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.14278005063533783,
      "learning_rate": 0.00019563782923179063,
      "loss": 1.1853,
      "step": 272
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.11730815470218658,
      "learning_rate": 0.00019551673621440147,
      "loss": 1.4478,
      "step": 273
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.12385374307632446,
      "learning_rate": 0.00019539402388875598,
      "loss": 1.4392,
      "step": 274
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.10802686959505081,
      "learning_rate": 0.00019526969433521297,
      "loss": 1.3126,
      "step": 275
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.12256834656000137,
      "learning_rate": 0.00019514374966154825,
      "loss": 1.1855,
      "step": 276
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.10675772279500961,
      "learning_rate": 0.00019501619200291908,
      "loss": 1.2578,
      "step": 277
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.1070014238357544,
      "learning_rate": 0.00019488702352182786,
      "loss": 1.1426,
      "step": 278
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.11141553521156311,
      "learning_rate": 0.00019475624640808543,
      "loss": 1.1046,
      "step": 279
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.10741408169269562,
      "learning_rate": 0.00019462386287877407,
      "loss": 1.1991,
      "step": 280
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.11526710540056229,
      "learning_rate": 0.00019448987517820982,
      "loss": 1.3507,
      "step": 281
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.1025758609175682,
      "learning_rate": 0.0001943542855779044,
      "loss": 1.3,
      "step": 282
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.11231940984725952,
      "learning_rate": 0.0001942170963765268,
      "loss": 1.2408,
      "step": 283
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.1058710590004921,
      "learning_rate": 0.0001940783098998643,
      "loss": 1.1668,
      "step": 284
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.1024676039814949,
      "learning_rate": 0.00019393792850078292,
      "loss": 1.2192,
      "step": 285
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.10104867815971375,
      "learning_rate": 0.00019379595455918774,
      "loss": 1.2881,
      "step": 286
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.1211661621928215,
      "learning_rate": 0.00019365239048198225,
      "loss": 1.3614,
      "step": 287
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.1058136522769928,
      "learning_rate": 0.0001935072387030279,
      "loss": 1.2043,
      "step": 288
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.3414977490901947,
      "learning_rate": 0.0001933605016831026,
      "loss": 1.1638,
      "step": 289
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.09908904135227203,
      "learning_rate": 0.00019321218190985904,
      "loss": 1.0565,
      "step": 290
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.11039788275957108,
      "learning_rate": 0.00019306228189778254,
      "loss": 1.1324,
      "step": 291
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.1092277467250824,
      "learning_rate": 0.0001929108041881485,
      "loss": 1.0857,
      "step": 292
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.11304416507482529,
      "learning_rate": 0.00019275775134897918,
      "loss": 1.3365,
      "step": 293
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.1239667758345604,
      "learning_rate": 0.00019260312597500022,
      "loss": 1.2251,
      "step": 294
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.1069445088505745,
      "learning_rate": 0.00019244693068759666,
      "loss": 1.1466,
      "step": 295
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.11692642420530319,
      "learning_rate": 0.00019228916813476853,
      "loss": 0.947,
      "step": 296
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.11571913212537766,
      "learning_rate": 0.00019212984099108594,
      "loss": 1.1865,
      "step": 297
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.14679546654224396,
      "learning_rate": 0.00019196895195764362,
      "loss": 1.2747,
      "step": 298
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.11377564072608948,
      "learning_rate": 0.00019180650376201535,
      "loss": 1.2149,
      "step": 299
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.14246807992458344,
      "learning_rate": 0.0001916424991582075,
      "loss": 1.3234,
      "step": 300
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.11343812942504883,
      "learning_rate": 0.00019147694092661255,
      "loss": 1.2125,
      "step": 301
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.49394461512565613,
      "learning_rate": 0.00019130983187396172,
      "loss": 1.2545,
      "step": 302
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.12038710713386536,
      "learning_rate": 0.0001911411748332776,
      "loss": 1.2441,
      "step": 303
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.112415611743927,
      "learning_rate": 0.00019097097266382597,
      "loss": 1.1986,
      "step": 304
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.1328456848859787,
      "learning_rate": 0.0001907992282510675,
      "loss": 1.0427,
      "step": 305
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.12180749326944351,
      "learning_rate": 0.00019062594450660857,
      "loss": 1.2805,
      "step": 306
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.16928140819072723,
      "learning_rate": 0.00019045112436815215,
      "loss": 1.1842,
      "step": 307
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.15169909596443176,
      "learning_rate": 0.00019027477079944788,
      "loss": 1.1867,
      "step": 308
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.11884856224060059,
      "learning_rate": 0.0001900968867902419,
      "loss": 1.2883,
      "step": 309
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.10564276576042175,
      "learning_rate": 0.0001899174753562261,
      "loss": 1.264,
      "step": 310
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.12126502394676208,
      "learning_rate": 0.0001897365395389869,
      "loss": 1.2866,
      "step": 311
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.11135692894458771,
      "learning_rate": 0.00018955408240595395,
      "loss": 1.3224,
      "step": 312
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.11942031234502792,
      "learning_rate": 0.00018937010705034787,
      "loss": 1.2606,
      "step": 313
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.11544723808765411,
      "learning_rate": 0.00018918461659112802,
      "loss": 1.2025,
      "step": 314
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.12025148421525955,
      "learning_rate": 0.00018899761417293941,
      "loss": 1.4092,
      "step": 315
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.126939594745636,
      "learning_rate": 0.00018880910296605955,
      "loss": 1.1811,
      "step": 316
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.29961127042770386,
      "learning_rate": 0.00018861908616634465,
      "loss": 1.1998,
      "step": 317
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.12669076025485992,
      "learning_rate": 0.00018842756699517538,
      "loss": 1.1652,
      "step": 318
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.12721942365169525,
      "learning_rate": 0.00018823454869940242,
      "loss": 1.2145,
      "step": 319
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.1199335977435112,
      "learning_rate": 0.0001880400345512913,
      "loss": 1.2953,
      "step": 320
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.13686604797840118,
      "learning_rate": 0.00018784402784846683,
      "loss": 1.2455,
      "step": 321
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.1259993016719818,
      "learning_rate": 0.00018764653191385736,
      "loss": 0.9901,
      "step": 322
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.13265901803970337,
      "learning_rate": 0.00018744755009563848,
      "loss": 1.2294,
      "step": 323
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.11366987973451614,
      "learning_rate": 0.00018724708576717605,
      "loss": 1.3416,
      "step": 324
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.11126185208559036,
      "learning_rate": 0.0001870451423269692,
      "loss": 1.1254,
      "step": 325
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.11551456153392792,
      "learning_rate": 0.0001868417231985926,
      "loss": 1.2113,
      "step": 326
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.11155180633068085,
      "learning_rate": 0.00018663683183063845,
      "loss": 1.1587,
      "step": 327
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.1066342443227768,
      "learning_rate": 0.0001864304716966581,
      "loss": 1.3596,
      "step": 328
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.11637596786022186,
      "learning_rate": 0.00018622264629510299,
      "loss": 1.2117,
      "step": 329
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.10947410017251968,
      "learning_rate": 0.00018601335914926557,
      "loss": 0.9963,
      "step": 330
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.1826743334531784,
      "learning_rate": 0.0001858026138072193,
      "loss": 1.1548,
      "step": 331
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.12238079309463501,
      "learning_rate": 0.00018559041384175876,
      "loss": 1.0507,
      "step": 332
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.1196766272187233,
      "learning_rate": 0.00018537676285033887,
      "loss": 1.1443,
      "step": 333
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.1195516511797905,
      "learning_rate": 0.00018516166445501405,
      "loss": 1.083,
      "step": 334
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.111576147377491,
      "learning_rate": 0.0001849451223023767,
      "loss": 1.192,
      "step": 335
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.12735474109649658,
      "learning_rate": 0.0001847271400634955,
      "loss": 1.2692,
      "step": 336
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.10551782697439194,
      "learning_rate": 0.00018450772143385307,
      "loss": 1.2476,
      "step": 337
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.11405996978282928,
      "learning_rate": 0.00018428687013328336,
      "loss": 1.1888,
      "step": 338
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.13341975212097168,
      "learning_rate": 0.00018406458990590858,
      "loss": 1.2129,
      "step": 339
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.13987426459789276,
      "learning_rate": 0.00018384088452007578,
      "loss": 1.2114,
      "step": 340
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.1929798275232315,
      "learning_rate": 0.0001836157577682928,
      "loss": 1.1411,
      "step": 341
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.11936288326978683,
      "learning_rate": 0.00018338921346716425,
      "loss": 1.2696,
      "step": 342
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.10896290838718414,
      "learning_rate": 0.0001831612554573265,
      "loss": 1.2438,
      "step": 343
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.11335362493991852,
      "learning_rate": 0.00018293188760338286,
      "loss": 1.0288,
      "step": 344
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.11729809641838074,
      "learning_rate": 0.00018270111379383775,
      "loss": 1.1118,
      "step": 345
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.11622367799282074,
      "learning_rate": 0.0001824689379410311,
      "loss": 1.1515,
      "step": 346
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.11727786809206009,
      "learning_rate": 0.00018223536398107176,
      "loss": 1.2468,
      "step": 347
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.11702138185501099,
      "learning_rate": 0.00018200039587377097,
      "loss": 1.2647,
      "step": 348
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.12109760195016861,
      "learning_rate": 0.0001817640376025751,
      "loss": 1.2282,
      "step": 349
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.11537209153175354,
      "learning_rate": 0.00018152629317449813,
      "loss": 1.178,
      "step": 350
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.1228351965546608,
      "learning_rate": 0.00018128716662005383,
      "loss": 1.2099,
      "step": 351
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.15511314570903778,
      "learning_rate": 0.0001810466619931872,
      "loss": 1.2881,
      "step": 352
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.1076040267944336,
      "learning_rate": 0.00018080478337120605,
      "loss": 1.264,
      "step": 353
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.14106139540672302,
      "learning_rate": 0.00018056153485471166,
      "loss": 1.2821,
      "step": 354
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.11988777667284012,
      "learning_rate": 0.00018031692056752924,
      "loss": 1.2771,
      "step": 355
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.10703491419553757,
      "learning_rate": 0.0001800709446566382,
      "loss": 1.2525,
      "step": 356
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.11741764843463898,
      "learning_rate": 0.00017982361129210172,
      "loss": 1.2151,
      "step": 357
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.19867832958698273,
      "learning_rate": 0.00017957492466699605,
      "loss": 1.2559,
      "step": 358
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.12436340004205704,
      "learning_rate": 0.00017932488899733947,
      "loss": 1.3145,
      "step": 359
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.12323563545942307,
      "learning_rate": 0.00017907350852202077,
      "loss": 1.2145,
      "step": 360
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.126682311296463,
      "learning_rate": 0.00017882078750272742,
      "loss": 1.2805,
      "step": 361
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.10707955062389374,
      "learning_rate": 0.00017856673022387338,
      "loss": 1.1415,
      "step": 362
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.1281125843524933,
      "learning_rate": 0.0001783113409925263,
      "loss": 1.3117,
      "step": 363
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.11295267939567566,
      "learning_rate": 0.00017805462413833468,
      "loss": 1.1453,
      "step": 364
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.12421952188014984,
      "learning_rate": 0.00017779658401345438,
      "loss": 1.2293,
      "step": 365
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.11468549817800522,
      "learning_rate": 0.00017753722499247478,
      "loss": 1.1391,
      "step": 366
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.12964022159576416,
      "learning_rate": 0.00017727655147234476,
      "loss": 1.3975,
      "step": 367
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.11601359397172928,
      "learning_rate": 0.00017701456787229804,
      "loss": 1.2622,
      "step": 368
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.12835174798965454,
      "learning_rate": 0.00017675127863377838,
      "loss": 1.1703,
      "step": 369
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.11938933283090591,
      "learning_rate": 0.00017648668822036408,
      "loss": 1.2702,
      "step": 370
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.12092038989067078,
      "learning_rate": 0.00017622080111769257,
      "loss": 1.3653,
      "step": 371
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.1410731077194214,
      "learning_rate": 0.00017595362183338412,
      "loss": 1.2465,
      "step": 372
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.11769446730613708,
      "learning_rate": 0.00017568515489696558,
      "loss": 1.235,
      "step": 373
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.10854797065258026,
      "learning_rate": 0.00017541540485979359,
      "loss": 1.1758,
      "step": 374
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.10094323009252548,
      "learning_rate": 0.0001751443762949772,
      "loss": 1.0939,
      "step": 375
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.10836099833250046,
      "learning_rate": 0.00017487207379730078,
      "loss": 1.1618,
      "step": 376
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.1138240173459053,
      "learning_rate": 0.00017459850198314563,
      "loss": 1.1423,
      "step": 377
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.11237498372793198,
      "learning_rate": 0.000174323665490412,
      "loss": 1.0702,
      "step": 378
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.1078115925192833,
      "learning_rate": 0.00017404756897844052,
      "loss": 1.2523,
      "step": 379
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.11401957273483276,
      "learning_rate": 0.00017377021712793297,
      "loss": 1.2443,
      "step": 380
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.1109926849603653,
      "learning_rate": 0.00017349161464087312,
      "loss": 1.264,
      "step": 381
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.10931116342544556,
      "learning_rate": 0.00017321176624044687,
      "loss": 1.294,
      "step": 382
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.10655161738395691,
      "learning_rate": 0.00017293067667096239,
      "loss": 1.2629,
      "step": 383
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.11112368851900101,
      "learning_rate": 0.00017264835069776945,
      "loss": 1.25,
      "step": 384
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.1312362551689148,
      "learning_rate": 0.0001723647931071788,
      "loss": 1.1807,
      "step": 385
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.1174694374203682,
      "learning_rate": 0.00017208000870638093,
      "loss": 1.1719,
      "step": 386
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.11656120419502258,
      "learning_rate": 0.00017179400232336463,
      "loss": 1.4266,
      "step": 387
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.11490406095981598,
      "learning_rate": 0.00017150677880683514,
      "loss": 1.3041,
      "step": 388
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.1138778105378151,
      "learning_rate": 0.00017121834302613186,
      "loss": 1.3364,
      "step": 389
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.11219751089811325,
      "learning_rate": 0.000170928699871146,
      "loss": 1.3409,
      "step": 390
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.10369167476892471,
      "learning_rate": 0.0001706378542522374,
      "loss": 1.1775,
      "step": 391
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.11245618760585785,
      "learning_rate": 0.00017034581110015156,
      "loss": 1.1841,
      "step": 392
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.10895467549562454,
      "learning_rate": 0.00017005257536593577,
      "loss": 1.1855,
      "step": 393
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.10433463007211685,
      "learning_rate": 0.00016975815202085554,
      "loss": 1.2853,
      "step": 394
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.10486618429422379,
      "learning_rate": 0.00016946254605630992,
      "loss": 1.2029,
      "step": 395
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.12566353380680084,
      "learning_rate": 0.00016916576248374719,
      "loss": 1.187,
      "step": 396
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.11196866631507874,
      "learning_rate": 0.00016886780633457973,
      "loss": 1.1956,
      "step": 397
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.12952035665512085,
      "learning_rate": 0.00016856868266009874,
      "loss": 1.24,
      "step": 398
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.1032567098736763,
      "learning_rate": 0.0001682683965313887,
      "loss": 1.207,
      "step": 399
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.10673535615205765,
      "learning_rate": 0.00016796695303924127,
      "loss": 1.23,
      "step": 400
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.1103532612323761,
      "learning_rate": 0.00016766435729406913,
      "loss": 1.2994,
      "step": 401
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.11205995827913284,
      "learning_rate": 0.0001673606144258192,
      "loss": 1.1168,
      "step": 402
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.11950290203094482,
      "learning_rate": 0.00016705572958388576,
      "loss": 1.2152,
      "step": 403
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.17785818874835968,
      "learning_rate": 0.0001667497079370231,
      "loss": 1.2523,
      "step": 404
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.10789604485034943,
      "learning_rate": 0.00016644255467325794,
      "loss": 1.289,
      "step": 405
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.12129693478345871,
      "learning_rate": 0.00016613427499980143,
      "loss": 1.257,
      "step": 406
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.11083897203207016,
      "learning_rate": 0.00016582487414296097,
      "loss": 1.3349,
      "step": 407
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.09888826310634613,
      "learning_rate": 0.0001655143573480515,
      "loss": 1.1737,
      "step": 408
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.11880184710025787,
      "learning_rate": 0.00016520272987930652,
      "loss": 1.0338,
      "step": 409
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.11628952622413635,
      "learning_rate": 0.00016488999701978903,
      "loss": 1.2898,
      "step": 410
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.113735631108284,
      "learning_rate": 0.0001645761640713019,
      "loss": 1.1826,
      "step": 411
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.10678713023662567,
      "learning_rate": 0.00016426123635429786,
      "loss": 1.2231,
      "step": 412
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.11631446331739426,
      "learning_rate": 0.00016394521920778948,
      "loss": 1.0241,
      "step": 413
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.11213452368974686,
      "learning_rate": 0.00016362811798925853,
      "loss": 1.4935,
      "step": 414
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.1304715871810913,
      "learning_rate": 0.0001633099380745652,
      "loss": 1.1686,
      "step": 415
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.10429395735263824,
      "learning_rate": 0.000162990684857857,
      "loss": 1.0118,
      "step": 416
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.10091353952884674,
      "learning_rate": 0.00016267036375147725,
      "loss": 1.1401,
      "step": 417
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.10777878761291504,
      "learning_rate": 0.00016234898018587337,
      "loss": 1.33,
      "step": 418
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.11149658262729645,
      "learning_rate": 0.00016202653960950474,
      "loss": 1.2201,
      "step": 419
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.10549131035804749,
      "learning_rate": 0.0001617030474887505,
      "loss": 1.0758,
      "step": 420
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.12649282813072205,
      "learning_rate": 0.0001613785093078166,
      "loss": 1.2241,
      "step": 421
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.11720848083496094,
      "learning_rate": 0.00016105293056864315,
      "loss": 1.3001,
      "step": 422
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.12630730867385864,
      "learning_rate": 0.0001607263167908109,
      "loss": 0.986,
      "step": 423
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.11931440979242325,
      "learning_rate": 0.00016039867351144778,
      "loss": 1.2037,
      "step": 424
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.1198689192533493,
      "learning_rate": 0.000160070006285135,
      "loss": 1.2556,
      "step": 425
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.11019983142614365,
      "learning_rate": 0.0001597403206838128,
      "loss": 1.004,
      "step": 426
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.11280027031898499,
      "learning_rate": 0.00015940962229668625,
      "loss": 1.3164,
      "step": 427
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.12924431264400482,
      "learning_rate": 0.00015907791673013016,
      "loss": 1.1612,
      "step": 428
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.1206149235367775,
      "learning_rate": 0.00015874520960759423,
      "loss": 1.2171,
      "step": 429
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.12173935025930405,
      "learning_rate": 0.0001584115065695077,
      "loss": 1.2215,
      "step": 430
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.126984640955925,
      "learning_rate": 0.0001580768132731837,
      "loss": 1.2379,
      "step": 431
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.12142551690340042,
      "learning_rate": 0.0001577411353927233,
      "loss": 1.1171,
      "step": 432
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.11669143289327621,
      "learning_rate": 0.00015740447861891944,
      "loss": 1.1148,
      "step": 433
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.12235064059495926,
      "learning_rate": 0.00015706684865916026,
      "loss": 1.2705,
      "step": 434
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.11695750802755356,
      "learning_rate": 0.00015672825123733258,
      "loss": 1.1602,
      "step": 435
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.10883978754281998,
      "learning_rate": 0.00015638869209372457,
      "loss": 1.3629,
      "step": 436
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.09961367398500443,
      "learning_rate": 0.00015604817698492885,
      "loss": 1.1844,
      "step": 437
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.10849742591381073,
      "learning_rate": 0.00015570671168374438,
      "loss": 1.2453,
      "step": 438
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.13158155977725983,
      "learning_rate": 0.00015536430197907904,
      "loss": 1.0024,
      "step": 439
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.11819962412118912,
      "learning_rate": 0.00015502095367585122,
      "loss": 1.2432,
      "step": 440
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.10652721673250198,
      "learning_rate": 0.0001546766725948916,
      "loss": 1.2042,
      "step": 441
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.10861877351999283,
      "learning_rate": 0.0001543314645728442,
      "loss": 1.4347,
      "step": 442
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.11992961913347244,
      "learning_rate": 0.00015398533546206778,
      "loss": 1.2897,
      "step": 443
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.11120472103357315,
      "learning_rate": 0.00015363829113053634,
      "loss": 1.1729,
      "step": 444
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.11962945014238358,
      "learning_rate": 0.00015329033746173975,
      "loss": 1.0863,
      "step": 445
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.10950727015733719,
      "learning_rate": 0.00015294148035458404,
      "loss": 1.2127,
      "step": 446
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.11483786255121231,
      "learning_rate": 0.0001525917257232913,
      "loss": 1.1465,
      "step": 447
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.11853177845478058,
      "learning_rate": 0.0001522410794972995,
      "loss": 1.2527,
      "step": 448
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.10667990148067474,
      "learning_rate": 0.00015188954762116195,
      "loss": 1.2347,
      "step": 449
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.1277850717306137,
      "learning_rate": 0.0001515371360544465,
      "loss": 1.2875,
      "step": 450
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.10839621722698212,
      "learning_rate": 0.00015118385077163445,
      "loss": 1.3415,
      "step": 451
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.11405637860298157,
      "learning_rate": 0.00015082969776201947,
      "loss": 1.1112,
      "step": 452
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.1116739809513092,
      "learning_rate": 0.00015047468302960577,
      "loss": 1.2876,
      "step": 453
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.10504814237356186,
      "learning_rate": 0.00015011881259300653,
      "loss": 1.2027,
      "step": 454
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.10286784917116165,
      "learning_rate": 0.00014976209248534182,
      "loss": 1.1372,
      "step": 455
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.1141451969742775,
      "learning_rate": 0.00014940452875413627,
      "loss": 1.0566,
      "step": 456
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.12613342702388763,
      "learning_rate": 0.00014904612746121659,
      "loss": 0.9762,
      "step": 457
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.11521808058023453,
      "learning_rate": 0.00014868689468260874,
      "loss": 1.144,
      "step": 458
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.10090262442827225,
      "learning_rate": 0.0001483268365084351,
      "loss": 1.0197,
      "step": 459
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.11367885768413544,
      "learning_rate": 0.0001479659590428109,
      "loss": 1.3636,
      "step": 460
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.11813089996576309,
      "learning_rate": 0.0001476042684037411,
      "loss": 1.314,
      "step": 461
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.10709206759929657,
      "learning_rate": 0.0001472417707230164,
      "loss": 1.2899,
      "step": 462
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.11429055780172348,
      "learning_rate": 0.00014687847214610943,
      "loss": 1.2237,
      "step": 463
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.10949011892080307,
      "learning_rate": 0.00014651437883207054,
      "loss": 1.3596,
      "step": 464
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.1138731986284256,
      "learning_rate": 0.00014614949695342336,
      "loss": 1.0538,
      "step": 465
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.10718726366758347,
      "learning_rate": 0.00014578383269606005,
      "loss": 1.2639,
      "step": 466
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.10555477440357208,
      "learning_rate": 0.0001454173922591367,
      "loss": 1.2917,
      "step": 467
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.10175728797912598,
      "learning_rate": 0.00014505018185496802,
      "loss": 1.111,
      "step": 468
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.113253653049469,
      "learning_rate": 0.00014468220770892206,
      "loss": 1.0426,
      "step": 469
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.11308235675096512,
      "learning_rate": 0.0001443134760593147,
      "loss": 1.2584,
      "step": 470
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.10383269935846329,
      "learning_rate": 0.0001439439931573039,
      "loss": 1.0795,
      "step": 471
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.10998781770467758,
      "learning_rate": 0.00014357376526678367,
      "loss": 1.2224,
      "step": 472
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.12233158946037292,
      "learning_rate": 0.00014320279866427796,
      "loss": 1.3125,
      "step": 473
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.11626295000314713,
      "learning_rate": 0.00014283109963883419,
      "loss": 1.3173,
      "step": 474
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.12676700949668884,
      "learning_rate": 0.0001424586744919166,
      "loss": 1.3502,
      "step": 475
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.11780429631471634,
      "learning_rate": 0.00014208552953729947,
      "loss": 1.3161,
      "step": 476
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.11618795245885849,
      "learning_rate": 0.00014171167110096016,
      "loss": 1.2574,
      "step": 477
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.10930506885051727,
      "learning_rate": 0.00014133710552097174,
      "loss": 1.243,
      "step": 478
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.11596094816923141,
      "learning_rate": 0.00014096183914739553,
      "loss": 1.2908,
      "step": 479
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.13881249725818634,
      "learning_rate": 0.00014058587834217355,
      "loss": 1.1379,
      "step": 480
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.09807305783033371,
      "learning_rate": 0.00014020922947902066,
      "loss": 1.1799,
      "step": 481
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.10549982637166977,
      "learning_rate": 0.00013983189894331636,
      "loss": 1.2467,
      "step": 482
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.7992545366287231,
      "learning_rate": 0.00013945389313199669,
      "loss": 1.2711,
      "step": 483
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.18441331386566162,
      "learning_rate": 0.00013907521845344571,
      "loss": 1.0915,
      "step": 484
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.12779419124126434,
      "learning_rate": 0.00013869588132738695,
      "loss": 1.2633,
      "step": 485
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.13042038679122925,
      "learning_rate": 0.00013831588818477436,
      "loss": 1.2233,
      "step": 486
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.14113464951515198,
      "learning_rate": 0.00013793524546768356,
      "loss": 1.0653,
      "step": 487
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.12155605107545853,
      "learning_rate": 0.0001375539596292025,
      "loss": 1.2181,
      "step": 488
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.12651774287223816,
      "learning_rate": 0.00013717203713332188,
      "loss": 1.3338,
      "step": 489
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.2244076430797577,
      "learning_rate": 0.00013678948445482598,
      "loss": 1.3459,
      "step": 490
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.1257290095090866,
      "learning_rate": 0.00013640630807918247,
      "loss": 1.0061,
      "step": 491
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.11858661472797394,
      "learning_rate": 0.00013602251450243273,
      "loss": 1.1775,
      "step": 492
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.1279936581850052,
      "learning_rate": 0.00013563811023108158,
      "loss": 1.3702,
      "step": 493
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.13254624605178833,
      "learning_rate": 0.00013525310178198705,
      "loss": 1.3612,
      "step": 494
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.1243981122970581,
      "learning_rate": 0.00013486749568225001,
      "loss": 1.1652,
      "step": 495
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.1259343922138214,
      "learning_rate": 0.0001344812984691031,
      "loss": 1.2998,
      "step": 496
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.12421772629022598,
      "learning_rate": 0.00013409451668980048,
      "loss": 1.2688,
      "step": 497
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.11533673852682114,
      "learning_rate": 0.0001337071569015063,
      "loss": 1.2123,
      "step": 498
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.2135774940252304,
      "learning_rate": 0.00013331922567118393,
      "loss": 1.173,
      "step": 499
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.1305667608976364,
      "learning_rate": 0.00013293072957548443,
      "loss": 1.2795,
      "step": 500
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.11869475990533829,
      "learning_rate": 0.00013254167520063509,
      "loss": 1.1984,
      "step": 501
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.13584646582603455,
      "learning_rate": 0.0001321520691423278,
      "loss": 1.2018,
      "step": 502
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.12778714299201965,
      "learning_rate": 0.0001317619180056072,
      "loss": 1.2817,
      "step": 503
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.12326094508171082,
      "learning_rate": 0.0001313712284047587,
      "loss": 1.3434,
      "step": 504
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.11255846172571182,
      "learning_rate": 0.0001309800069631964,
      "loss": 1.1986,
      "step": 505
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.11870656907558441,
      "learning_rate": 0.0001305882603133508,
      "loss": 1.2409,
      "step": 506
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.11277451366186142,
      "learning_rate": 0.0001301959950965562,
      "loss": 1.4011,
      "step": 507
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.11900844424962997,
      "learning_rate": 0.00012980321796293836,
      "loss": 1.2379,
      "step": 508
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.14404135942459106,
      "learning_rate": 0.00012940993557130164,
      "loss": 1.3902,
      "step": 509
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.12229623645544052,
      "learning_rate": 0.000129016154589016,
      "loss": 1.2842,
      "step": 510
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.1197381243109703,
      "learning_rate": 0.00012862188169190418,
      "loss": 1.3836,
      "step": 511
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.12395986169576645,
      "learning_rate": 0.00012822712356412838,
      "loss": 1.3508,
      "step": 512
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.12418900430202484,
      "learning_rate": 0.00012783188689807697,
      "loss": 1.1969,
      "step": 513
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.1138986125588417,
      "learning_rate": 0.00012743617839425098,
      "loss": 1.3654,
      "step": 514
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.11219038814306259,
      "learning_rate": 0.0001270400047611508,
      "loss": 1.2621,
      "step": 515
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.11220891773700714,
      "learning_rate": 0.00012664337271516194,
      "loss": 1.2384,
      "step": 516
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.11203695833683014,
      "learning_rate": 0.0001262462889804416,
      "loss": 1.1551,
      "step": 517
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.10459273308515549,
      "learning_rate": 0.00012584876028880453,
      "loss": 1.2276,
      "step": 518
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.10843029618263245,
      "learning_rate": 0.00012545079337960882,
      "loss": 1.2414,
      "step": 519
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.10928460955619812,
      "learning_rate": 0.00012505239499964178,
      "loss": 1.0391,
      "step": 520
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.1170789897441864,
      "learning_rate": 0.0001246535719030055,
      "loss": 1.2558,
      "step": 521
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.118638776242733,
      "learning_rate": 0.00012425433085100224,
      "loss": 1.2906,
      "step": 522
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.12077479064464569,
      "learning_rate": 0.0001238546786120201,
      "loss": 1.3222,
      "step": 523
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.12042129039764404,
      "learning_rate": 0.00012345462196141797,
      "loss": 1.2399,
      "step": 524
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.11980915814638138,
      "learning_rate": 0.00012305416768141082,
      "loss": 1.2426,
      "step": 525
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.10789747536182404,
      "learning_rate": 0.00012265332256095464,
      "loss": 1.3437,
      "step": 526
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.10967132449150085,
      "learning_rate": 0.00012225209339563145,
      "loss": 1.1329,
      "step": 527
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.11903247982263565,
      "learning_rate": 0.00012185048698753403,
      "loss": 1.2107,
      "step": 528
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.10917498171329498,
      "learning_rate": 0.00012144851014515055,
      "loss": 1.1689,
      "step": 529
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.11308782547712326,
      "learning_rate": 0.00012104616968324927,
      "loss": 1.2487,
      "step": 530
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.10030585527420044,
      "learning_rate": 0.00012064347242276293,
      "loss": 1.1451,
      "step": 531
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.1195712685585022,
      "learning_rate": 0.00012024042519067309,
      "loss": 1.2525,
      "step": 532
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.12115343660116196,
      "learning_rate": 0.00011983703481989443,
      "loss": 1.2348,
      "step": 533
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.1060243546962738,
      "learning_rate": 0.00011943330814915896,
      "loss": 1.0223,
      "step": 534
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.2178022414445877,
      "learning_rate": 0.00011902925202289996,
      "loss": 1.1158,
      "step": 535
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.13022154569625854,
      "learning_rate": 0.00011862487329113606,
      "loss": 1.2742,
      "step": 536
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.11495082825422287,
      "learning_rate": 0.00011822017880935505,
      "loss": 1.2562,
      "step": 537
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.11310116201639175,
      "learning_rate": 0.00011781517543839768,
      "loss": 1.2719,
      "step": 538
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.13280071318149567,
      "learning_rate": 0.00011740987004434137,
      "loss": 1.1481,
      "step": 539
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.12752439081668854,
      "learning_rate": 0.00011700426949838363,
      "loss": 1.3024,
      "step": 540
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.11650800704956055,
      "learning_rate": 0.00011659838067672599,
      "loss": 1.2154,
      "step": 541
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.11279530823230743,
      "learning_rate": 0.00011619221046045687,
      "loss": 1.1627,
      "step": 542
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.12641020119190216,
      "learning_rate": 0.0001157857657354354,
      "loss": 1.0148,
      "step": 543
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.10685007274150848,
      "learning_rate": 0.00011537905339217448,
      "loss": 1.288,
      "step": 544
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.10929278284311295,
      "learning_rate": 0.00011497208032572384,
      "loss": 1.123,
      "step": 545
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.11334774643182755,
      "learning_rate": 0.00011456485343555344,
      "loss": 1.2894,
      "step": 546
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.10330837219953537,
      "learning_rate": 0.0001141573796254363,
      "loss": 1.1484,
      "step": 547
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.09870249778032303,
      "learning_rate": 0.00011374966580333147,
      "loss": 1.0223,
      "step": 548
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.12955228984355927,
      "learning_rate": 0.00011334171888126697,
      "loss": 1.2058,
      "step": 549
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.11018647998571396,
      "learning_rate": 0.00011293354577522263,
      "loss": 1.1589,
      "step": 550
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.12055356800556183,
      "learning_rate": 0.00011252515340501281,
      "loss": 1.091,
      "step": 551
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.10831710696220398,
      "learning_rate": 0.00011211654869416901,
      "loss": 1.2076,
      "step": 552
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.11912648379802704,
      "learning_rate": 0.00011170773856982268,
      "loss": 1.2953,
      "step": 553
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.11210989207029343,
      "learning_rate": 0.00011129872996258757,
      "loss": 1.0549,
      "step": 554
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.10992615669965744,
      "learning_rate": 0.0001108895298064424,
      "loss": 1.1818,
      "step": 555
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.1124982088804245,
      "learning_rate": 0.00011048014503861321,
      "loss": 1.3218,
      "step": 556
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.12512408196926117,
      "learning_rate": 0.00011007058259945584,
      "loss": 1.2262,
      "step": 557
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.1207667887210846,
      "learning_rate": 0.00010966084943233817,
      "loss": 1.1636,
      "step": 558
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.10861886292695999,
      "learning_rate": 0.00010925095248352239,
      "loss": 1.1363,
      "step": 559
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.11505457758903503,
      "learning_rate": 0.0001088408987020475,
      "loss": 1.1823,
      "step": 560
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.10509267449378967,
      "learning_rate": 0.00010843069503961111,
      "loss": 1.2269,
      "step": 561
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.10376826673746109,
      "learning_rate": 0.00010802034845045188,
      "loss": 1.0244,
      "step": 562
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.10977223515510559,
      "learning_rate": 0.00010760986589123146,
      "loss": 1.2713,
      "step": 563
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.11492714285850525,
      "learning_rate": 0.0001071992543209167,
      "loss": 1.2242,
      "step": 564
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.10544589161872864,
      "learning_rate": 0.00010678852070066159,
      "loss": 1.2959,
      "step": 565
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.11260811984539032,
      "learning_rate": 0.00010637767199368912,
      "loss": 1.3398,
      "step": 566
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.10535457730293274,
      "learning_rate": 0.00010596671516517356,
      "loss": 1.3178,
      "step": 567
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.10959797352552414,
      "learning_rate": 0.00010555565718212197,
      "loss": 1.0378,
      "step": 568
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.10014626383781433,
      "learning_rate": 0.00010514450501325645,
      "loss": 1.2384,
      "step": 569
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.1097472757101059,
      "learning_rate": 0.00010473326562889582,
      "loss": 1.1366,
      "step": 570
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.10998505353927612,
      "learning_rate": 0.0001043219460008374,
      "loss": 1.1083,
      "step": 571
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.11129113286733627,
      "learning_rate": 0.00010391055310223899,
      "loss": 1.1951,
      "step": 572
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.10348080098628998,
      "learning_rate": 0.00010349909390750046,
      "loss": 1.1871,
      "step": 573
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.10865020006895065,
      "learning_rate": 0.00010308757539214572,
      "loss": 1.2269,
      "step": 574
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.1128375232219696,
      "learning_rate": 0.00010267600453270421,
      "loss": 1.2358,
      "step": 575
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.13256721198558807,
      "learning_rate": 0.0001022643883065929,
      "loss": 1.3445,
      "step": 576
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.10738494992256165,
      "learning_rate": 0.00010185273369199781,
      "loss": 1.1042,
      "step": 577
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.10427848994731903,
      "learning_rate": 0.00010144104766775572,
      "loss": 1.0901,
      "step": 578
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.11618473380804062,
      "learning_rate": 0.000101029337213236,
      "loss": 0.781,
      "step": 579
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.20147624611854553,
      "learning_rate": 0.00010061760930822211,
      "loss": 1.2707,
      "step": 580
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.11436042189598083,
      "learning_rate": 0.0001002058709327934,
      "loss": 1.193,
      "step": 581
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.12396758794784546,
      "learning_rate": 9.979412906720663e-05,
      "loss": 1.0342,
      "step": 582
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.11723118275403976,
      "learning_rate": 9.938239069177791e-05,
      "loss": 1.1791,
      "step": 583
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.12710683047771454,
      "learning_rate": 9.897066278676404e-05,
      "loss": 1.023,
      "step": 584
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.10067608207464218,
      "learning_rate": 9.85589523322443e-05,
      "loss": 1.1753,
      "step": 585
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.11074398458003998,
      "learning_rate": 9.81472663080022e-05,
      "loss": 1.2567,
      "step": 586
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.16948086023330688,
      "learning_rate": 9.773561169340712e-05,
      "loss": 1.3789,
      "step": 587
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.10852295160293579,
      "learning_rate": 9.732399546729578e-05,
      "loss": 1.3642,
      "step": 588
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.10131103545427322,
      "learning_rate": 9.691242460785433e-05,
      "loss": 1.129,
      "step": 589
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.11037011444568634,
      "learning_rate": 9.650090609249956e-05,
      "loss": 1.1974,
      "step": 590
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.12829488515853882,
      "learning_rate": 9.608944689776103e-05,
      "loss": 1.2126,
      "step": 591
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.16550996899604797,
      "learning_rate": 9.567805399916259e-05,
      "loss": 1.228,
      "step": 592
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.1353956162929535,
      "learning_rate": 9.526673437110419e-05,
      "loss": 1.2877,
      "step": 593
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.17490524053573608,
      "learning_rate": 9.485549498674358e-05,
      "loss": 1.212,
      "step": 594
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.10544750839471817,
      "learning_rate": 9.444434281787807e-05,
      "loss": 1.3147,
      "step": 595
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.10780143737792969,
      "learning_rate": 9.403328483482646e-05,
      "loss": 1.1851,
      "step": 596
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.12775209546089172,
      "learning_rate": 9.362232800631088e-05,
      "loss": 1.1219,
      "step": 597
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.11476773023605347,
      "learning_rate": 9.321147929933847e-05,
      "loss": 1.2289,
      "step": 598
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.11137551069259644,
      "learning_rate": 9.28007456790833e-05,
      "loss": 1.2016,
      "step": 599
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.13261277973651886,
      "learning_rate": 9.239013410876857e-05,
      "loss": 1.389,
      "step": 600
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.10396253317594528,
      "learning_rate": 9.197965154954815e-05,
      "loss": 1.3316,
      "step": 601
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.11388034373521805,
      "learning_rate": 9.15693049603889e-05,
      "loss": 1.2501,
      "step": 602
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.1430211067199707,
      "learning_rate": 9.115910129795251e-05,
      "loss": 1.2259,
      "step": 603
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.12369144707918167,
      "learning_rate": 9.074904751647762e-05,
      "loss": 1.138,
      "step": 604
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.11555325239896774,
      "learning_rate": 9.033915056766187e-05,
      "loss": 1.1845,
      "step": 605
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.15159161388874054,
      "learning_rate": 8.992941740054418e-05,
      "loss": 1.3823,
      "step": 606
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.12028995901346207,
      "learning_rate": 8.951985496138678e-05,
      "loss": 1.0985,
      "step": 607
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.12002217024564743,
      "learning_rate": 8.911047019355764e-05,
      "loss": 1.2431,
      "step": 608
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.11859162151813507,
      "learning_rate": 8.870127003741244e-05,
      "loss": 1.1669,
      "step": 609
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.10495835542678833,
      "learning_rate": 8.829226143017735e-05,
      "loss": 1.1424,
      "step": 610
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.11981993168592453,
      "learning_rate": 8.788345130583099e-05,
      "loss": 1.0444,
      "step": 611
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.11924871802330017,
      "learning_rate": 8.747484659498724e-05,
      "loss": 1.1574,
      "step": 612
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.15531030297279358,
      "learning_rate": 8.706645422477739e-05,
      "loss": 1.1786,
      "step": 613
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.11026681959629059,
      "learning_rate": 8.665828111873307e-05,
      "loss": 1.2112,
      "step": 614
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.12397878617048264,
      "learning_rate": 8.625033419666857e-05,
      "loss": 1.2931,
      "step": 615
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.10594267398118973,
      "learning_rate": 8.584262037456373e-05,
      "loss": 1.1422,
      "step": 616
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.11834708601236343,
      "learning_rate": 8.54351465644466e-05,
      "loss": 1.16,
      "step": 617
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.12605226039886475,
      "learning_rate": 8.502791967427618e-05,
      "loss": 1.3575,
      "step": 618
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.12767426669597626,
      "learning_rate": 8.462094660782556e-05,
      "loss": 1.3343,
      "step": 619
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.12377568334341049,
      "learning_rate": 8.42142342645646e-05,
      "loss": 1.2027,
      "step": 620
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.11611881852149963,
      "learning_rate": 8.380778953954313e-05,
      "loss": 1.1748,
      "step": 621
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.10847889631986618,
      "learning_rate": 8.340161932327404e-05,
      "loss": 1.138,
      "step": 622
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.11104707419872284,
      "learning_rate": 8.299573050161638e-05,
      "loss": 1.2403,
      "step": 623
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.10397801548242569,
      "learning_rate": 8.259012995565867e-05,
      "loss": 1.2159,
      "step": 624
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.10560380667448044,
      "learning_rate": 8.21848245616023e-05,
      "loss": 1.0932,
      "step": 625
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.10818043351173401,
      "learning_rate": 8.177982119064496e-05,
      "loss": 1.2667,
      "step": 626
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.1053985133767128,
      "learning_rate": 8.137512670886397e-05,
      "loss": 1.3267,
      "step": 627
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.10604721307754517,
      "learning_rate": 8.097074797710006e-05,
      "loss": 1.2254,
      "step": 628
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.1083521693944931,
      "learning_rate": 8.056669185084108e-05,
      "loss": 1.389,
      "step": 629
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.12045227736234665,
      "learning_rate": 8.016296518010558e-05,
      "loss": 1.245,
      "step": 630
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.11613083630800247,
      "learning_rate": 7.975957480932695e-05,
      "loss": 1.2618,
      "step": 631
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.11401872336864471,
      "learning_rate": 7.935652757723708e-05,
      "loss": 1.2073,
      "step": 632
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.1841176450252533,
      "learning_rate": 7.895383031675074e-05,
      "loss": 1.2132,
      "step": 633
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.10668601095676422,
      "learning_rate": 7.855148985484946e-05,
      "loss": 1.2177,
      "step": 634
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.10970567911863327,
      "learning_rate": 7.814951301246598e-05,
      "loss": 1.2306,
      "step": 635
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.10744346678256989,
      "learning_rate": 7.774790660436858e-05,
      "loss": 1.2155,
      "step": 636
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.09847363084554672,
      "learning_rate": 7.734667743904539e-05,
      "loss": 1.1383,
      "step": 637
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.12585873901844025,
      "learning_rate": 7.69458323185892e-05,
      "loss": 1.2411,
      "step": 638
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.13609972596168518,
      "learning_rate": 7.654537803858204e-05,
      "loss": 1.1972,
      "step": 639
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.10885506868362427,
      "learning_rate": 7.614532138797993e-05,
      "loss": 1.2197,
      "step": 640
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.10706409066915512,
      "learning_rate": 7.57456691489978e-05,
      "loss": 1.0704,
      "step": 641
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.11232727020978928,
      "learning_rate": 7.534642809699455e-05,
      "loss": 1.2443,
      "step": 642
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.1036246195435524,
      "learning_rate": 7.494760500035823e-05,
      "loss": 1.1623,
      "step": 643
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.11857923120260239,
      "learning_rate": 7.454920662039117e-05,
      "loss": 1.3161,
      "step": 644
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.1063128188252449,
      "learning_rate": 7.415123971119549e-05,
      "loss": 1.37,
      "step": 645
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.11207219213247299,
      "learning_rate": 7.375371101955841e-05,
      "loss": 1.1433,
      "step": 646
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.1198432594537735,
      "learning_rate": 7.335662728483808e-05,
      "loss": 1.2341,
      "step": 647
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.14110444486141205,
      "learning_rate": 7.295999523884921e-05,
      "loss": 1.2201,
      "step": 648
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.10053621232509613,
      "learning_rate": 7.256382160574901e-05,
      "loss": 1.1276,
      "step": 649
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.1087542399764061,
      "learning_rate": 7.216811310192308e-05,
      "loss": 1.3124,
      "step": 650
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.10470487177371979,
      "learning_rate": 7.177287643587163e-05,
      "loss": 1.0677,
      "step": 651
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.12348546087741852,
      "learning_rate": 7.137811830809583e-05,
      "loss": 1.0808,
      "step": 652
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.12709330022335052,
      "learning_rate": 7.0983845410984e-05,
      "loss": 1.1746,
      "step": 653
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.0996909812092781,
      "learning_rate": 7.059006442869839e-05,
      "loss": 1.2121,
      "step": 654
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.12972918152809143,
      "learning_rate": 7.019678203706163e-05,
      "loss": 1.149,
      "step": 655
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.10252442955970764,
      "learning_rate": 6.980400490344383e-05,
      "loss": 1.2643,
      "step": 656
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.11937206983566284,
      "learning_rate": 6.941173968664923e-05,
      "loss": 1.2484,
      "step": 657
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.11440838873386383,
      "learning_rate": 6.901999303680358e-05,
      "loss": 1.3256,
      "step": 658
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.10103975981473923,
      "learning_rate": 6.862877159524132e-05,
      "loss": 1.3196,
      "step": 659
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.10081090033054352,
      "learning_rate": 6.823808199439284e-05,
      "loss": 1.1371,
      "step": 660
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.16644109785556793,
      "learning_rate": 6.784793085767221e-05,
      "loss": 1.1239,
      "step": 661
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.1086033284664154,
      "learning_rate": 6.745832479936491e-05,
      "loss": 1.3244,
      "step": 662
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.11194080859422684,
      "learning_rate": 6.70692704245156e-05,
      "loss": 1.3153,
      "step": 663
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.1124996691942215,
      "learning_rate": 6.668077432881609e-05,
      "loss": 1.2942,
      "step": 664
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.3689382076263428,
      "learning_rate": 6.629284309849372e-05,
      "loss": 1.2606,
      "step": 665
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.11704660952091217,
      "learning_rate": 6.590548331019957e-05,
      "loss": 1.3214,
      "step": 666
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.11636264622211456,
      "learning_rate": 6.55187015308969e-05,
      "loss": 1.3366,
      "step": 667
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.10623035579919815,
      "learning_rate": 6.513250431775004e-05,
      "loss": 1.1485,
      "step": 668
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.10182057321071625,
      "learning_rate": 6.474689821801295e-05,
      "loss": 1.1521,
      "step": 669
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.11016492545604706,
      "learning_rate": 6.436188976891846e-05,
      "loss": 1.2067,
      "step": 670
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.11019518971443176,
      "learning_rate": 6.39774854975673e-05,
      "loss": 1.2559,
      "step": 671
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.11215123534202576,
      "learning_rate": 6.359369192081756e-05,
      "loss": 1.2557,
      "step": 672
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.1162608414888382,
      "learning_rate": 6.321051554517407e-05,
      "loss": 1.1967,
      "step": 673
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.1348588615655899,
      "learning_rate": 6.282796286667814e-05,
      "loss": 1.1944,
      "step": 674
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.12534910440444946,
      "learning_rate": 6.244604037079753e-05,
      "loss": 1.1354,
      "step": 675
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.11349747329950333,
      "learning_rate": 6.206475453231644e-05,
      "loss": 1.3001,
      "step": 676
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.11473750323057175,
      "learning_rate": 6.168411181522569e-05,
      "loss": 1.3084,
      "step": 677
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.11263560503721237,
      "learning_rate": 6.13041186726131e-05,
      "loss": 1.0361,
      "step": 678
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.11515972018241882,
      "learning_rate": 6.092478154655431e-05,
      "loss": 1.1342,
      "step": 679
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.10681987553834915,
      "learning_rate": 6.0546106868003324e-05,
      "loss": 1.2221,
      "step": 680
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.10119231045246124,
      "learning_rate": 6.016810105668365e-05,
      "loss": 1.1237,
      "step": 681
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.10480265319347382,
      "learning_rate": 5.979077052097936e-05,
      "loss": 1.2506,
      "step": 682
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.11073455959558487,
      "learning_rate": 5.941412165782645e-05,
      "loss": 1.0603,
      "step": 683
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.11075324565172195,
      "learning_rate": 5.9038160852604476e-05,
      "loss": 1.1351,
      "step": 684
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.10219030827283859,
      "learning_rate": 5.866289447902828e-05,
      "loss": 1.2163,
      "step": 685
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.10446413606405258,
      "learning_rate": 5.8288328899039834e-05,
      "loss": 1.2578,
      "step": 686
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.10746556520462036,
      "learning_rate": 5.791447046270056e-05,
      "loss": 1.2526,
      "step": 687
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.12443575263023376,
      "learning_rate": 5.754132550808344e-05,
      "loss": 1.3376,
      "step": 688
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.12434745579957962,
      "learning_rate": 5.7168900361165824e-05,
      "loss": 1.369,
      "step": 689
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.11743592470884323,
      "learning_rate": 5.679720133572206e-05,
      "loss": 1.2044,
      "step": 690
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.12590321898460388,
      "learning_rate": 5.6426234733216376e-05,
      "loss": 1.0729,
      "step": 691
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.10635906457901001,
      "learning_rate": 5.605600684269614e-05,
      "loss": 1.2448,
      "step": 692
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.10504303872585297,
      "learning_rate": 5.568652394068532e-05,
      "loss": 1.0663,
      "step": 693
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.10200861841440201,
      "learning_rate": 5.5317792291077965e-05,
      "loss": 1.0939,
      "step": 694
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.11620642244815826,
      "learning_rate": 5.4949818145031986e-05,
      "loss": 1.1002,
      "step": 695
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.10879629105329514,
      "learning_rate": 5.4582607740863324e-05,
      "loss": 1.1695,
      "step": 696
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.11083561182022095,
      "learning_rate": 5.421616730394e-05,
      "loss": 1.2967,
      "step": 697
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.11363320797681808,
      "learning_rate": 5.385050304657669e-05,
      "loss": 1.0099,
      "step": 698
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.10837159305810928,
      "learning_rate": 5.348562116792946e-05,
      "loss": 1.1575,
      "step": 699
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.7101715207099915,
      "learning_rate": 5.312152785389055e-05,
      "loss": 1.3229,
      "step": 700
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.11040350794792175,
      "learning_rate": 5.2758229276983614e-05,
      "loss": 1.183,
      "step": 701
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.10885446518659592,
      "learning_rate": 5.2395731596258925e-05,
      "loss": 1.139,
      "step": 702
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.10934137552976608,
      "learning_rate": 5.20340409571891e-05,
      "loss": 1.291,
      "step": 703
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.11229011416435242,
      "learning_rate": 5.167316349156495e-05,
      "loss": 1.1406,
      "step": 704
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.12293226271867752,
      "learning_rate": 5.1313105317391284e-05,
      "loss": 1.1102,
      "step": 705
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.10751207172870636,
      "learning_rate": 5.095387253878345e-05,
      "loss": 1.1296,
      "step": 706
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.10970565676689148,
      "learning_rate": 5.059547124586375e-05,
      "loss": 1.3482,
      "step": 707
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.10699622333049774,
      "learning_rate": 5.0237907514658176e-05,
      "loss": 1.3642,
      "step": 708
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.1132645532488823,
      "learning_rate": 4.988118740699346e-05,
      "loss": 1.1504,
      "step": 709
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.12392119318246841,
      "learning_rate": 4.9525316970394245e-05,
      "loss": 1.3906,
      "step": 710
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.10416343808174133,
      "learning_rate": 4.917030223798057e-05,
      "loss": 1.3419,
      "step": 711
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.11010722070932388,
      "learning_rate": 4.881614922836555e-05,
      "loss": 1.1175,
      "step": 712
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.11078814417123795,
      "learning_rate": 4.8462863945553516e-05,
      "loss": 1.3513,
      "step": 713
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.12069610506296158,
      "learning_rate": 4.8110452378838035e-05,
      "loss": 1.3913,
      "step": 714
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.11017792671918869,
      "learning_rate": 4.775892050270051e-05,
      "loss": 1.2217,
      "step": 715
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.10235986858606339,
      "learning_rate": 4.740827427670871e-05,
      "loss": 1.1375,
      "step": 716
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.11740509420633316,
      "learning_rate": 4.7058519645416e-05,
      "loss": 1.1649,
      "step": 717
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.16506601870059967,
      "learning_rate": 4.6709662538260267e-05,
      "loss": 1.0392,
      "step": 718
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.22638647258281708,
      "learning_rate": 4.63617088694637e-05,
      "loss": 1.21,
      "step": 719
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.11359802633523941,
      "learning_rate": 4.601466453793224e-05,
      "loss": 1.2481,
      "step": 720
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.10808828473091125,
      "learning_rate": 4.566853542715581e-05,
      "loss": 1.2008,
      "step": 721
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.09867287427186966,
      "learning_rate": 4.5323327405108416e-05,
      "loss": 1.0815,
      "step": 722
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.11430750787258148,
      "learning_rate": 4.4979046324148785e-05,
      "loss": 1.3082,
      "step": 723
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.1082565113902092,
      "learning_rate": 4.4635698020921014e-05,
      "loss": 1.0386,
      "step": 724
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.11467121541500092,
      "learning_rate": 4.4293288316255653e-05,
      "loss": 1.299,
      "step": 725
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.10791204124689102,
      "learning_rate": 4.395182301507118e-05,
      "loss": 1.2306,
      "step": 726
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.11149678379297256,
      "learning_rate": 4.3611307906275414e-05,
      "loss": 1.3074,
      "step": 727
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.11212314665317535,
      "learning_rate": 4.327174876266743e-05,
      "loss": 1.1512,
      "step": 728
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.1070462167263031,
      "learning_rate": 4.293315134083975e-05,
      "loss": 1.2378,
      "step": 729
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.11427627503871918,
      "learning_rate": 4.2595521381080604e-05,
      "loss": 1.2729,
      "step": 730
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.1135973185300827,
      "learning_rate": 4.2258864607276705e-05,
      "loss": 1.1399,
      "step": 731
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.11445244401693344,
      "learning_rate": 4.192318672681631e-05,
      "loss": 1.1549,
      "step": 732
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.10163676738739014,
      "learning_rate": 4.158849343049232e-05,
      "loss": 1.2751,
      "step": 733
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.11175056546926498,
      "learning_rate": 4.1254790392405797e-05,
      "loss": 1.2371,
      "step": 734
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.11040692776441574,
      "learning_rate": 4.092208326986986e-05,
      "loss": 1.3218,
      "step": 735
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.10620775073766708,
      "learning_rate": 4.0590377703313785e-05,
      "loss": 1.1667,
      "step": 736
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.10938204824924469,
      "learning_rate": 4.0259679316187214e-05,
      "loss": 1.0897,
      "step": 737
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.10027039796113968,
      "learning_rate": 3.992999371486508e-05,
      "loss": 1.1646,
      "step": 738
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.13585035502910614,
      "learning_rate": 3.960132648855226e-05,
      "loss": 1.1759,
      "step": 739
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.12253659218549728,
      "learning_rate": 3.9273683209189115e-05,
      "loss": 1.3585,
      "step": 740
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.14041317999362946,
      "learning_rate": 3.8947069431356855e-05,
      "loss": 1.2736,
      "step": 741
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.10640061646699905,
      "learning_rate": 3.862149069218343e-05,
      "loss": 1.2859,
      "step": 742
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.10962343961000443,
      "learning_rate": 3.829695251124953e-05,
      "loss": 1.2581,
      "step": 743
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.12311571091413498,
      "learning_rate": 3.797346039049528e-05,
      "loss": 1.3115,
      "step": 744
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.12078052759170532,
      "learning_rate": 3.7651019814126654e-05,
      "loss": 1.03,
      "step": 745
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.10591412335634232,
      "learning_rate": 3.732963624852275e-05,
      "loss": 1.0906,
      "step": 746
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.1085178554058075,
      "learning_rate": 3.700931514214303e-05,
      "loss": 1.1478,
      "step": 747
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.10265811532735825,
      "learning_rate": 3.6690061925434814e-05,
      "loss": 1.1375,
      "step": 748
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.10623796284198761,
      "learning_rate": 3.637188201074149e-05,
      "loss": 1.2655,
      "step": 749
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.1077788770198822,
      "learning_rate": 3.605478079221054e-05,
      "loss": 1.2805,
      "step": 750
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.1078350618481636,
      "learning_rate": 3.573876364570214e-05,
      "loss": 1.356,
      "step": 751
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.10767360776662827,
      "learning_rate": 3.5423835928698125e-05,
      "loss": 1.2346,
      "step": 752
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.10546786338090897,
      "learning_rate": 3.5110002980210975e-05,
      "loss": 1.2541,
      "step": 753
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.09910236299037933,
      "learning_rate": 3.479727012069349e-05,
      "loss": 1.2015,
      "step": 754
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.10072822123765945,
      "learning_rate": 3.4485642651948516e-05,
      "loss": 1.3005,
      "step": 755
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.11063572019338608,
      "learning_rate": 3.417512585703903e-05,
      "loss": 1.0739,
      "step": 756
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.09937085956335068,
      "learning_rate": 3.38657250001986e-05,
      "loss": 1.3106,
      "step": 757
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.10743856430053711,
      "learning_rate": 3.355744532674211e-05,
      "loss": 1.1492,
      "step": 758
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.11000777781009674,
      "learning_rate": 3.325029206297694e-05,
      "loss": 1.1876,
      "step": 759
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.11820725351572037,
      "learning_rate": 3.294427041611425e-05,
      "loss": 1.2457,
      "step": 760
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.10308333486318588,
      "learning_rate": 3.263938557418082e-05,
      "loss": 1.2801,
      "step": 761
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.1042330265045166,
      "learning_rate": 3.233564270593088e-05,
      "loss": 1.2237,
      "step": 762
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.10803329199552536,
      "learning_rate": 3.2033046960758764e-05,
      "loss": 1.2841,
      "step": 763
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.11426287144422531,
      "learning_rate": 3.1731603468611337e-05,
      "loss": 1.355,
      "step": 764
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.10273044556379318,
      "learning_rate": 3.143131733990127e-05,
      "loss": 1.2975,
      "step": 765
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.11743076890707016,
      "learning_rate": 3.113219366542031e-05,
      "loss": 1.2532,
      "step": 766
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.1157737672328949,
      "learning_rate": 3.083423751625282e-05,
      "loss": 1.4476,
      "step": 767
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.10297975689172745,
      "learning_rate": 3.0537453943690074e-05,
      "loss": 1.1543,
      "step": 768
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.10176219791173935,
      "learning_rate": 3.0241847979144487e-05,
      "loss": 1.0901,
      "step": 769
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.11567193269729614,
      "learning_rate": 2.994742463406427e-05,
      "loss": 1.25,
      "step": 770
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.12137153744697571,
      "learning_rate": 2.96541888998485e-05,
      "loss": 1.3956,
      "step": 771
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.10772846639156342,
      "learning_rate": 2.9362145747762625e-05,
      "loss": 1.4177,
      "step": 772
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.10832470655441284,
      "learning_rate": 2.9071300128854007e-05,
      "loss": 1.1614,
      "step": 773
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.09898839145898819,
      "learning_rate": 2.878165697386812e-05,
      "loss": 1.2994,
      "step": 774
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.1113181784749031,
      "learning_rate": 2.849322119316489e-05,
      "loss": 1.1914,
      "step": 775
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.10062463581562042,
      "learning_rate": 2.82059976766354e-05,
      "loss": 1.1165,
      "step": 776
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.10819394886493683,
      "learning_rate": 2.7919991293619098e-05,
      "loss": 1.283,
      "step": 777
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.1365766078233719,
      "learning_rate": 2.763520689282122e-05,
      "loss": 1.1963,
      "step": 778
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.11200159043073654,
      "learning_rate": 2.7351649302230553e-05,
      "loss": 1.1885,
      "step": 779
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.11218269914388657,
      "learning_rate": 2.7069323329037634e-05,
      "loss": 1.2944,
      "step": 780
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.12175281345844269,
      "learning_rate": 2.678823375955314e-05,
      "loss": 1.2379,
      "step": 781
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.125010147690773,
      "learning_rate": 2.650838535912692e-05,
      "loss": 1.2309,
      "step": 782
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.12338108569383621,
      "learning_rate": 2.6229782872067045e-05,
      "loss": 1.1489,
      "step": 783
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.11083382368087769,
      "learning_rate": 2.595243102155951e-05,
      "loss": 1.1595,
      "step": 784
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.11741047352552414,
      "learning_rate": 2.5676334509588008e-05,
      "loss": 1.2438,
      "step": 785
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.10495740175247192,
      "learning_rate": 2.540149801685441e-05,
      "loss": 1.299,
      "step": 786
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.11125294119119644,
      "learning_rate": 2.5127926202699238e-05,
      "loss": 1.1244,
      "step": 787
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.10036681592464447,
      "learning_rate": 2.485562370502279e-05,
      "loss": 1.2892,
      "step": 788
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.11835762113332748,
      "learning_rate": 2.4584595140206458e-05,
      "loss": 1.1257,
      "step": 789
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.11104938387870789,
      "learning_rate": 2.4314845103034456e-05,
      "loss": 1.1601,
      "step": 790
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.10659179091453552,
      "learning_rate": 2.40463781666159e-05,
      "loss": 1.2066,
      "step": 791
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.10941512882709503,
      "learning_rate": 2.377919888230744e-05,
      "loss": 1.3887,
      "step": 792
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.18816524744033813,
      "learning_rate": 2.3513311779635905e-05,
      "loss": 1.2191,
      "step": 793
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.1119375005364418,
      "learning_rate": 2.3248721366221638e-05,
      "loss": 1.2773,
      "step": 794
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.10966973006725311,
      "learning_rate": 2.2985432127701946e-05,
      "loss": 1.2283,
      "step": 795
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.10751962661743164,
      "learning_rate": 2.2723448527655266e-05,
      "loss": 1.375,
      "step": 796
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.19195051491260529,
      "learning_rate": 2.246277500752524e-05,
      "loss": 1.3717,
      "step": 797
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4861750900745392,
      "learning_rate": 2.220341598654565e-05,
      "loss": 1.1786,
      "step": 798
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.10594101250171661,
      "learning_rate": 2.194537586166532e-05,
      "loss": 1.2528,
      "step": 799
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.11542479693889618,
      "learning_rate": 2.16886590074737e-05,
      "loss": 1.2277,
      "step": 800
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.10653986781835556,
      "learning_rate": 2.143326977612662e-05,
      "loss": 1.1785,
      "step": 801
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.11760350316762924,
      "learning_rate": 2.1179212497272584e-05,
      "loss": 1.2573,
      "step": 802
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.1091405376791954,
      "learning_rate": 2.092649147797927e-05,
      "loss": 1.0923,
      "step": 803
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.10867162048816681,
      "learning_rate": 2.0675111002660562e-05,
      "loss": 1.2358,
      "step": 804
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.10017253458499908,
      "learning_rate": 2.0425075333003953e-05,
      "loss": 1.2741,
      "step": 805
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.11259835958480835,
      "learning_rate": 2.0176388707898274e-05,
      "loss": 1.0486,
      "step": 806
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.1280423402786255,
      "learning_rate": 1.992905534336178e-05,
      "loss": 1.3091,
      "step": 807
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.10603731125593185,
      "learning_rate": 1.9683079432470775e-05,
      "loss": 1.2541,
      "step": 808
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.10933735966682434,
      "learning_rate": 1.9438465145288374e-05,
      "loss": 1.2186,
      "step": 809
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.11166153848171234,
      "learning_rate": 1.9195216628793956e-05,
      "loss": 1.2563,
      "step": 810
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.11184462904930115,
      "learning_rate": 1.8953338006812805e-05,
      "loss": 1.3052,
      "step": 811
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.11681969463825226,
      "learning_rate": 1.8712833379946215e-05,
      "loss": 1.1702,
      "step": 812
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.11982112377882004,
      "learning_rate": 1.847370682550187e-05,
      "loss": 1.3025,
      "step": 813
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.11721136420965195,
      "learning_rate": 1.82359623974249e-05,
      "loss": 1.1182,
      "step": 814
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.11441758275032043,
      "learning_rate": 1.799960412622904e-05,
      "loss": 1.1366,
      "step": 815
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.11492711305618286,
      "learning_rate": 1.776463601892825e-05,
      "loss": 1.1784,
      "step": 816
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.11135811358690262,
      "learning_rate": 1.753106205896895e-05,
      "loss": 1.2411,
      "step": 817
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.0985221266746521,
      "learning_rate": 1.729888620616228e-05,
      "loss": 1.283,
      "step": 818
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.10769841074943542,
      "learning_rate": 1.7068112396617163e-05,
      "loss": 1.2442,
      "step": 819
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.09943833947181702,
      "learning_rate": 1.683874454267349e-05,
      "loss": 1.1121,
      "step": 820
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.11267413944005966,
      "learning_rate": 1.6610786532835774e-05,
      "loss": 1.2912,
      "step": 821
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.10905832797288895,
      "learning_rate": 1.6384242231707204e-05,
      "loss": 1.1264,
      "step": 822
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.11152827739715576,
      "learning_rate": 1.6159115479924257e-05,
      "loss": 1.0625,
      "step": 823
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.10507339239120483,
      "learning_rate": 1.593541009409143e-05,
      "loss": 1.282,
      "step": 824
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.11833300441503525,
      "learning_rate": 1.5713129866716648e-05,
      "loss": 1.0683,
      "step": 825
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.10734719038009644,
      "learning_rate": 1.5492278566146946e-05,
      "loss": 1.2312,
      "step": 826
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.09920140355825424,
      "learning_rate": 1.5272859936504514e-05,
      "loss": 1.3253,
      "step": 827
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.10309410840272903,
      "learning_rate": 1.5054877697623304e-05,
      "loss": 1.3389,
      "step": 828
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.09966228157281876,
      "learning_rate": 1.4838335544985981e-05,
      "loss": 1.1592,
      "step": 829
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.10500723123550415,
      "learning_rate": 1.462323714966114e-05,
      "loss": 1.2318,
      "step": 830
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.10912671685218811,
      "learning_rate": 1.4409586158241272e-05,
      "loss": 1.2195,
      "step": 831
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.1132998913526535,
      "learning_rate": 1.4197386192780715e-05,
      "loss": 1.2476,
      "step": 832
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.11414579302072525,
      "learning_rate": 1.3986640850734444e-05,
      "loss": 1.229,
      "step": 833
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.10963235795497894,
      "learning_rate": 1.3777353704897e-05,
      "loss": 1.1468,
      "step": 834
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.11412128806114197,
      "learning_rate": 1.3569528303341927e-05,
      "loss": 1.141,
      "step": 835
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.11492784321308136,
      "learning_rate": 1.3363168169361573e-05,
      "loss": 1.0967,
      "step": 836
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.11600396037101746,
      "learning_rate": 1.3158276801407431e-05,
      "loss": 1.2414,
      "step": 837
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.11931241303682327,
      "learning_rate": 1.2954857673030807e-05,
      "loss": 1.2318,
      "step": 838
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.10490193963050842,
      "learning_rate": 1.2752914232823942e-05,
      "loss": 1.2481,
      "step": 839
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.11982220411300659,
      "learning_rate": 1.2552449904361529e-05,
      "loss": 1.0291,
      "step": 840
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.0949595496058464,
      "learning_rate": 1.235346808614264e-05,
      "loss": 1.0835,
      "step": 841
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.11206669360399246,
      "learning_rate": 1.2155972151533224e-05,
      "loss": 1.2174,
      "step": 842
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.10925093293190002,
      "learning_rate": 1.1959965448708733e-05,
      "loss": 1.359,
      "step": 843
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.10597091913223267,
      "learning_rate": 1.1765451300597573e-05,
      "loss": 1.2567,
      "step": 844
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.12670493125915527,
      "learning_rate": 1.1572433004824635e-05,
      "loss": 1.1798,
      "step": 845
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.10978806763887405,
      "learning_rate": 1.1380913833655382e-05,
      "loss": 1.2303,
      "step": 846
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.1855391412973404,
      "learning_rate": 1.1190897033940461e-05,
      "loss": 1.2306,
      "step": 847
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.11130934208631516,
      "learning_rate": 1.1002385827060602e-05,
      "loss": 1.373,
      "step": 848
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.10244546085596085,
      "learning_rate": 1.0815383408871982e-05,
      "loss": 1.3044,
      "step": 849
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.11533805727958679,
      "learning_rate": 1.0629892949652132e-05,
      "loss": 1.0704,
      "step": 850
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.12121516466140747,
      "learning_rate": 1.0445917594046072e-05,
      "loss": 1.265,
      "step": 851
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.1047421246767044,
      "learning_rate": 1.026346046101312e-05,
      "loss": 1.1294,
      "step": 852
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.10979347676038742,
      "learning_rate": 1.0082524643773916e-05,
      "loss": 0.9608,
      "step": 853
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.12287425249814987,
      "learning_rate": 9.903113209758096e-06,
      "loss": 0.8795,
      "step": 854
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.11486369371414185,
      "learning_rate": 9.725229200552122e-06,
      "loss": 1.3881,
      "step": 855
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.11395483464002609,
      "learning_rate": 9.548875631847875e-06,
      "loss": 1.1552,
      "step": 856
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.10814516991376877,
      "learning_rate": 9.374055493391455e-06,
      "loss": 1.2155,
      "step": 857
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.10679764300584793,
      "learning_rate": 9.200771748932513e-06,
      "loss": 1.1449,
      "step": 858
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.1089312732219696,
      "learning_rate": 9.029027336174022e-06,
      "loss": 1.2221,
      "step": 859
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.11471953988075256,
      "learning_rate": 8.858825166722417e-06,
      "loss": 0.9955,
      "step": 860
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.11357836425304413,
      "learning_rate": 8.6901681260383e-06,
      "loss": 1.0855,
      "step": 861
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.11274585127830505,
      "learning_rate": 8.523059073387474e-06,
      "loss": 1.1812,
      "step": 862
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.10156439989805222,
      "learning_rate": 8.35750084179251e-06,
      "loss": 1.1702,
      "step": 863
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.10595791786909103,
      "learning_rate": 8.193496237984677e-06,
      "loss": 1.1719,
      "step": 864
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.10934492200613022,
      "learning_rate": 8.031048042356392e-06,
      "loss": 1.1735,
      "step": 865
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.10104786604642868,
      "learning_rate": 7.870159008914069e-06,
      "loss": 1.1967,
      "step": 866
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.10521814227104187,
      "learning_rate": 7.710831865231461e-06,
      "loss": 1.3094,
      "step": 867
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.1166355162858963,
      "learning_rate": 7.553069312403349e-06,
      "loss": 1.2912,
      "step": 868
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.09746871888637543,
      "learning_rate": 7.3968740249998115e-06,
      "loss": 1.1133,
      "step": 869
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.10336558520793915,
      "learning_rate": 7.242248651020844e-06,
      "loss": 1.1893,
      "step": 870
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.11591080576181412,
      "learning_rate": 7.089195811851501e-06,
      "loss": 1.1395,
      "step": 871
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.22089900076389313,
      "learning_rate": 6.937718102217461e-06,
      "loss": 1.1993,
      "step": 872
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.108499675989151,
      "learning_rate": 6.787818090140985e-06,
      "loss": 1.1195,
      "step": 873
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.10180932283401489,
      "learning_rate": 6.639498316897419e-06,
      "loss": 1.3188,
      "step": 874
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.10950213670730591,
      "learning_rate": 6.492761296972116e-06,
      "loss": 1.1373,
      "step": 875
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.09740964323282242,
      "learning_rate": 6.347609518017761e-06,
      "loss": 1.1551,
      "step": 876
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.10599476099014282,
      "learning_rate": 6.2040454408123e-06,
      "loss": 1.2161,
      "step": 877
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.11540497094392776,
      "learning_rate": 6.062071499217081e-06,
      "loss": 1.256,
      "step": 878
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.10363679379224777,
      "learning_rate": 5.921690100135713e-06,
      "loss": 1.2462,
      "step": 879
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.10643389075994492,
      "learning_rate": 5.782903623473201e-06,
      "loss": 1.34,
      "step": 880
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.14184726774692535,
      "learning_rate": 5.645714422095627e-06,
      "loss": 1.0978,
      "step": 881
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.11375614255666733,
      "learning_rate": 5.510124821790208e-06,
      "loss": 1.2334,
      "step": 882
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.11842206120491028,
      "learning_rate": 5.376137121225933e-06,
      "loss": 1.2043,
      "step": 883
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.1027233749628067,
      "learning_rate": 5.24375359191458e-06,
      "loss": 1.2038,
      "step": 884
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.09779934585094452,
      "learning_rate": 5.11297647817216e-06,
      "loss": 1.2799,
      "step": 885
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.28459811210632324,
      "learning_rate": 4.983807997080925e-06,
      "loss": 1.2972,
      "step": 886
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.12350057065486908,
      "learning_rate": 4.856250338451762e-06,
      "loss": 1.1445,
      "step": 887
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.10094188153743744,
      "learning_rate": 4.73030566478706e-06,
      "loss": 1.0144,
      "step": 888
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.10270640254020691,
      "learning_rate": 4.605976111244015e-06,
      "loss": 1.1673,
      "step": 889
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.12842753529548645,
      "learning_rate": 4.483263785598524e-06,
      "loss": 1.1681,
      "step": 890
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.10717935115098953,
      "learning_rate": 4.362170768209406e-06,
      "loss": 1.2661,
      "step": 891
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.20119836926460266,
      "learning_rate": 4.24269911198305e-06,
      "loss": 1.1492,
      "step": 892
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.11531922966241837,
      "learning_rate": 4.124850842338779e-06,
      "loss": 1.2629,
      "step": 893
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.10086618363857269,
      "learning_rate": 4.008627957174372e-06,
      "loss": 1.1223,
      "step": 894
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.10228842496871948,
      "learning_rate": 3.8940324268322285e-06,
      "loss": 1.2285,
      "step": 895
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.1152658760547638,
      "learning_rate": 3.7810661940660517e-06,
      "loss": 1.3102,
      "step": 896
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.09905364364385605,
      "learning_rate": 3.66973117400774e-06,
      "loss": 1.2029,
      "step": 897
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.11741360276937485,
      "learning_rate": 3.5600292541351e-06,
      "loss": 1.0504,
      "step": 898
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.10663048177957535,
      "learning_rate": 3.451962294239741e-06,
      "loss": 1.0434,
      "step": 899
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.13997888565063477,
      "learning_rate": 3.3455321263955786e-06,
      "loss": 1.1909,
      "step": 900
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.10722199827432632,
      "learning_rate": 3.240740554927768e-06,
      "loss": 1.1066,
      "step": 901
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.10431613773107529,
      "learning_rate": 3.1375893563820756e-06,
      "loss": 1.1097,
      "step": 902
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.10055112838745117,
      "learning_rate": 3.0360802794948686e-06,
      "loss": 1.119,
      "step": 903
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.14287303388118744,
      "learning_rate": 2.936215045163371e-06,
      "loss": 1.1977,
      "step": 904
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.11945986747741699,
      "learning_rate": 2.837995346416533e-06,
      "loss": 1.0618,
      "step": 905
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.11288944631814957,
      "learning_rate": 2.7414228483862657e-06,
      "loss": 1.1562,
      "step": 906
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.11274345964193344,
      "learning_rate": 2.6464991882793278e-06,
      "loss": 1.2794,
      "step": 907
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.10257246345281601,
      "learning_rate": 2.5532259753494827e-06,
      "loss": 1.1981,
      "step": 908
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.11306274682283401,
      "learning_rate": 2.4616047908702087e-06,
      "loss": 1.2817,
      "step": 909
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.1063089668750763,
      "learning_rate": 2.3716371881079557e-06,
      "loss": 1.2429,
      "step": 910
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.11371003836393356,
      "learning_rate": 2.2833246922957407e-06,
      "loss": 1.2017,
      "step": 911
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.10964678227901459,
      "learning_rate": 2.1966688006073823e-06,
      "loss": 1.0438,
      "step": 912
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.10942777991294861,
      "learning_rate": 2.111670982132008e-06,
      "loss": 1.1411,
      "step": 913
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.20406045019626617,
      "learning_rate": 2.028332677849254e-06,
      "loss": 1.26,
      "step": 914
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.1080511286854744,
      "learning_rate": 1.9466553006047383e-06,
      "loss": 1.2078,
      "step": 915
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.10461273789405823,
      "learning_rate": 1.86664023508617e-06,
      "loss": 1.0991,
      "step": 916
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.14060625433921814,
      "learning_rate": 1.7882888377998786e-06,
      "loss": 1.1999,
      "step": 917
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.10702642798423767,
      "learning_rate": 1.7116024370477878e-06,
      "loss": 1.1649,
      "step": 918
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.12004446983337402,
      "learning_rate": 1.6365823329049124e-06,
      "loss": 1.1923,
      "step": 919
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.12491834163665771,
      "learning_rate": 1.5632297971972965e-06,
      "loss": 1.1329,
      "step": 920
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.11271212995052338,
      "learning_rate": 1.4915460734805097e-06,
      "loss": 1.1198,
      "step": 921
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.11066664755344391,
      "learning_rate": 1.4215323770184641e-06,
      "loss": 1.2727,
      "step": 922
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.12216173112392426,
      "learning_rate": 1.3531898947629296e-06,
      "loss": 1.1179,
      "step": 923
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.10309845954179764,
      "learning_rate": 1.2865197853333178e-06,
      "loss": 1.328,
      "step": 924
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.1231243684887886,
      "learning_rate": 1.221523178997075e-06,
      "loss": 1.1911,
      "step": 925
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.12335013598203659,
      "learning_rate": 1.158201177650553e-06,
      "loss": 1.2257,
      "step": 926
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.11128874868154526,
      "learning_rate": 1.0965548548002803e-06,
      "loss": 1.2306,
      "step": 927
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.10876341909170151,
      "learning_rate": 1.036585255544764e-06,
      "loss": 1.2044,
      "step": 928
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.12024974822998047,
      "learning_rate": 9.782933965567954e-07,
      "loss": 1.1953,
      "step": 929
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.11236795783042908,
      "learning_rate": 9.21680266066216e-07,
      "loss": 1.3033,
      "step": 930
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.11030973494052887,
      "learning_rate": 8.667468238431453e-07,
      "loss": 1.2,
      "step": 931
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.10306066274642944,
      "learning_rate": 8.13494001181725e-07,
      "loss": 1.3141,
      "step": 932
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.10356069356203079,
      "learning_rate": 7.619227008843322e-07,
      "loss": 1.2244,
      "step": 933
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.10682324320077896,
      "learning_rate": 7.120337972462365e-07,
      "loss": 1.1282,
      "step": 934
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.11171166598796844,
      "learning_rate": 6.638281360408339e-07,
      "loss": 1.1534,
      "step": 935
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.09517676383256912,
      "learning_rate": 6.173065345052687e-07,
      "loss": 1.1292,
      "step": 936
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.11539721488952637,
      "learning_rate": 5.724697813265901e-07,
      "loss": 1.1716,
      "step": 937
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.11124927550554276,
      "learning_rate": 5.293186366284176e-07,
      "loss": 1.1894,
      "step": 938
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.261902779340744,
      "learning_rate": 4.878538319579629e-07,
      "loss": 1.2252,
      "step": 939
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.11507264524698257,
      "learning_rate": 4.4807607027372855e-07,
      "loss": 1.2918,
      "step": 940
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.1210726797580719,
      "learning_rate": 4.0998602593352863e-07,
      "loss": 1.1893,
      "step": 941
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.11546792089939117,
      "learning_rate": 3.7358434468308667e-07,
      "loss": 1.1486,
      "step": 942
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.12196337431669235,
      "learning_rate": 3.388716436450667e-07,
      "loss": 1.3495,
      "step": 943
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.10631462186574936,
      "learning_rate": 3.0584851130861513e-07,
      "loss": 1.1866,
      "step": 944
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.14002542197704315,
      "learning_rate": 2.745155075194128e-07,
      "loss": 0.9632,
      "step": 945
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.15344469249248505,
      "learning_rate": 2.448731634701384e-07,
      "loss": 1.3407,
      "step": 946
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.11114881932735443,
      "learning_rate": 2.169219816914869e-07,
      "loss": 1.1723,
      "step": 947
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.09970024973154068,
      "learning_rate": 1.9066243604367594e-07,
      "loss": 1.3012,
      "step": 948
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.10021547228097916,
      "learning_rate": 1.6609497170834155e-07,
      "loss": 1.1621,
      "step": 949
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.11028821766376495,
      "learning_rate": 1.4322000518106616e-07,
      "loss": 1.092,
      "step": 950
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.10497645288705826,
      "learning_rate": 1.220379242642844e-07,
      "loss": 1.1194,
      "step": 951
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.10754618048667908,
      "learning_rate": 1.025490880606883e-07,
      "loss": 1.0586,
      "step": 952
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.1118793711066246,
      "learning_rate": 8.475382696719881e-08,
      "loss": 1.1977,
      "step": 953
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.11205057799816132,
      "learning_rate": 6.865244266928139e-08,
      "loss": 0.944,
      "step": 954
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.10550331324338913,
      "learning_rate": 5.424520813590572e-08,
      "loss": 1.1463,
      "step": 955
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.10324594378471375,
      "learning_rate": 4.153236761488266e-08,
      "loss": 1.1325,
      "step": 956
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.10522052645683289,
      "learning_rate": 3.051413662874536e-08,
      "loss": 1.1922,
      "step": 957
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.12036368250846863,
      "learning_rate": 2.1190701971052218e-08,
      "loss": 1.123,
      "step": 958
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.11026199162006378,
      "learning_rate": 1.3562221703267152e-08,
      "loss": 1.261,
      "step": 959
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.28202083706855774,
      "learning_rate": 7.628825152050657e-09,
      "loss": 1.2547,
      "step": 960
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.10787337273359299,
      "learning_rate": 3.390612907094859e-09,
      "loss": 1.2832,
      "step": 961
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.10484518110752106,
      "learning_rate": 8.476568193804824e-10,
      "loss": 1.2884,
      "step": 962
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.11853493005037308,
      "learning_rate": 0.0,
      "loss": 1.2526,
      "step": 963
    },
    {
      "epoch": 1.0,
      "step": 963,
      "total_flos": 1.3034774175330337e+19,
      "train_loss": 1.2195818047285822,
      "train_runtime": 50845.1793,
      "train_samples_per_second": 0.152,
      "train_steps_per_second": 0.019
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 963,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 1.3034774175330337e+19,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}