{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9991569613138914,
  "eval_steps": 100,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 2.4793388429752066e-06,
      "loss": 0.7395,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.958677685950413e-06,
      "loss": 0.5605,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 7.438016528925619e-06,
      "loss": 1.1672,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 9.917355371900826e-06,
      "loss": 0.9085,
      "step": 4
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.2396694214876033e-05,
      "loss": 0.7465,
      "step": 5
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.4876033057851239e-05,
      "loss": 1.1959,
      "step": 6
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.7355371900826445e-05,
      "loss": 0.7802,
      "step": 7
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.9834710743801653e-05,
      "loss": 0.7606,
      "step": 8
    },
    {
      "epoch": 0.0,
      "learning_rate": 2.2314049586776857e-05,
      "loss": 0.7707,
      "step": 9
    },
    {
      "epoch": 0.0,
      "learning_rate": 2.4793388429752065e-05,
      "loss": 1.0276,
      "step": 10
    },
    {
      "epoch": 0.0,
      "learning_rate": 2.727272727272727e-05,
      "loss": 0.7283,
      "step": 11
    },
    {
      "epoch": 0.0,
      "learning_rate": 2.9752066115702478e-05,
      "loss": 0.842,
      "step": 12
    },
    {
      "epoch": 0.0,
      "learning_rate": 3.2231404958677685e-05,
      "loss": 0.9745,
      "step": 13
    },
    {
      "epoch": 0.0,
      "learning_rate": 3.471074380165289e-05,
      "loss": 0.7545,
      "step": 14
    },
    {
      "epoch": 0.0,
      "learning_rate": 3.7190082644628094e-05,
      "loss": 0.8536,
      "step": 15
    },
    {
      "epoch": 0.0,
      "learning_rate": 3.9669421487603306e-05,
      "loss": 0.6765,
      "step": 16
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.214876033057851e-05,
      "loss": 1.1224,
      "step": 17
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.4628099173553715e-05,
      "loss": 0.834,
      "step": 18
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.7107438016528926e-05,
      "loss": 0.7396,
      "step": 19
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.958677685950413e-05,
      "loss": 1.1275,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.2066115702479335e-05,
      "loss": 0.6943,
      "step": 21
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.454545454545454e-05,
      "loss": 0.9206,
      "step": 22
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.702479338842975e-05,
      "loss": 0.5686,
      "step": 23
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.9504132231404955e-05,
      "loss": 0.7646,
      "step": 24
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.198347107438017e-05,
      "loss": 0.8085,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.446280991735537e-05,
      "loss": 0.7863,
      "step": 26
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.694214876033058e-05,
      "loss": 0.6307,
      "step": 27
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.942148760330578e-05,
      "loss": 0.6614,
      "step": 28
    },
    {
      "epoch": 0.01,
      "learning_rate": 7.190082644628098e-05,
      "loss": 0.4608,
      "step": 29
    },
    {
      "epoch": 0.01,
      "learning_rate": 7.438016528925619e-05,
      "loss": 0.4663,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 7.68595041322314e-05,
      "loss": 0.6377,
      "step": 31
    },
    {
      "epoch": 0.01,
      "learning_rate": 7.933884297520661e-05,
      "loss": 0.9795,
      "step": 32
    },
    {
      "epoch": 0.01,
      "learning_rate": 8.18181818181818e-05,
      "loss": 0.7535,
      "step": 33
    },
    {
      "epoch": 0.01,
      "learning_rate": 8.429752066115702e-05,
      "loss": 0.9393,
      "step": 34
    },
    {
      "epoch": 0.01,
      "learning_rate": 8.677685950413221e-05,
      "loss": 0.8176,
      "step": 35
    },
    {
      "epoch": 0.01,
      "learning_rate": 8.925619834710743e-05,
      "loss": 0.6978,
      "step": 36
    },
    {
      "epoch": 0.01,
      "learning_rate": 9.173553719008263e-05,
      "loss": 0.6523,
      "step": 37
    },
    {
      "epoch": 0.01,
      "learning_rate": 9.421487603305785e-05,
      "loss": 0.7531,
      "step": 38
    },
    {
      "epoch": 0.01,
      "learning_rate": 9.669421487603304e-05,
      "loss": 0.8382,
      "step": 39
    },
    {
      "epoch": 0.01,
      "learning_rate": 9.917355371900826e-05,
      "loss": 0.6094,
      "step": 40
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00010165289256198345,
      "loss": 0.8196,
      "step": 41
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00010413223140495867,
      "loss": 0.6414,
      "step": 42
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00010661157024793387,
      "loss": 0.9574,
      "step": 43
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00010909090909090908,
      "loss": 0.655,
      "step": 44
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00011157024793388428,
      "loss": 0.7176,
      "step": 45
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0001140495867768595,
      "loss": 0.4037,
      "step": 46
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00011652892561983469,
      "loss": 0.2765,
      "step": 47
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00011900826446280991,
      "loss": 0.743,
      "step": 48
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00012148760330578511,
      "loss": 0.6934,
      "step": 49
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00012396694214876033,
      "loss": 0.8232,
      "step": 50
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00012396694214876033,
      "loss": 1.4057,
      "step": 51
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00012644628099173552,
      "loss": 0.4353,
      "step": 52
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00012892561983471074,
      "loss": 1.235,
      "step": 53
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00013140495867768593,
      "loss": 0.8951,
      "step": 54
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00013388429752066115,
      "loss": 0.2833,
      "step": 55
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00013636363636363634,
      "loss": 0.3174,
      "step": 56
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00013884297520661156,
      "loss": 0.6171,
      "step": 57
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00014132231404958675,
      "loss": 0.8656,
      "step": 58
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00014380165289256197,
      "loss": 1.1583,
      "step": 59
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00014628099173553716,
      "loss": 0.6687,
      "step": 60
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00014876033057851238,
      "loss": 0.8528,
      "step": 61
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0001512396694214876,
      "loss": 0.6429,
      "step": 62
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0001537190082644628,
      "loss": 0.4651,
      "step": 63
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.000156198347107438,
      "loss": 0.9373,
      "step": 64
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00015867768595041322,
      "loss": 0.5622,
      "step": 65
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00016115702479338841,
      "loss": 0.9218,
      "step": 66
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0001636363636363636,
      "loss": 0.2916,
      "step": 67
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00016611570247933885,
      "loss": 1.9224,
      "step": 68
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00016859504132231404,
      "loss": 0.2946,
      "step": 69
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00017107438016528923,
      "loss": 0.5455,
      "step": 70
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00017355371900826442,
      "loss": 0.9446,
      "step": 71
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00017603305785123967,
      "loss": 0.5889,
      "step": 72
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00017851239669421486,
      "loss": 0.5363,
      "step": 73
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00018099173553719008,
      "loss": 0.551,
      "step": 74
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00018347107438016527,
      "loss": 0.5523,
      "step": 75
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00018595041322314049,
      "loss": 0.4451,
      "step": 76
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0001884297520661157,
      "loss": 0.4374,
      "step": 77
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0001909090909090909,
      "loss": 0.6954,
      "step": 78
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019338842975206609,
      "loss": 0.8697,
      "step": 79
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019586776859504133,
      "loss": 0.9998,
      "step": 80
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019834710743801652,
      "loss": 0.3688,
      "step": 81
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002008264462809917,
      "loss": 0.7807,
      "step": 82
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002033057851239669,
      "loss": 0.4585,
      "step": 83
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00020578512396694215,
      "loss": 0.9171,
      "step": 84
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00020826446280991734,
      "loss": 0.689,
      "step": 85
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00021074380165289253,
      "loss": 0.7177,
      "step": 86
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00021322314049586775,
      "loss": 0.4259,
      "step": 87
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00021570247933884297,
      "loss": 0.8513,
      "step": 88
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00021818181818181816,
      "loss": 0.6856,
      "step": 89
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00022066115702479338,
      "loss": 0.4438,
      "step": 90
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00022314049586776857,
      "loss": 0.7912,
      "step": 91
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00022561983471074378,
      "loss": 0.4966,
      "step": 92
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.000228099173553719,
      "loss": 0.4537,
      "step": 93
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002305785123966942,
      "loss": 0.3843,
      "step": 94
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00023305785123966938,
      "loss": 0.5328,
      "step": 95
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00023553719008264463,
      "loss": 0.3251,
      "step": 96
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00023801652892561982,
      "loss": 0.3926,
      "step": 97
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.000240495867768595,
      "loss": 0.4101,
      "step": 98
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00024297520661157023,
      "loss": 1.1283,
      "step": 99
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00024545454545454545,
      "loss": 0.4546,
      "step": 100
    },
    {
      "epoch": 0.02,
      "eval_loss": 0.5531858801841736,
      "eval_runtime": 50.2969,
      "eval_samples_per_second": 0.656,
      "eval_steps_per_second": 0.656,
      "step": 100
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00024793388429752067,
      "loss": 0.5849,
      "step": 101
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00025041322314049583,
      "loss": 0.5361,
      "step": 102
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00025289256198347105,
      "loss": 1.1766,
      "step": 103
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00025537190082644627,
      "loss": 0.6957,
      "step": 104
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002578512396694215,
      "loss": 0.4606,
      "step": 105
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00026033057851239665,
      "loss": 0.4176,
      "step": 106
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00026280991735537187,
      "loss": 0.3716,
      "step": 107
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002652892561983471,
      "loss": 0.4438,
      "step": 108
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002677685950413223,
      "loss": 1.0926,
      "step": 109
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002702479338842975,
      "loss": 0.8791,
      "step": 110
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002727272727272727,
      "loss": 0.8508,
      "step": 111
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002752066115702479,
      "loss": 0.3589,
      "step": 112
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002776859504132231,
      "loss": 1.1555,
      "step": 113
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00028016528925619834,
      "loss": 1.0118,
      "step": 114
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002826446280991735,
      "loss": 0.9295,
      "step": 115
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002851239669421488,
      "loss": 0.5785,
      "step": 116
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00028760330578512394,
      "loss": 0.9671,
      "step": 117
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00029008264462809916,
      "loss": 0.2695,
      "step": 118
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002925619834710743,
      "loss": 0.7017,
      "step": 119
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002950413223140496,
      "loss": 0.4282,
      "step": 120
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00029752066115702476,
      "loss": 0.7802,
      "step": 121
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0003,
      "loss": 0.5957,
      "step": 122
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.000299999950880954,
      "loss": 0.721,
      "step": 123
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002999998035238483,
      "loss": 1.2606,
      "step": 124
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00029999955792877925,
      "loss": 0.5951,
      "step": 125
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00029999921409590784,
      "loss": 0.4787,
      "step": 126
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00029999877202545924,
      "loss": 0.9637,
      "step": 127
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002999982317177229,
      "loss": 0.5446,
      "step": 128
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002999975931730527,
      "loss": 0.7236,
      "step": 129
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00029999685639186684,
      "loss": 0.2906,
      "step": 130
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00029999602137464784,
      "loss": 0.4878,
      "step": 131
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00029999508812194263,
      "loss": 0.4158,
      "step": 132
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002999940566343624,
      "loss": 0.4712,
      "step": 133
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002999929269125826,
      "loss": 0.8673,
      "step": 134
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002999916989573432,
      "loss": 0.7469,
      "step": 135
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002999903727694484,
      "loss": 0.7669,
      "step": 136
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002999889483497667,
      "loss": 0.5049,
      "step": 137
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002999874256992311,
      "loss": 0.6145,
      "step": 138
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002999858048188387,
      "loss": 0.509,
      "step": 139
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00029998408570965107,
      "loss": 0.7502,
      "step": 140
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029998226837279406,
      "loss": 0.1854,
      "step": 141
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029998035280945796,
      "loss": 0.548,
      "step": 142
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999783390208973,
      "loss": 0.4033,
      "step": 143
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999762270084309,
      "loss": 0.357,
      "step": 144
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029997401677344195,
      "loss": 0.6885,
      "step": 145
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029997170831737806,
      "loss": 0.6436,
      "step": 146
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999693016417511,
      "loss": 0.7707,
      "step": 147
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029996679674813706,
      "loss": 0.6878,
      "step": 148
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999641936381767,
      "loss": 0.8969,
      "step": 149
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999614923135747,
      "loss": 0.627,
      "step": 150
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999586927761002,
      "loss": 0.9229,
      "step": 151
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999557950275867,
      "loss": 0.6693,
      "step": 152
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999527990699321,
      "loss": 0.5707,
      "step": 153
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999497049050983,
      "loss": 0.4924,
      "step": 154
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029994651253511194,
      "loss": 0.7528,
      "step": 155
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029994322196206374,
      "loss": 0.7359,
      "step": 156
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029993983318810866,
      "loss": 0.5627,
      "step": 157
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999363462154661,
      "loss": 0.4116,
      "step": 158
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029993276104641983,
      "loss": 0.2696,
      "step": 159
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999290776833177,
      "loss": 0.449,
      "step": 160
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999252961285723,
      "loss": 0.3702,
      "step": 161
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029992141638465996,
      "loss": 0.8138,
      "step": 162
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999174384541217,
      "loss": 0.354,
      "step": 163
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029991336233956277,
      "loss": 0.4235,
      "step": 164
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029990918804365276,
      "loss": 0.8132,
      "step": 165
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999049155691254,
      "loss": 0.5083,
      "step": 166
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002999005449187788,
      "loss": 0.811,
      "step": 167
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029989607609547553,
      "loss": 0.8583,
      "step": 168
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029989150910214216,
      "loss": 0.4596,
      "step": 169
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029988684394176984,
      "loss": 0.4818,
      "step": 170
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002998820806174138,
      "loss": 0.6915,
      "step": 171
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029987721913219366,
      "loss": 0.5329,
      "step": 172
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002998722594892933,
      "loss": 0.8095,
      "step": 173
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002998672016919609,
      "loss": 0.6161,
      "step": 174
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029986204574350894,
      "loss": 0.2796,
      "step": 175
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002998567916473141,
      "loss": 0.8323,
      "step": 176
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002998514394068174,
      "loss": 0.4128,
      "step": 177
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002998459890255242,
      "loss": 0.5931,
      "step": 178
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002998404405070041,
      "loss": 0.2607,
      "step": 179
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029983479385489075,
      "loss": 0.4156,
      "step": 180
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002998290490728824,
      "loss": 0.2804,
      "step": 181
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002998232061647414,
      "loss": 0.5357,
      "step": 182
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002998172651342944,
      "loss": 0.5178,
      "step": 183
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029981122598543233,
      "loss": 0.54,
      "step": 184
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029980508872211027,
      "loss": 0.6063,
      "step": 185
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029979885334834776,
      "loss": 0.5029,
      "step": 186
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029979251986822833,
      "loss": 0.5525,
      "step": 187
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002997860882859,
      "loss": 0.7016,
      "step": 188
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002997795586055749,
      "loss": 1.0331,
      "step": 189
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029977293083152955,
      "loss": 0.6561,
      "step": 190
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002997662049681045,
      "loss": 0.5247,
      "step": 191
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002997593810197047,
      "loss": 0.8221,
      "step": 192
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002997524589907993,
      "loss": 0.3154,
      "step": 193
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029974543888592166,
      "loss": 0.6263,
      "step": 194
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002997383207096694,
      "loss": 0.9541,
      "step": 195
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029973110446670435,
      "loss": 0.668,
      "step": 196
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029972379016175266,
      "loss": 0.7415,
      "step": 197
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029971637779960454,
      "loss": 0.1387,
      "step": 198
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029970886738511446,
      "loss": 0.6723,
      "step": 199
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002997012589232012,
      "loss": 1.0751,
      "step": 200
    },
    {
      "epoch": 0.05,
      "eval_loss": 0.4962003231048584,
      "eval_runtime": 51.2848,
      "eval_samples_per_second": 0.643,
      "eval_steps_per_second": 0.643,
      "step": 200
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029969355241884777,
      "loss": 0.5455,
      "step": 201
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002996857478771012,
      "loss": 0.4332,
      "step": 202
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002996778453030728,
      "loss": 0.4506,
      "step": 203
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029966984470193835,
      "loss": 0.1448,
      "step": 204
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002996617460789374,
      "loss": 0.46,
      "step": 205
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.000299653549439374,
      "loss": 0.2477,
      "step": 206
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002996452547886162,
      "loss": 0.4066,
      "step": 207
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029963686213209655,
      "loss": 0.6116,
      "step": 208
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029962837147531137,
      "loss": 0.645,
      "step": 209
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002996197828238215,
      "loss": 0.843,
      "step": 210
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002996110961832517,
      "loss": 0.181,
      "step": 211
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029960231155929106,
      "loss": 0.2407,
      "step": 212
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002995934289576929,
      "loss": 0.2553,
      "step": 213
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029958444838427457,
      "loss": 0.623,
      "step": 214
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029957536984491755,
      "loss": 0.2728,
      "step": 215
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002995661933455677,
      "loss": 0.3567,
      "step": 216
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029955691889223485,
      "loss": 0.8278,
      "step": 217
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029954754649099297,
      "loss": 0.7734,
      "step": 218
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002995380761479803,
      "loss": 0.6057,
      "step": 219
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002995285078693992,
      "loss": 0.4046,
      "step": 220
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029951884166151596,
      "loss": 0.4979,
      "step": 221
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029950907753066136,
      "loss": 0.5038,
      "step": 222
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029949921548323005,
      "loss": 0.6706,
      "step": 223
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029948925552568097,
      "loss": 0.4975,
      "step": 224
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029947919766453704,
      "loss": 0.6313,
      "step": 225
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029946904190638533,
      "loss": 0.3172,
      "step": 226
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029945878825787705,
      "loss": 0.8056,
      "step": 227
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002994484367257276,
      "loss": 0.4614,
      "step": 228
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029943798731671634,
      "loss": 0.6569,
      "step": 229
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029942744003768685,
      "loss": 0.3018,
      "step": 230
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002994167948955468,
      "loss": 0.4507,
      "step": 231
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002994060518972678,
      "loss": 0.9597,
      "step": 232
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029939521104988573,
      "loss": 0.2078,
      "step": 233
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002993842723605005,
      "loss": 0.5651,
      "step": 234
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.000299373235836276,
      "loss": 0.315,
      "step": 235
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002993621014844404,
      "loss": 0.4966,
      "step": 236
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002993508693122857,
      "loss": 0.387,
      "step": 237
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002993395393271682,
      "loss": 0.5883,
      "step": 238
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.000299328111536508,
      "loss": 0.4505,
      "step": 239
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029931658594778947,
      "loss": 0.477,
      "step": 240
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.000299304962568561,
      "loss": 0.7098,
      "step": 241
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002992932414064349,
      "loss": 0.908,
      "step": 242
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002992814224690877,
      "loss": 0.5153,
      "step": 243
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002992695057642597,
      "loss": 0.2368,
      "step": 244
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002992574912997556,
      "loss": 0.2808,
      "step": 245
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029924537908344374,
      "loss": 0.4408,
      "step": 246
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002992331691232568,
      "loss": 0.2643,
      "step": 247
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029922086142719124,
      "loss": 0.7206,
      "step": 248
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002992084560033076,
      "loss": 0.5388,
      "step": 249
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002991959528597306,
      "loss": 0.9601,
      "step": 250
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002991833520046487,
      "loss": 0.3647,
      "step": 251
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029917065344631444,
      "loss": 0.6112,
      "step": 252
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002991578571930444,
      "loss": 0.5584,
      "step": 253
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002991449632532191,
      "loss": 0.7914,
      "step": 254
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002991319716352831,
      "loss": 0.435,
      "step": 255
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002991188823477448,
      "loss": 0.2787,
      "step": 256
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002991056953991767,
      "loss": 0.4698,
      "step": 257
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029909241079821515,
      "loss": 0.2037,
      "step": 258
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002990790285535605,
      "loss": 0.2784,
      "step": 259
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002990655486739772,
      "loss": 0.271,
      "step": 260
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029905197116829335,
      "loss": 0.3445,
      "step": 261
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002990382960454012,
      "loss": 0.6735,
      "step": 262
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029902452331425684,
      "loss": 1.0058,
      "step": 263
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002990106529838803,
      "loss": 0.1816,
      "step": 264
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029899668506335565,
      "loss": 0.4174,
      "step": 265
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029898261956183065,
      "loss": 0.471,
      "step": 266
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029896845648851717,
      "loss": 0.6281,
      "step": 267
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002989541958526909,
      "loss": 0.5509,
      "step": 268
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002989398376636913,
      "loss": 0.5514,
      "step": 269
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029892538193092196,
      "loss": 0.9307,
      "step": 270
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002989108286638502,
      "loss": 0.3955,
      "step": 271
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002988961778720073,
      "loss": 0.811,
      "step": 272
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002988814295649883,
      "loss": 0.675,
      "step": 273
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029886658375245225,
      "loss": 0.4251,
      "step": 274
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002988516404441219,
      "loss": 0.7038,
      "step": 275
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002988365996497839,
      "loss": 0.4296,
      "step": 276
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002988214613792889,
      "loss": 0.8127,
      "step": 277
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002988062256425512,
      "loss": 0.4031,
      "step": 278
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029879089244954894,
      "loss": 0.6578,
      "step": 279
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002987754618103243,
      "loss": 0.81,
      "step": 280
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029875993373498295,
      "loss": 0.4904,
      "step": 281
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029874430823369463,
      "loss": 0.4027,
      "step": 282
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029872858531669285,
      "loss": 0.4876,
      "step": 283
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029871276499427476,
      "loss": 0.8471,
      "step": 284
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002986968472768015,
      "loss": 0.3108,
      "step": 285
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029868083217469786,
      "loss": 0.9183,
      "step": 286
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029866471969845255,
      "loss": 0.6407,
      "step": 287
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002986485098586178,
      "loss": 0.9156,
      "step": 288
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029863220266581,
      "loss": 0.3703,
      "step": 289
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029861579813070883,
      "loss": 0.5688,
      "step": 290
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002985992962640581,
      "loss": 0.3565,
      "step": 291
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029858269707666514,
      "loss": 0.5898,
      "step": 292
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002985660005794012,
      "loss": 0.79,
      "step": 293
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002985492067832011,
      "loss": 0.5207,
      "step": 294
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002985323156990634,
      "loss": 0.5156,
      "step": 295
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002985153273380505,
      "loss": 0.6698,
      "step": 296
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029849824171128837,
      "loss": 0.9935,
      "step": 297
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002984810588299668,
      "loss": 0.317,
      "step": 298
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029846377870533915,
      "loss": 0.4263,
      "step": 299
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029844640134872255,
      "loss": 0.374,
      "step": 300
    },
    {
      "epoch": 0.07,
      "eval_loss": 0.4789610207080841,
      "eval_runtime": 49.8741,
      "eval_samples_per_second": 0.662,
      "eval_steps_per_second": 0.662,
      "step": 300
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029842892677149787,
      "loss": 0.3992,
      "step": 301
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002984113549851094,
      "loss": 0.7148,
      "step": 302
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002983936860010654,
      "loss": 0.1513,
      "step": 303
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029837591983093763,
      "loss": 0.4799,
      "step": 304
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029835805648636146,
      "loss": 1.0202,
      "step": 305
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002983400959790361,
      "loss": 0.3289,
      "step": 306
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029832203832072404,
      "loss": 0.4568,
      "step": 307
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002983038835232518,
      "loss": 0.6284,
      "step": 308
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002982856315985093,
      "loss": 0.532,
      "step": 309
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029826728255845,
      "loss": 0.5261,
      "step": 310
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002982488364150912,
      "loss": 0.3229,
      "step": 311
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002982302931805136,
      "loss": 0.5588,
      "step": 312
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002982116528668614,
      "loss": 0.686,
      "step": 313
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029819291548634284,
      "loss": 0.3283,
      "step": 314
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002981740810512291,
      "loss": 1.1459,
      "step": 315
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002981551495738555,
      "loss": 0.3736,
      "step": 316
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002981361210666205,
      "loss": 0.4242,
      "step": 317
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029811699554198627,
      "loss": 0.3667,
      "step": 318
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029809777301247856,
      "loss": 0.544,
      "step": 319
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002980784534906866,
      "loss": 0.478,
      "step": 320
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002980590369892631,
      "loss": 0.7715,
      "step": 321
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002980395235209243,
      "loss": 0.4486,
      "step": 322
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002980199130984501,
      "loss": 0.5514,
      "step": 323
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029800020573468366,
      "loss": 0.455,
      "step": 324
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029798040144253177,
      "loss": 0.4602,
      "step": 325
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029796050023496467,
      "loss": 0.7685,
      "step": 326
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029794050212501606,
      "loss": 0.3284,
      "step": 327
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029792040712578314,
      "loss": 0.5873,
      "step": 328
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002979002152504265,
      "loss": 0.3321,
      "step": 329
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002978799265121703,
      "loss": 0.8255,
      "step": 330
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029785954092430197,
      "loss": 0.4082,
      "step": 331
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029783905850017247,
      "loss": 0.6256,
      "step": 332
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029781847925319614,
      "loss": 0.6969,
      "step": 333
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029779780319685075,
      "loss": 0.5103,
      "step": 334
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029777703034467755,
      "loss": 0.5422,
      "step": 335
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029775616071028106,
      "loss": 0.8769,
      "step": 336
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029773519430732927,
      "loss": 0.3917,
      "step": 337
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002977141311495535,
      "loss": 0.2997,
      "step": 338
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029769297125074835,
      "loss": 0.3013,
      "step": 339
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029767171462477196,
      "loss": 0.607,
      "step": 340
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002976503612855457,
      "loss": 0.8018,
      "step": 341
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002976289112470544,
      "loss": 0.5352,
      "step": 342
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029760736452334605,
      "loss": 0.6256,
      "step": 343
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.000297585721128532,
      "loss": 0.6408,
      "step": 344
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029756398107678707,
      "loss": 0.552,
      "step": 345
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002975421443823492,
      "loss": 1.2506,
      "step": 346
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002975202110595197,
      "loss": 0.5492,
      "step": 347
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002974981811226631,
      "loss": 0.3975,
      "step": 348
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002974760545862074,
      "loss": 1.0081,
      "step": 349
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002974538314646436,
      "loss": 1.1184,
      "step": 350
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002974315117725262,
      "loss": 0.7059,
      "step": 351
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002974090955244726,
      "loss": 0.5686,
      "step": 352
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002973865827351639,
      "loss": 0.2542,
      "step": 353
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029736397341934413,
      "loss": 0.4502,
      "step": 354
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002973412675918205,
      "loss": 0.4286,
      "step": 355
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029731846526746364,
      "loss": 0.8042,
      "step": 356
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002972955664612072,
      "loss": 0.3343,
      "step": 357
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002972725711880481,
      "loss": 0.9204,
      "step": 358
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002972494794630464,
      "loss": 0.5911,
      "step": 359
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002972262913013254,
      "loss": 0.5391,
      "step": 360
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002972030067180714,
      "loss": 0.5557,
      "step": 361
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002971796257285341,
      "loss": 0.4905,
      "step": 362
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029715614834802606,
      "loss": 0.3964,
      "step": 363
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029713257459192317,
      "loss": 0.447,
      "step": 364
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029710890447566437,
      "loss": 0.3296,
      "step": 365
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029708513801475164,
      "loss": 0.6638,
      "step": 366
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002970612752247502,
      "loss": 0.2703,
      "step": 367
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002970373161212882,
      "loss": 0.536,
      "step": 368
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002970132607200571,
      "loss": 0.4098,
      "step": 369
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029698910903681115,
      "loss": 0.5572,
      "step": 370
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002969648610873678,
      "loss": 0.5759,
      "step": 371
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029694051688760757,
      "loss": 0.6074,
      "step": 372
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029691607645347396,
      "loss": 0.5367,
      "step": 373
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029689153980097353,
      "loss": 0.3434,
      "step": 374
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002968669069461758,
      "loss": 0.2744,
      "step": 375
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002968421779052133,
      "loss": 0.9684,
      "step": 376
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029681735269428177,
      "loss": 0.5153,
      "step": 377
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002967924313296395,
      "loss": 0.8236,
      "step": 378
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002967674138276082,
      "loss": 0.2987,
      "step": 379
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002967423002045723,
      "loss": 0.4323,
      "step": 380
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002967170904769792,
      "loss": 0.5062,
      "step": 381
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002966917846613393,
      "loss": 0.8246,
      "step": 382
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029666638277422584,
      "loss": 0.3215,
      "step": 383
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029664088483227514,
      "loss": 0.3224,
      "step": 384
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029661529085218616,
      "loss": 0.4085,
      "step": 385
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002965896008507211,
      "loss": 0.5961,
      "step": 386
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029656381484470485,
      "loss": 0.633,
      "step": 387
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002965379328510251,
      "loss": 0.4322,
      "step": 388
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002965119548866326,
      "loss": 0.2502,
      "step": 389
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029648588096854077,
      "loss": 0.3247,
      "step": 390
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.000296459711113826,
      "loss": 0.4435,
      "step": 391
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029643344533962745,
      "loss": 0.2223,
      "step": 392
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029640708366314715,
      "loss": 0.4167,
      "step": 393
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029638062610164986,
      "loss": 0.6656,
      "step": 394
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029635407267246324,
      "loss": 0.2642,
      "step": 395
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002963274233929776,
      "loss": 0.4816,
      "step": 396
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002963006782806462,
      "loss": 0.408,
      "step": 397
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002962738373529849,
      "loss": 0.5925,
      "step": 398
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029624690062757237,
      "loss": 0.5068,
      "step": 399
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029621986812205,
      "loss": 0.5416,
      "step": 400
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.4656181037425995,
      "eval_runtime": 49.9246,
      "eval_samples_per_second": 0.661,
      "eval_steps_per_second": 0.661,
      "step": 400
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029619273985412206,
      "loss": 0.5515,
      "step": 401
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002961655158415553,
      "loss": 0.5194,
      "step": 402
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029613819610217935,
      "loss": 0.2119,
      "step": 403
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002961107806538864,
      "loss": 0.6621,
      "step": 404
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029608326951463147,
      "loss": 0.6816,
      "step": 405
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002960556627024321,
      "loss": 0.4493,
      "step": 406
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002960279602353686,
      "loss": 0.5293,
      "step": 407
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002960001621315839,
      "loss": 0.6136,
      "step": 408
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029597226840928355,
      "loss": 0.398,
      "step": 409
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002959442790867357,
      "loss": 0.6128,
      "step": 410
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029591619418227114,
      "loss": 0.5258,
      "step": 411
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002958880137142832,
      "loss": 0.461,
      "step": 412
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029585973770122795,
      "loss": 0.3082,
      "step": 413
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002958313661616239,
      "loss": 0.4292,
      "step": 414
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002958028991140521,
      "loss": 0.3288,
      "step": 415
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002957743365771563,
      "loss": 0.6629,
      "step": 416
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002957456785696426,
      "loss": 0.1399,
      "step": 417
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002957169251102798,
      "loss": 0.5864,
      "step": 418
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002956880762178991,
      "loss": 0.4516,
      "step": 419
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029565913191139417,
      "loss": 0.5132,
      "step": 420
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002956300922097213,
      "loss": 0.4434,
      "step": 421
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029560095713189927,
      "loss": 0.1672,
      "step": 422
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002955717266970091,
      "loss": 0.7356,
      "step": 423
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029554240092419443,
      "loss": 0.6056,
      "step": 424
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002955129798326614,
      "loss": 0.432,
      "step": 425
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002954834634416784,
      "loss": 0.9932,
      "step": 426
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029545385177057637,
      "loss": 0.8388,
      "step": 427
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029542414483874855,
      "loss": 0.5872,
      "step": 428
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002953943426656507,
      "loss": 0.3774,
      "step": 429
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029536444527080086,
      "loss": 0.6615,
      "step": 430
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002953344526737794,
      "loss": 0.7478,
      "step": 431
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002953043648942292,
      "loss": 0.4358,
      "step": 432
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029527418195185516,
      "loss": 0.8076,
      "step": 433
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002952439038664249,
      "loss": 0.691,
      "step": 434
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029521353065776806,
      "loss": 0.2658,
      "step": 435
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029518306234577673,
      "loss": 0.3197,
      "step": 436
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029515249895040527,
      "loss": 0.8976,
      "step": 437
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002951218404916702,
      "loss": 0.1272,
      "step": 438
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002950910869896504,
      "loss": 0.6221,
      "step": 439
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.000295060238464487,
      "loss": 0.1978,
      "step": 440
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029502929493638335,
      "loss": 0.7059,
      "step": 441
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002949982564256049,
      "loss": 0.8963,
      "step": 442
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029496712295247956,
      "loss": 0.2643,
      "step": 443
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002949358945373972,
      "loss": 0.5589,
      "step": 444
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029490457120080987,
      "loss": 0.522,
      "step": 445
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029487315296323205,
      "loss": 0.6127,
      "step": 446
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029484163984524006,
      "loss": 0.5169,
      "step": 447
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029481003186747253,
      "loss": 0.595,
      "step": 448
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029477832905063016,
      "loss": 0.5026,
      "step": 449
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029474653141547583,
      "loss": 0.4744,
      "step": 450
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002947146389828344,
      "loss": 0.5491,
      "step": 451
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029468265177359294,
      "loss": 0.3451,
      "step": 452
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002946505698087005,
      "loss": 0.7748,
      "step": 453
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002946183931091682,
      "loss": 0.4013,
      "step": 454
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029458612169606924,
      "loss": 0.3768,
      "step": 455
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002945537555905389,
      "loss": 0.3243,
      "step": 456
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029452129481377426,
      "loss": 0.4168,
      "step": 457
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002944887393870347,
      "loss": 0.4272,
      "step": 458
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002944560893316414,
      "loss": 0.319,
      "step": 459
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002944233446689775,
      "loss": 0.177,
      "step": 460
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002943905054204882,
      "loss": 0.8348,
      "step": 461
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029435757160768065,
      "loss": 0.5352,
      "step": 462
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029432454325212373,
      "loss": 0.3998,
      "step": 463
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029429142037544855,
      "loss": 0.2362,
      "step": 464
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029425820299934784,
      "loss": 0.5784,
      "step": 465
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002942248911455765,
      "loss": 0.4959,
      "step": 466
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.000294191484835951,
      "loss": 0.6902,
      "step": 467
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002941579840923499,
      "loss": 0.5815,
      "step": 468
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002941243889367135,
      "loss": 0.4156,
      "step": 469
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002940906993910439,
      "loss": 0.9869,
      "step": 470
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002940569154774052,
      "loss": 0.3447,
      "step": 471
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002940230372179231,
      "loss": 0.5376,
      "step": 472
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029398906463478524,
      "loss": 0.3723,
      "step": 473
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002939549977502409,
      "loss": 0.6089,
      "step": 474
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002939208365866012,
      "loss": 0.4969,
      "step": 475
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.000293886581166239,
      "loss": 0.9688,
      "step": 476
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002938522315115888,
      "loss": 0.3806,
      "step": 477
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029381778764514706,
      "loss": 0.5527,
      "step": 478
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002937832495894717,
      "loss": 0.4304,
      "step": 479
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002937486173671823,
      "loss": 0.5994,
      "step": 480
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002937138910009604,
      "loss": 0.6778,
      "step": 481
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029367907051354886,
      "loss": 0.4,
      "step": 482
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029364415592775245,
      "loss": 0.4904,
      "step": 483
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002936091472664374,
      "loss": 1.0745,
      "step": 484
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002935740445525316,
      "loss": 0.3808,
      "step": 485
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029353884780902446,
      "loss": 0.7404,
      "step": 486
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002935035570589672,
      "loss": 0.7535,
      "step": 487
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029346817232547243,
      "loss": 0.3729,
      "step": 488
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002934326936317143,
      "loss": 0.2859,
      "step": 489
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002933971210009285,
      "loss": 0.7242,
      "step": 490
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029336145445641235,
      "loss": 0.2228,
      "step": 491
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002933256940215246,
      "loss": 0.294,
      "step": 492
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002932898397196854,
      "loss": 0.5114,
      "step": 493
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029325389157437666,
      "loss": 0.3415,
      "step": 494
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002932178496091414,
      "loss": 0.3491,
      "step": 495
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002931817138475843,
      "loss": 0.6835,
      "step": 496
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002931454843133714,
      "loss": 0.641,
      "step": 497
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002931091610302302,
      "loss": 0.4553,
      "step": 498
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029307274402194953,
      "loss": 0.2534,
      "step": 499
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029303623331237966,
      "loss": 0.4531,
      "step": 500
    },
    {
      "epoch": 0.12,
      "eval_loss": 0.4592454135417938,
      "eval_runtime": 50.2966,
      "eval_samples_per_second": 0.656,
      "eval_steps_per_second": 0.656,
      "step": 500
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002929996289254322,
      "loss": 0.3783,
      "step": 501
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029296293088508014,
      "loss": 0.5877,
      "step": 502
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002929261392153577,
      "loss": 0.6199,
      "step": 503
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002928892539403606,
      "loss": 0.4918,
      "step": 504
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029285227508424577,
      "loss": 0.4727,
      "step": 505
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029281520267123134,
      "loss": 0.7408,
      "step": 506
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029277803672559684,
      "loss": 0.3145,
      "step": 507
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029274077727168304,
      "loss": 0.7061,
      "step": 508
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.000292703424333892,
      "loss": 0.479,
      "step": 509
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029266597793668666,
      "loss": 0.3517,
      "step": 510
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002926284381045917,
      "loss": 0.2987,
      "step": 511
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029259080486219264,
      "loss": 0.3532,
      "step": 512
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029255307823413627,
      "loss": 0.4703,
      "step": 513
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002925152582451305,
      "loss": 0.6087,
      "step": 514
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029247734491994447,
      "loss": 0.6901,
      "step": 515
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029243933828340836,
      "loss": 0.4192,
      "step": 516
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029240123836041357,
      "loss": 0.4996,
      "step": 517
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029236304517591246,
      "loss": 0.3612,
      "step": 518
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002923247587549185,
      "loss": 0.9717,
      "step": 519
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029228637912250634,
      "loss": 0.4058,
      "step": 520
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029224790630381155,
      "loss": 0.3055,
      "step": 521
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029220934032403076,
      "loss": 0.6644,
      "step": 522
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002921706812084217,
      "loss": 0.3586,
      "step": 523
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002921319289823029,
      "loss": 0.4584,
      "step": 524
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029209308367105413,
      "loss": 0.3417,
      "step": 525
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002920541453001159,
      "loss": 0.1639,
      "step": 526
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002920151138949897,
      "loss": 0.5596,
      "step": 527
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002919759894812382,
      "loss": 0.1738,
      "step": 528
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002919367720844845,
      "loss": 0.5244,
      "step": 529
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002918974617304131,
      "loss": 0.4763,
      "step": 530
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002918580584447691,
      "loss": 0.452,
      "step": 531
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002918185622533585,
      "loss": 0.4377,
      "step": 532
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002917789731820482,
      "loss": 0.5392,
      "step": 533
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002917392912567659,
      "loss": 0.3871,
      "step": 534
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029169951650350004,
      "loss": 0.7525,
      "step": 535
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029165964894830004,
      "loss": 0.5328,
      "step": 536
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029161968861727595,
      "loss": 0.3854,
      "step": 537
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029157963553659853,
      "loss": 0.548,
      "step": 538
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029153948973249946,
      "loss": 0.617,
      "step": 539
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.000291499251231271,
      "loss": 0.3326,
      "step": 540
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029145892005926627,
      "loss": 0.4232,
      "step": 541
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029141849624289887,
      "loss": 0.6844,
      "step": 542
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002913779798086432,
      "loss": 0.8596,
      "step": 543
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002913373707830344,
      "loss": 0.6514,
      "step": 544
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029129666919266805,
      "loss": 0.4856,
      "step": 545
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029125587506420055,
      "loss": 0.4111,
      "step": 546
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029121498842434875,
      "loss": 0.4227,
      "step": 547
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002911740092998902,
      "loss": 0.5127,
      "step": 548
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029113293771766295,
      "loss": 0.292,
      "step": 549
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029109177370456563,
      "loss": 0.4071,
      "step": 550
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002910505172875574,
      "loss": 0.3155,
      "step": 551
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029100916849365794,
      "loss": 0.3224,
      "step": 552
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002909677273499474,
      "loss": 0.3452,
      "step": 553
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002909261938835665,
      "loss": 0.5044,
      "step": 554
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002908845681217163,
      "loss": 0.3062,
      "step": 555
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002908428500916584,
      "loss": 0.3906,
      "step": 556
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002908010398207148,
      "loss": 0.5677,
      "step": 557
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002907591373362679,
      "loss": 0.4429,
      "step": 558
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029071714266576057,
      "loss": 0.757,
      "step": 559
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029067505583669584,
      "loss": 0.6684,
      "step": 560
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002906328768766373,
      "loss": 0.2015,
      "step": 561
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002905906058132089,
      "loss": 0.642,
      "step": 562
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002905482426740948,
      "loss": 0.6202,
      "step": 563
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029050578748703943,
      "loss": 0.397,
      "step": 564
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002904632402798476,
      "loss": 0.6144,
      "step": 565
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029042060108038436,
      "loss": 0.2786,
      "step": 566
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.000290377869916575,
      "loss": 1.2726,
      "step": 567
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029033504681640505,
      "loss": 0.2469,
      "step": 568
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029029213180792015,
      "loss": 0.43,
      "step": 569
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029024912491922635,
      "loss": 0.4193,
      "step": 570
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029020602617848976,
      "loss": 0.5942,
      "step": 571
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002901628356139365,
      "loss": 0.6177,
      "step": 572
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029011955325385303,
      "loss": 0.323,
      "step": 573
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002900761791265859,
      "loss": 0.4635,
      "step": 574
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002900327132605417,
      "loss": 0.4616,
      "step": 575
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00028998915568418705,
      "loss": 0.6873,
      "step": 576
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00028994550642604883,
      "loss": 1.2433,
      "step": 577
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002899017655147137,
      "loss": 0.6319,
      "step": 578
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00028985793297882855,
      "loss": 0.3809,
      "step": 579
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00028981400884710024,
      "loss": 0.3195,
      "step": 580
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002897699931482956,
      "loss": 0.8321,
      "step": 581
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028972588591124135,
      "loss": 0.3303,
      "step": 582
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028968168716482425,
      "loss": 0.7224,
      "step": 583
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028963739693799097,
      "loss": 0.4347,
      "step": 584
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028959301525974815,
      "loss": 0.3358,
      "step": 585
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028954854215916217,
      "loss": 1.2214,
      "step": 586
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002895039776653595,
      "loss": 0.2874,
      "step": 587
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002894593218075262,
      "loss": 0.4277,
      "step": 588
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028941457461490843,
      "loss": 0.7291,
      "step": 589
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028936973611681197,
      "loss": 0.2189,
      "step": 590
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028932480634260253,
      "loss": 0.4165,
      "step": 591
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028927978532170554,
      "loss": 0.4825,
      "step": 592
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002892346730836062,
      "loss": 0.4849,
      "step": 593
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028918946965784937,
      "loss": 0.7935,
      "step": 594
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002891441750740398,
      "loss": 0.5248,
      "step": 595
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028909878936184184,
      "loss": 0.2086,
      "step": 596
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028905331255097945,
      "loss": 0.3782,
      "step": 597
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002890077446712364,
      "loss": 0.7334,
      "step": 598
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028896208575245595,
      "loss": 0.4023,
      "step": 599
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002889163358245412,
      "loss": 0.3906,
      "step": 600
    },
    {
      "epoch": 0.15,
      "eval_loss": 0.4527268409729004,
      "eval_runtime": 49.9253,
      "eval_samples_per_second": 0.661,
      "eval_steps_per_second": 0.661,
      "step": 600
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028887049491745467,
      "loss": 0.4575,
      "step": 601
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028882456306121844,
      "loss": 0.5165,
      "step": 602
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028877854028591427,
      "loss": 0.5145,
      "step": 603
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002887324266216835,
      "loss": 0.3195,
      "step": 604
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002886862220987268,
      "loss": 0.4103,
      "step": 605
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002886399267473046,
      "loss": 0.731,
      "step": 606
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002885935405977365,
      "loss": 0.3744,
      "step": 607
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002885470636804019,
      "loss": 0.4203,
      "step": 608
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028850049602573944,
      "loss": 0.4885,
      "step": 609
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028845383766424725,
      "loss": 0.285,
      "step": 610
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028840708862648285,
      "loss": 0.5204,
      "step": 611
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002883602489430631,
      "loss": 0.2521,
      "step": 612
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002883133186446643,
      "loss": 0.5715,
      "step": 613
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002882662977620221,
      "loss": 0.5915,
      "step": 614
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002882191863259314,
      "loss": 0.5118,
      "step": 615
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028817198436724656,
      "loss": 0.5718,
      "step": 616
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028812469191688095,
      "loss": 0.4001,
      "step": 617
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028807730900580746,
      "loss": 0.6099,
      "step": 618
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002880298356650582,
      "loss": 0.2241,
      "step": 619
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00028798227192572434,
      "loss": 0.4859,
      "step": 620
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002879346178189563,
      "loss": 0.5275,
      "step": 621
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002878868733759639,
      "loss": 0.2612,
      "step": 622
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00028783903862801585,
      "loss": 0.6347,
      "step": 623
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00028779111360644014,
      "loss": 0.6216,
      "step": 624
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00028774309834262384,
      "loss": 0.4927,
      "step": 625
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002876949928680132,
      "loss": 0.2508,
      "step": 626
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00028764679721411337,
      "loss": 0.3373,
      "step": 627
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002875985114124888,
      "loss": 0.612,
      "step": 628
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002875501354947628,
      "loss": 0.2452,
      "step": 629
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002875016694926177,
      "loss": 0.268,
      "step": 630
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.000287453113437795,
      "loss": 0.1654,
      "step": 631
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.000287404467362095,
      "loss": 0.3627,
      "step": 632
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.000287355731297377,
      "loss": 0.0965,
      "step": 633
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00028730690527555924,
      "loss": 0.7048,
      "step": 634
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.000287257989328619,
      "loss": 0.2507,
      "step": 635
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002872089834885922,
      "loss": 0.1501,
      "step": 636
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00028715988778757386,
      "loss": 0.8129,
      "step": 637
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002871107022577178,
      "loss": 0.6114,
      "step": 638
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002870614269312365,
      "loss": 0.274,
      "step": 639
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00028701206184040146,
      "loss": 0.2962,
      "step": 640
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.000286962607017543,
      "loss": 0.462,
      "step": 641
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002869130624950499,
      "loss": 0.363,
      "step": 642
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00028686342830537006,
      "loss": 0.3677,
      "step": 643
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00028681370448100984,
      "loss": 0.6258,
      "step": 644
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00028676389105453446,
      "loss": 0.6815,
      "step": 645
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002867139880585677,
      "loss": 0.6775,
      "step": 646
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002866639955257921,
      "loss": 0.4311,
      "step": 647
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002866139134889488,
      "loss": 0.5161,
      "step": 648
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002865637419808375,
      "loss": 0.4064,
      "step": 649
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00028651348103431667,
      "loss": 0.5398,
      "step": 650
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002864631306823031,
      "loss": 0.202,
      "step": 651
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00028641269095777244,
      "loss": 0.526,
      "step": 652
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00028636216189375857,
      "loss": 0.4555,
      "step": 653
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.000286311543523354,
      "loss": 0.2988,
      "step": 654
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002862608358797099,
      "loss": 0.6965,
      "step": 655
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002862100389960356,
      "loss": 0.3745,
      "step": 656
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.000286159152905599,
      "loss": 0.3784,
      "step": 657
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002861081776417266,
      "loss": 0.2835,
      "step": 658
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002860571132378031,
      "loss": 0.5179,
      "step": 659
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002860059597272716,
      "loss": 0.2156,
      "step": 660
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002859547171436336,
      "loss": 0.6458,
      "step": 661
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002859033855204489,
      "loss": 0.5924,
      "step": 662
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028585196489133567,
      "loss": 0.7933,
      "step": 663
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002858004552899703,
      "loss": 0.8402,
      "step": 664
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028574885675008754,
      "loss": 0.4346,
      "step": 665
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002856971693054803,
      "loss": 0.6804,
      "step": 666
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002856453929899998,
      "loss": 0.5563,
      "step": 667
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002855935278375554,
      "loss": 0.2278,
      "step": 668
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002855415738821146,
      "loss": 0.657,
      "step": 669
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002854895311577032,
      "loss": 0.5049,
      "step": 670
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.000285437399698405,
      "loss": 0.6456,
      "step": 671
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028538517953836197,
      "loss": 0.4297,
      "step": 672
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002853328707117742,
      "loss": 0.6972,
      "step": 673
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002852804732528997,
      "loss": 0.3064,
      "step": 674
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028522798719605487,
      "loss": 0.7358,
      "step": 675
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028517541257561373,
      "loss": 0.4911,
      "step": 676
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028512274942600857,
      "loss": 0.293,
      "step": 677
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028506999778172946,
      "loss": 0.3888,
      "step": 678
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002850171576773247,
      "loss": 0.2724,
      "step": 679
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028496422914740024,
      "loss": 0.1489,
      "step": 680
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002849112122266201,
      "loss": 0.8564,
      "step": 681
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028485810694970623,
      "loss": 0.4322,
      "step": 682
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002848049133514382,
      "loss": 0.3873,
      "step": 683
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002847516314666538,
      "loss": 0.3015,
      "step": 684
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002846982613302483,
      "loss": 0.677,
      "step": 685
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002846448029771749,
      "loss": 0.3877,
      "step": 686
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028459125644244464,
      "loss": 0.42,
      "step": 687
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028453762176112616,
      "loss": 0.5593,
      "step": 688
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028448389896834603,
      "loss": 0.4787,
      "step": 689
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028443008809928835,
      "loss": 0.2531,
      "step": 690
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.000284376189189195,
      "loss": 0.3165,
      "step": 691
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002843222022733654,
      "loss": 1.0479,
      "step": 692
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002842681273871569,
      "loss": 0.4833,
      "step": 693
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.000284213964565984,
      "loss": 0.371,
      "step": 694
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028415971384531915,
      "loss": 0.703,
      "step": 695
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028410537526069236,
      "loss": 0.5817,
      "step": 696
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.000284050948847691,
      "loss": 0.6836,
      "step": 697
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028399643464196,
      "loss": 0.5951,
      "step": 698
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028394183267920197,
      "loss": 0.706,
      "step": 699
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00028388714299517676,
      "loss": 0.5671,
      "step": 700
    },
    {
      "epoch": 0.17,
      "eval_loss": 0.44737541675567627,
      "eval_runtime": 49.9335,
      "eval_samples_per_second": 0.661,
      "eval_steps_per_second": 0.661,
      "step": 700
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002838323656257018,
      "loss": 0.2182,
      "step": 701
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002837775006066519,
      "loss": 0.6357,
      "step": 702
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002837225479739593,
      "loss": 0.231,
      "step": 703
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002836675077636137,
      "loss": 0.5741,
      "step": 704
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028361238001166193,
      "loss": 0.3911,
      "step": 705
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028355716475420836,
      "loss": 0.6464,
      "step": 706
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002835018620274146,
      "loss": 0.8379,
      "step": 707
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002834464718674995,
      "loss": 0.3157,
      "step": 708
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002833909943107393,
      "loss": 0.7584,
      "step": 709
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028333542939346735,
      "loss": 0.4937,
      "step": 710
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028327977715207424,
      "loss": 0.5259,
      "step": 711
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002832240376230078,
      "loss": 0.5897,
      "step": 712
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028316821084277297,
      "loss": 0.6554,
      "step": 713
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002831122968479319,
      "loss": 0.4274,
      "step": 714
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028305629567510377,
      "loss": 0.2948,
      "step": 715
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028300020736096493,
      "loss": 0.4879,
      "step": 716
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002829440319422488,
      "loss": 0.6712,
      "step": 717
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028288776945574576,
      "loss": 0.3336,
      "step": 718
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028283141993830336,
      "loss": 0.8737,
      "step": 719
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002827749834268259,
      "loss": 0.4714,
      "step": 720
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.000282718459958275,
      "loss": 0.4066,
      "step": 721
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002826618495696689,
      "loss": 0.2366,
      "step": 722
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.000282605152298083,
      "loss": 0.6039,
      "step": 723
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028254836818064947,
      "loss": 0.617,
      "step": 724
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028249149725455747,
      "loss": 0.3623,
      "step": 725
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002824345395570528,
      "loss": 0.1902,
      "step": 726
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028237749512543837,
      "loss": 0.3135,
      "step": 727
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028232036399707366,
      "loss": 0.6001,
      "step": 728
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002822631462093751,
      "loss": 0.7159,
      "step": 729
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002822058417998158,
      "loss": 0.4653,
      "step": 730
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028214845080592544,
      "loss": 0.7539,
      "step": 731
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028209097326529077,
      "loss": 0.5475,
      "step": 732
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002820334092155549,
      "loss": 0.7042,
      "step": 733
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028197575869441774,
      "loss": 0.799,
      "step": 734
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028191802173963584,
      "loss": 0.4746,
      "step": 735
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028186019838902227,
      "loss": 0.7063,
      "step": 736
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028180228868044673,
      "loss": 0.2255,
      "step": 737
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028174429265183555,
      "loss": 0.6474,
      "step": 738
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00028168621034117147,
      "loss": 0.2174,
      "step": 739
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002816280417864938,
      "loss": 0.2694,
      "step": 740
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002815697870258983,
      "loss": 0.3319,
      "step": 741
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00028151144609753724,
      "loss": 0.4895,
      "step": 742
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002814530190396193,
      "loss": 0.7168,
      "step": 743
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002813945058904096,
      "loss": 0.5407,
      "step": 744
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00028133590668822956,
      "loss": 0.2574,
      "step": 745
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002812772214714571,
      "loss": 0.383,
      "step": 746
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002812184502785262,
      "loss": 0.5274,
      "step": 747
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002811595931479274,
      "loss": 0.2782,
      "step": 748
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002811006501182075,
      "loss": 0.5169,
      "step": 749
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002810416212279694,
      "loss": 0.4632,
      "step": 750
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00028098250651587247,
      "loss": 0.6054,
      "step": 751
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00028092330602063215,
      "loss": 0.5388,
      "step": 752
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00028086401978101995,
      "loss": 0.4535,
      "step": 753
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002808046478358637,
      "loss": 0.2559,
      "step": 754
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002807451902240473,
      "loss": 1.0736,
      "step": 755
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00028068564698451086,
      "loss": 0.3034,
      "step": 756
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00028062601815625034,
      "loss": 0.3564,
      "step": 757
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.000280566303778318,
      "loss": 0.542,
      "step": 758
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002805065038898219,
      "loss": 0.2588,
      "step": 759
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00028044661852992634,
      "loss": 0.2294,
      "step": 760
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002803866477378514,
      "loss": 0.4883,
      "step": 761
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002803265915528732,
      "loss": 0.6093,
      "step": 762
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002802664500143238,
      "loss": 0.7072,
      "step": 763
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00028020622316159106,
      "loss": 0.413,
      "step": 764
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00028014591103411895,
      "loss": 0.6161,
      "step": 765
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00028008551367140696,
      "loss": 0.2802,
      "step": 766
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002800250311130106,
      "loss": 0.4173,
      "step": 767
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002799644633985412,
      "loss": 0.4289,
      "step": 768
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002799038105676658,
      "loss": 0.4351,
      "step": 769
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002798430726601071,
      "loss": 0.1989,
      "step": 770
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00027978224971564374,
      "loss": 0.728,
      "step": 771
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00027972134177410985,
      "loss": 0.4013,
      "step": 772
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00027966034887539525,
      "loss": 0.4097,
      "step": 773
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002795992710594455,
      "loss": 0.431,
      "step": 774
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00027953810836626173,
      "loss": 0.4483,
      "step": 775
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00027947686083590063,
      "loss": 0.2901,
      "step": 776
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00027941552850847454,
      "loss": 0.5839,
      "step": 777
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002793541114241511,
      "loss": 0.3329,
      "step": 778
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00027929260962315376,
      "loss": 0.443,
      "step": 779
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002792310231457613,
      "loss": 0.6173,
      "step": 780
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.000279169352032308,
      "loss": 0.1568,
      "step": 781
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027910759632318344,
      "loss": 0.5921,
      "step": 782
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027904575605883274,
      "loss": 0.3825,
      "step": 783
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002789838312797564,
      "loss": 0.8074,
      "step": 784
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002789218220265103,
      "loss": 0.5906,
      "step": 785
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027885972833970545,
      "loss": 0.5794,
      "step": 786
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002787975502600083,
      "loss": 0.2951,
      "step": 787
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002787352878281406,
      "loss": 0.7162,
      "step": 788
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027867294108487934,
      "loss": 0.6017,
      "step": 789
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027861051007105667,
      "loss": 0.3834,
      "step": 790
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002785479948275599,
      "loss": 0.5256,
      "step": 791
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002784853953953315,
      "loss": 0.4704,
      "step": 792
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027842271181536925,
      "loss": 0.532,
      "step": 793
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002783599441287259,
      "loss": 0.3942,
      "step": 794
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002782970923765092,
      "loss": 0.4559,
      "step": 795
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027823415659988213,
      "loss": 0.5131,
      "step": 796
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002781711368400627,
      "loss": 0.639,
      "step": 797
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002781080331383237,
      "loss": 0.5095,
      "step": 798
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027804484553599313,
      "loss": 0.595,
      "step": 799
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002779815740744539,
      "loss": 0.4934,
      "step": 800
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.4436054825782776,
      "eval_runtime": 53.2826,
      "eval_samples_per_second": 0.619,
      "eval_steps_per_second": 0.619,
      "step": 800
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027791821879514367,
      "loss": 0.5021,
      "step": 801
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002778547797395552,
      "loss": 0.5376,
      "step": 802
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027779125694923603,
      "loss": 0.4085,
      "step": 803
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027772765046578856,
      "loss": 0.4362,
      "step": 804
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027766396033086993,
      "loss": 0.5263,
      "step": 805
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027760018658619216,
      "loss": 0.3076,
      "step": 806
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.000277536329273522,
      "loss": 0.4043,
      "step": 807
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027747238843468087,
      "loss": 0.4446,
      "step": 808
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027740836411154504,
      "loss": 0.3942,
      "step": 809
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002773442563460453,
      "loss": 0.5563,
      "step": 810
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027728006518016706,
      "loss": 0.486,
      "step": 811
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002772157906559506,
      "loss": 0.4773,
      "step": 812
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027715143281549054,
      "loss": 0.2839,
      "step": 813
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002770869917009361,
      "loss": 0.4563,
      "step": 814
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027702246735449124,
      "loss": 0.4408,
      "step": 815
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027695785981841415,
      "loss": 0.34,
      "step": 816
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027689316913501774,
      "loss": 0.7834,
      "step": 817
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00027682839534666926,
      "loss": 0.4888,
      "step": 818
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002767635384957903,
      "loss": 0.5982,
      "step": 819
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.000276698598624857,
      "loss": 0.2622,
      "step": 820
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027663357577639995,
      "loss": 0.3839,
      "step": 821
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002765684699930037,
      "loss": 0.6698,
      "step": 822
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027650328131730755,
      "loss": 0.1602,
      "step": 823
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027643800979200493,
      "loss": 0.4332,
      "step": 824
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002763726554598434,
      "loss": 0.0873,
      "step": 825
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027630721836362487,
      "loss": 0.3851,
      "step": 826
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027624169854620557,
      "loss": 0.4645,
      "step": 827
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002761760960504956,
      "loss": 0.3136,
      "step": 828
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027611041091945947,
      "loss": 0.461,
      "step": 829
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027604464319611575,
      "loss": 0.3601,
      "step": 830
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.000275978792923537,
      "loss": 0.3854,
      "step": 831
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002759128601448501,
      "loss": 0.5236,
      "step": 832
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002758468449032356,
      "loss": 0.7572,
      "step": 833
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027578074724192827,
      "loss": 0.4725,
      "step": 834
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002757145672042169,
      "loss": 0.8225,
      "step": 835
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027564830483344416,
      "loss": 0.1801,
      "step": 836
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027558196017300654,
      "loss": 0.3628,
      "step": 837
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002755155332663546,
      "loss": 0.6348,
      "step": 838
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002754490241569927,
      "loss": 0.5353,
      "step": 839
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027538243288847904,
      "loss": 0.4078,
      "step": 840
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027531575950442554,
      "loss": 0.3949,
      "step": 841
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027524900404849806,
      "loss": 0.3628,
      "step": 842
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002751821665644161,
      "loss": 0.5819,
      "step": 843
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027511524709595284,
      "loss": 0.3141,
      "step": 844
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002750482456869353,
      "loss": 0.745,
      "step": 845
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027498116238124396,
      "loss": 0.2078,
      "step": 846
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002749139972228133,
      "loss": 0.3276,
      "step": 847
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027484675025563093,
      "loss": 0.1589,
      "step": 848
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002747794215237384,
      "loss": 0.5331,
      "step": 849
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002747120110712306,
      "loss": 0.6283,
      "step": 850
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027464451894225613,
      "loss": 0.1678,
      "step": 851
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027457694518101685,
      "loss": 0.622,
      "step": 852
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002745092898317684,
      "loss": 0.3826,
      "step": 853
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002744415529388194,
      "loss": 0.8568,
      "step": 854
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027437373454653233,
      "loss": 0.4216,
      "step": 855
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002743058346993228,
      "loss": 0.4463,
      "step": 856
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002742378534416599,
      "loss": 0.3034,
      "step": 857
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027416979081806576,
      "loss": 0.3104,
      "step": 858
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002741016468731162,
      "loss": 0.5303,
      "step": 859
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00027403342165144,
      "loss": 0.7047,
      "step": 860
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027396511519771924,
      "loss": 0.3747,
      "step": 861
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027389672755668925,
      "loss": 0.3578,
      "step": 862
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027382825877313855,
      "loss": 0.5814,
      "step": 863
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002737597088919087,
      "loss": 0.5836,
      "step": 864
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027369107795789443,
      "loss": 0.3993,
      "step": 865
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002736223660160436,
      "loss": 0.5046,
      "step": 866
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027355357311135704,
      "loss": 0.5678,
      "step": 867
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002734846992888887,
      "loss": 0.295,
      "step": 868
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027341574459374533,
      "loss": 0.3115,
      "step": 869
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027334670907108694,
      "loss": 0.2611,
      "step": 870
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002732775927661262,
      "loss": 0.5816,
      "step": 871
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027320839572412886,
      "loss": 0.413,
      "step": 872
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002731391179904135,
      "loss": 0.2203,
      "step": 873
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002730697596103515,
      "loss": 0.4208,
      "step": 874
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002730003206293671,
      "loss": 0.4631,
      "step": 875
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002729308010929373,
      "loss": 0.6029,
      "step": 876
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.000272861201046592,
      "loss": 0.1706,
      "step": 877
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002727915205359135,
      "loss": 0.3782,
      "step": 878
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027272175960653714,
      "loss": 0.24,
      "step": 879
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027265191830415077,
      "loss": 0.2265,
      "step": 880
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002725819966744949,
      "loss": 0.3668,
      "step": 881
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027251199476336263,
      "loss": 0.5207,
      "step": 882
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002724419126165997,
      "loss": 0.6072,
      "step": 883
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027237175028010425,
      "loss": 0.2702,
      "step": 884
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027230150779982716,
      "loss": 0.4032,
      "step": 885
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027223118522177157,
      "loss": 0.2309,
      "step": 886
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002721607825919933,
      "loss": 0.3178,
      "step": 887
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002720902999566004,
      "loss": 0.3818,
      "step": 888
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002720197373617534,
      "loss": 0.2788,
      "step": 889
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027194909485366524,
      "loss": 0.5562,
      "step": 890
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027187837247860116,
      "loss": 0.4351,
      "step": 891
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027180757028287864,
      "loss": 0.3358,
      "step": 892
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002717366883128675,
      "loss": 0.2901,
      "step": 893
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002716657266149899,
      "loss": 0.8234,
      "step": 894
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027159468523572006,
      "loss": 0.307,
      "step": 895
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002715235642215844,
      "loss": 0.1647,
      "step": 896
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00027145236361916157,
      "loss": 0.5726,
      "step": 897
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002713810834750824,
      "loss": 0.2315,
      "step": 898
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002713097238360296,
      "loss": 0.3334,
      "step": 899
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002712382847487381,
      "loss": 0.3679,
      "step": 900
    },
    {
      "epoch": 0.22,
      "eval_loss": 0.4337725043296814,
      "eval_runtime": 49.8218,
      "eval_samples_per_second": 0.662,
      "eval_steps_per_second": 0.662,
      "step": 900
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00027116676625999484,
      "loss": 0.2769,
      "step": 901
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00027109516841663877,
      "loss": 0.137,
      "step": 902
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00027102349126556074,
      "loss": 0.1429,
      "step": 903
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00027095173485370373,
      "loss": 0.2783,
      "step": 904
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002708798992280623,
      "loss": 0.2115,
      "step": 905
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002708079844356832,
      "loss": 0.5883,
      "step": 906
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00027073599052366496,
      "loss": 0.4809,
      "step": 907
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00027066391753915775,
      "loss": 0.6323,
      "step": 908
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00027059176552936373,
      "loss": 0.622,
      "step": 909
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00027051953454153673,
      "loss": 0.3471,
      "step": 910
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002704472246229823,
      "loss": 0.3914,
      "step": 911
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00027037483582105764,
      "loss": 0.6128,
      "step": 912
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00027030236818317176,
      "loss": 0.6604,
      "step": 913
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002702298217567852,
      "loss": 0.5338,
      "step": 914
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00027015719658941004,
      "loss": 0.3285,
      "step": 915
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00027008449272861005,
      "loss": 0.3008,
      "step": 916
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002700117102220006,
      "loss": 0.4096,
      "step": 917
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002699388491172482,
      "loss": 0.8348,
      "step": 918
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00026986590946207124,
      "loss": 0.5129,
      "step": 919
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002697928913042395,
      "loss": 0.4089,
      "step": 920
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00026971979469157386,
      "loss": 0.2807,
      "step": 921
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002696466196719469,
      "loss": 0.4886,
      "step": 922
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00026957336629328244,
      "loss": 0.4796,
      "step": 923
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00026950003460355574,
      "loss": 0.2952,
      "step": 924
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002694266246507931,
      "loss": 0.6004,
      "step": 925
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002693531364830722,
      "loss": 0.2121,
      "step": 926
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.000269279570148522,
      "loss": 0.6105,
      "step": 927
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00026920592569532265,
      "loss": 0.3267,
      "step": 928
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00026913220317170537,
      "loss": 0.5789,
      "step": 929
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00026905840262595257,
      "loss": 0.1605,
      "step": 930
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00026898452410639777,
      "loss": 0.3615,
      "step": 931
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002689105676614255,
      "loss": 0.5904,
      "step": 932
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002688365333394714,
      "loss": 0.5573,
      "step": 933
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.000268762421189022,
      "loss": 0.6503,
      "step": 934
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00026868823125861496,
      "loss": 0.4488,
      "step": 935
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002686139635968388,
      "loss": 0.7794,
      "step": 936
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00026853961825233283,
      "loss": 0.1449,
      "step": 937
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00026846519527378744,
      "loss": 0.4202,
      "step": 938
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002683906947099438,
      "loss": 0.4491,
      "step": 939
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002683161166095938,
      "loss": 0.519,
      "step": 940
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00026824146102158013,
      "loss": 0.8464,
      "step": 941
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00026816672799479636,
      "loss": 0.3559,
      "step": 942
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002680919175781867,
      "loss": 0.3006,
      "step": 943
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.000268017029820746,
      "loss": 0.6824,
      "step": 944
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002679420647715198,
      "loss": 0.6694,
      "step": 945
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00026786702247960425,
      "loss": 0.2853,
      "step": 946
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00026779190299414606,
      "loss": 0.491,
      "step": 947
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002677167063643426,
      "loss": 0.3094,
      "step": 948
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002676414326394417,
      "loss": 0.4684,
      "step": 949
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00026756608186874156,
      "loss": 0.7121,
      "step": 950
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00026749065410159104,
      "loss": 0.6264,
      "step": 951
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002674151493873893,
      "loss": 0.3331,
      "step": 952
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00026733956777558605,
      "loss": 0.6035,
      "step": 953
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002672639093156811,
      "loss": 0.5262,
      "step": 954
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002671881740572248,
      "loss": 0.2494,
      "step": 955
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002671123620498177,
      "loss": 0.4418,
      "step": 956
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002670364733431107,
      "loss": 0.5292,
      "step": 957
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00026696050798680486,
      "loss": 0.3212,
      "step": 958
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002668844660306514,
      "loss": 0.21,
      "step": 959
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002668083475244518,
      "loss": 0.7373,
      "step": 960
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002667321525180577,
      "loss": 0.6574,
      "step": 961
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002666558810613707,
      "loss": 0.7501,
      "step": 962
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002665795332043425,
      "loss": 0.5003,
      "step": 963
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002665031089969751,
      "loss": 0.3134,
      "step": 964
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002664266084893201,
      "loss": 0.6985,
      "step": 965
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00026635003173147926,
      "loss": 0.2412,
      "step": 966
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002662733787736044,
      "loss": 0.2633,
      "step": 967
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00026619664966589704,
      "loss": 0.9232,
      "step": 968
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002661198444586086,
      "loss": 0.2965,
      "step": 969
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002660429632020405,
      "loss": 0.4174,
      "step": 970
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00026596600594654384,
      "loss": 0.4158,
      "step": 971
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00026588897274251944,
      "loss": 0.3562,
      "step": 972
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.000265811863640418,
      "loss": 0.744,
      "step": 973
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00026573467869073984,
      "loss": 0.3454,
      "step": 974
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.000265657417944035,
      "loss": 0.6374,
      "step": 975
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002655800814509031,
      "loss": 0.147,
      "step": 976
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002655026692619934,
      "loss": 0.4846,
      "step": 977
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002654251814280048,
      "loss": 0.5487,
      "step": 978
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002653476179996856,
      "loss": 0.4397,
      "step": 979
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00026526997902783374,
      "loss": 0.744,
      "step": 980
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026519226456329654,
      "loss": 1.2305,
      "step": 981
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002651144746569709,
      "loss": 0.4283,
      "step": 982
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026503660935980287,
      "loss": 0.5078,
      "step": 983
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002649586687227882,
      "loss": 0.2373,
      "step": 984
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026488065279697175,
      "loss": 0.3141,
      "step": 985
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002648025616334477,
      "loss": 0.4184,
      "step": 986
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002647243952833597,
      "loss": 0.5842,
      "step": 987
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026464615379790036,
      "loss": 0.6459,
      "step": 988
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026456783722831175,
      "loss": 0.4927,
      "step": 989
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026448944562588496,
      "loss": 0.6331,
      "step": 990
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002644109790419602,
      "loss": 0.8719,
      "step": 991
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.000264332437527927,
      "loss": 0.2696,
      "step": 992
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002642538211352238,
      "loss": 0.3793,
      "step": 993
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026417512991533794,
      "loss": 0.2712,
      "step": 994
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.000264096363919806,
      "loss": 0.4139,
      "step": 995
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026401752320021353,
      "loss": 0.4446,
      "step": 996
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026393860780819485,
      "loss": 0.5126,
      "step": 997
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002638596177954333,
      "loss": 0.365,
      "step": 998
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002637805532136611,
      "loss": 0.4676,
      "step": 999
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026370141411465926,
      "loss": 0.2262,
      "step": 1000
    },
    {
      "epoch": 0.25,
      "eval_loss": 0.4300112724304199,
      "eval_runtime": 51.4642,
      "eval_samples_per_second": 0.641,
      "eval_steps_per_second": 0.641,
      "step": 1000
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002636222005502576,
      "loss": 0.1472,
      "step": 1001
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002635429125723347,
      "loss": 0.7052,
      "step": 1002
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026346355023281793,
      "loss": 0.636,
      "step": 1003
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002633841135836833,
      "loss": 0.3,
      "step": 1004
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002633046026769555,
      "loss": 0.3563,
      "step": 1005
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002632250175647079,
      "loss": 0.2216,
      "step": 1006
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026314535829906235,
      "loss": 0.6236,
      "step": 1007
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026306562493218943,
      "loss": 0.3658,
      "step": 1008
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002629858175163081,
      "loss": 0.2656,
      "step": 1009
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026290593610368596,
      "loss": 0.4687,
      "step": 1010
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026282598074663885,
      "loss": 0.4673,
      "step": 1011
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026274595149753137,
      "loss": 0.6829,
      "step": 1012
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002626658484087762,
      "loss": 0.2682,
      "step": 1013
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002625856715328345,
      "loss": 0.15,
      "step": 1014
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002625054209222159,
      "loss": 0.4633,
      "step": 1015
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.000262425096629478,
      "loss": 0.5618,
      "step": 1016
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026234469870722697,
      "loss": 0.2997,
      "step": 1017
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026226422720811697,
      "loss": 0.219,
      "step": 1018
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002621836821848504,
      "loss": 0.6435,
      "step": 1019
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026210306369017804,
      "loss": 0.6946,
      "step": 1020
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002620223717768985,
      "loss": 0.4267,
      "step": 1021
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002619416064978585,
      "loss": 0.4581,
      "step": 1022
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00026186076790595306,
      "loss": 0.453,
      "step": 1023
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00026177985605412485,
      "loss": 0.4297,
      "step": 1024
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002616988709953649,
      "loss": 0.4329,
      "step": 1025
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002616178127827118,
      "loss": 0.2961,
      "step": 1026
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00026153668146925243,
      "loss": 0.683,
      "step": 1027
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002614554771081212,
      "loss": 0.483,
      "step": 1028
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002613741997525006,
      "loss": 0.577,
      "step": 1029
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00026129284945562085,
      "loss": 0.4315,
      "step": 1030
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00026121142627075997,
      "loss": 0.4963,
      "step": 1031
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002611299302512436,
      "loss": 0.437,
      "step": 1032
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00026104836145044515,
      "loss": 0.4701,
      "step": 1033
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002609667199217858,
      "loss": 0.2408,
      "step": 1034
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002608850057187342,
      "loss": 0.3476,
      "step": 1035
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002608032188948068,
      "loss": 0.7385,
      "step": 1036
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00026072135950356727,
      "loss": 0.2979,
      "step": 1037
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00026063942759862714,
      "loss": 0.3588,
      "step": 1038
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00026055742323364525,
      "loss": 0.4494,
      "step": 1039
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.000260475346462328,
      "loss": 0.416,
      "step": 1040
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002603931973384291,
      "loss": 0.4363,
      "step": 1041
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002603109759157498,
      "loss": 0.3791,
      "step": 1042
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002602286822481385,
      "loss": 0.5441,
      "step": 1043
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002601463163894911,
      "loss": 0.4857,
      "step": 1044
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002600638783937506,
      "loss": 0.2416,
      "step": 1045
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00025998136831490744,
      "loss": 0.333,
      "step": 1046
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00025989878620699913,
      "loss": 0.5953,
      "step": 1047
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00025981613212411043,
      "loss": 0.3596,
      "step": 1048
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00025973340612037317,
      "loss": 0.5401,
      "step": 1049
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002596506082499663,
      "loss": 0.2429,
      "step": 1050
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002595677385671158,
      "loss": 0.2159,
      "step": 1051
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00025948479712609493,
      "loss": 0.5335,
      "step": 1052
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00025940178398122356,
      "loss": 0.4809,
      "step": 1053
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002593186991868688,
      "loss": 0.4059,
      "step": 1054
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002592355427974445,
      "loss": 0.2163,
      "step": 1055
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002591523148674116,
      "loss": 0.3328,
      "step": 1056
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002590690154512778,
      "loss": 0.3646,
      "step": 1057
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002589856446035975,
      "loss": 0.6167,
      "step": 1058
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.000258902202378972,
      "loss": 0.7292,
      "step": 1059
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002588186888320494,
      "loss": 0.3735,
      "step": 1060
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002587351040175245,
      "loss": 0.2662,
      "step": 1061
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002586514479901385,
      "loss": 0.3436,
      "step": 1062
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002585677208046797,
      "loss": 0.2121,
      "step": 1063
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002584839225159826,
      "loss": 0.361,
      "step": 1064
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025840005317892843,
      "loss": 0.2495,
      "step": 1065
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002583161128484451,
      "loss": 0.5275,
      "step": 1066
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002582321015795067,
      "loss": 0.5211,
      "step": 1067
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025814801942713393,
      "loss": 0.0687,
      "step": 1068
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002580638664463941,
      "loss": 0.5202,
      "step": 1069
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002579796426924006,
      "loss": 0.1317,
      "step": 1070
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002578953482203133,
      "loss": 0.3154,
      "step": 1071
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002578109830853384,
      "loss": 0.3186,
      "step": 1072
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025772654734272844,
      "loss": 0.945,
      "step": 1073
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.000257642041047782,
      "loss": 0.4347,
      "step": 1074
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025755746425584416,
      "loss": 0.5393,
      "step": 1075
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025747281702230584,
      "loss": 0.4397,
      "step": 1076
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025738809940260425,
      "loss": 0.0944,
      "step": 1077
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002573033114522229,
      "loss": 0.5852,
      "step": 1078
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.000257218453226691,
      "loss": 0.3277,
      "step": 1079
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025713352478158407,
      "loss": 0.3471,
      "step": 1080
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002570485261725233,
      "loss": 0.5725,
      "step": 1081
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025696345745517625,
      "loss": 0.3454,
      "step": 1082
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.000256878318685256,
      "loss": 0.4408,
      "step": 1083
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025679310991852194,
      "loss": 0.3971,
      "step": 1084
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025670783121077876,
      "loss": 0.9201,
      "step": 1085
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002566224826178773,
      "loss": 0.6996,
      "step": 1086
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025653706419571427,
      "loss": 0.4931,
      "step": 1087
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002564515760002319,
      "loss": 0.2824,
      "step": 1088
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.000256366018087418,
      "loss": 0.3835,
      "step": 1089
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025628039051330643,
      "loss": 0.5515,
      "step": 1090
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025619469333397634,
      "loss": 0.3696,
      "step": 1091
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002561089266055526,
      "loss": 0.2446,
      "step": 1092
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002560230903842056,
      "loss": 0.3802,
      "step": 1093
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025593718472615134,
      "loss": 0.7023,
      "step": 1094
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025585120968765106,
      "loss": 0.29,
      "step": 1095
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002557651653250117,
      "loss": 0.5026,
      "step": 1096
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002556790516945854,
      "loss": 0.5375,
      "step": 1097
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002555928688527698,
      "loss": 0.6592,
      "step": 1098
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025550661685600786,
      "loss": 0.5104,
      "step": 1099
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002554202957607877,
      "loss": 0.2757,
      "step": 1100
    },
    {
      "epoch": 0.27,
      "eval_loss": 0.4113215208053589,
      "eval_runtime": 49.836,
      "eval_samples_per_second": 0.662,
      "eval_steps_per_second": 0.662,
      "step": 1100
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002553339056236428,
      "loss": 0.4743,
      "step": 1101
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002552474465011519,
      "loss": 0.2896,
      "step": 1102
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025516091844993887,
      "loss": 0.617,
      "step": 1103
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025507432152667263,
      "loss": 0.3326,
      "step": 1104
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002549876557880674,
      "loss": 0.5212,
      "step": 1105
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002549009212908822,
      "loss": 0.5146,
      "step": 1106
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025481411809192135,
      "loss": 0.2869,
      "step": 1107
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002547272462480341,
      "loss": 0.208,
      "step": 1108
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002546403058161145,
      "loss": 0.3591,
      "step": 1109
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002545532968531017,
      "loss": 0.3905,
      "step": 1110
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025446621941597965,
      "loss": 0.3546,
      "step": 1111
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002543790735617771,
      "loss": 1.0187,
      "step": 1112
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002542918593475678,
      "loss": 1.0123,
      "step": 1113
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025420457683047,
      "loss": 0.3983,
      "step": 1114
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002541172260676469,
      "loss": 0.4907,
      "step": 1115
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002540298071163063,
      "loss": 0.2546,
      "step": 1116
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025394232003370066,
      "loss": 0.1426,
      "step": 1117
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025385476487712705,
      "loss": 0.3396,
      "step": 1118
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002537671417039272,
      "loss": 0.5162,
      "step": 1119
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002536794505714873,
      "loss": 0.4091,
      "step": 1120
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.000253591691537238,
      "loss": 0.5816,
      "step": 1121
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002535038646586547,
      "loss": 0.1453,
      "step": 1122
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002534159699932569,
      "loss": 0.4542,
      "step": 1123
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025332800759860863,
      "loss": 0.1872,
      "step": 1124
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002532399775323183,
      "loss": 0.217,
      "step": 1125
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025315187985203863,
      "loss": 0.4624,
      "step": 1126
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025306371461546656,
      "loss": 0.5145,
      "step": 1127
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025297548188034336,
      "loss": 0.6493,
      "step": 1128
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002528871817044544,
      "loss": 0.4606,
      "step": 1129
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002527988141456294,
      "loss": 0.4133,
      "step": 1130
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025271037926174203,
      "loss": 0.2933,
      "step": 1131
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025262187711071015,
      "loss": 0.3546,
      "step": 1132
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025253330775049554,
      "loss": 0.6914,
      "step": 1133
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025244467123910423,
      "loss": 0.5819,
      "step": 1134
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025235596763458596,
      "loss": 0.3443,
      "step": 1135
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025226719699503473,
      "loss": 0.1533,
      "step": 1136
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002521783593785881,
      "loss": 0.8362,
      "step": 1137
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025208945484342783,
      "loss": 0.5132,
      "step": 1138
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002520004834477791,
      "loss": 0.3976,
      "step": 1139
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002519114452499113,
      "loss": 0.353,
      "step": 1140
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002518223403081373,
      "loss": 0.383,
      "step": 1141
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00025173316868081375,
      "loss": 0.4937,
      "step": 1142
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002516439304263411,
      "loss": 0.913,
      "step": 1143
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002515546256031631,
      "loss": 0.4414,
      "step": 1144
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00025146525426976755,
      "loss": 0.5283,
      "step": 1145
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002513758164846855,
      "loss": 0.4903,
      "step": 1146
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002512863123064915,
      "loss": 0.4557,
      "step": 1147
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00025119674179380387,
      "loss": 0.8759,
      "step": 1148
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.000251107105005284,
      "loss": 0.1915,
      "step": 1149
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.000251017401999637,
      "loss": 0.7197,
      "step": 1150
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00025092763283561114,
      "loss": 0.6246,
      "step": 1151
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00025083779757199815,
      "loss": 0.2517,
      "step": 1152
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.000250747896267633,
      "loss": 0.3731,
      "step": 1153
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00025065792898139376,
      "loss": 0.5229,
      "step": 1154
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00025056789577220206,
      "loss": 0.317,
      "step": 1155
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00025047779669902244,
      "loss": 0.2828,
      "step": 1156
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002503876318208626,
      "loss": 0.6216,
      "step": 1157
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00025029740119677336,
      "loss": 0.1502,
      "step": 1158
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002502071048858487,
      "loss": 0.539,
      "step": 1159
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00025011674294722554,
      "loss": 0.0884,
      "step": 1160
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00025002631544008365,
      "loss": 0.9017,
      "step": 1161
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00024993582242364596,
      "loss": 0.212,
      "step": 1162
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002498452639571782,
      "loss": 0.2962,
      "step": 1163
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.000249754640099989,
      "loss": 0.6267,
      "step": 1164
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00024966395091142974,
      "loss": 0.1873,
      "step": 1165
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002495731964508946,
      "loss": 0.391,
      "step": 1166
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002494823767778208,
      "loss": 0.6439,
      "step": 1167
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002493914919516877,
      "loss": 0.5624,
      "step": 1168
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002493005420320178,
      "loss": 0.6951,
      "step": 1169
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00024920952707837597,
      "loss": 0.2214,
      "step": 1170
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00024911844715036995,
      "loss": 0.6301,
      "step": 1171
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00024902730230764975,
      "loss": 1.0625,
      "step": 1172
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.000248936092609908,
      "loss": 0.6766,
      "step": 1173
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00024884481811687987,
      "loss": 0.5935,
      "step": 1174
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002487534788883429,
      "loss": 0.21,
      "step": 1175
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.000248662074984117,
      "loss": 0.4316,
      "step": 1176
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00024857060646406445,
      "loss": 0.4068,
      "step": 1177
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00024847907338808995,
      "loss": 0.3283,
      "step": 1178
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00024838747581614026,
      "loss": 0.111,
      "step": 1179
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002482958138082047,
      "loss": 0.4139,
      "step": 1180
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024820408742431445,
      "loss": 0.5632,
      "step": 1181
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002481122967245431,
      "loss": 0.3268,
      "step": 1182
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002480204417690063,
      "loss": 0.2829,
      "step": 1183
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002479285226178616,
      "loss": 0.5764,
      "step": 1184
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002478365393313089,
      "loss": 0.5715,
      "step": 1185
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002477444919695899,
      "loss": 0.4258,
      "step": 1186
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002476523805929883,
      "loss": 0.1921,
      "step": 1187
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024756020526182974,
      "loss": 0.3412,
      "step": 1188
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002474679660364818,
      "loss": 0.2047,
      "step": 1189
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024737566297735377,
      "loss": 0.7168,
      "step": 1190
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002472832961448969,
      "loss": 0.3329,
      "step": 1191
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024719086559960403,
      "loss": 0.4247,
      "step": 1192
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024709837140200996,
      "loss": 0.3672,
      "step": 1193
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002470058136126909,
      "loss": 0.5531,
      "step": 1194
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.000246913192292265,
      "loss": 0.615,
      "step": 1195
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024682050750139174,
      "loss": 0.3769,
      "step": 1196
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002467277593007724,
      "loss": 0.33,
      "step": 1197
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002466349477511497,
      "loss": 0.5074,
      "step": 1198
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024654207291330776,
      "loss": 0.3861,
      "step": 1199
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024644913484807225,
      "loss": 0.437,
      "step": 1200
    },
    {
      "epoch": 0.3,
      "eval_loss": 0.4089837074279785,
      "eval_runtime": 53.3106,
      "eval_samples_per_second": 0.619,
      "eval_steps_per_second": 0.619,
      "step": 1200
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002463561336163102,
      "loss": 0.4545,
      "step": 1201
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002462630692789301,
      "loss": 0.1844,
      "step": 1202
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002461699418968817,
      "loss": 0.635,
      "step": 1203
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.000246076751531156,
      "loss": 0.316,
      "step": 1204
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024598349824278527,
      "loss": 0.4147,
      "step": 1205
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.000245890182092843,
      "loss": 0.4838,
      "step": 1206
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.000245796803142444,
      "loss": 0.352,
      "step": 1207
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024570336145274393,
      "loss": 0.3463,
      "step": 1208
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024560985708493974,
      "loss": 0.6046,
      "step": 1209
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002455162901002693,
      "loss": 0.3752,
      "step": 1210
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002454226605600116,
      "loss": 0.437,
      "step": 1211
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002453289685254866,
      "loss": 0.3678,
      "step": 1212
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024523521405805514,
      "loss": 0.5212,
      "step": 1213
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024514139721911886,
      "loss": 0.4304,
      "step": 1214
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002450475180701204,
      "loss": 0.5924,
      "step": 1215
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024495357667254317,
      "loss": 0.3587,
      "step": 1216
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002448595730879113,
      "loss": 0.6065,
      "step": 1217
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002447655073777897,
      "loss": 0.5318,
      "step": 1218
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024467137960378397,
      "loss": 0.2421,
      "step": 1219
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00024457718982754024,
      "loss": 0.6254,
      "step": 1220
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002444829381107453,
      "loss": 0.1927,
      "step": 1221
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002443886245151268,
      "loss": 0.2648,
      "step": 1222
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024429424910245236,
      "loss": 0.7091,
      "step": 1223
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002441998119345305,
      "loss": 0.3326,
      "step": 1224
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024410531307321006,
      "loss": 0.4482,
      "step": 1225
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024401075258038036,
      "loss": 1.0785,
      "step": 1226
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024391613051797097,
      "loss": 0.2355,
      "step": 1227
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024382144694795176,
      "loss": 0.3861,
      "step": 1228
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024372670193233308,
      "loss": 0.385,
      "step": 1229
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002436318955331653,
      "loss": 0.3923,
      "step": 1230
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024353702781253913,
      "loss": 0.6747,
      "step": 1231
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024344209883258534,
      "loss": 0.6449,
      "step": 1232
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024334710865547488,
      "loss": 0.8778,
      "step": 1233
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024325205734341887,
      "loss": 0.207,
      "step": 1234
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024315694495866827,
      "loss": 0.4497,
      "step": 1235
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002430617715635142,
      "loss": 0.3937,
      "step": 1236
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024296653722028766,
      "loss": 0.4238,
      "step": 1237
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024287124199135957,
      "loss": 0.5204,
      "step": 1238
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024277588593914077,
      "loss": 0.8021,
      "step": 1239
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024268046912608192,
      "loss": 0.2171,
      "step": 1240
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024258499161467338,
      "loss": 0.5846,
      "step": 1241
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024248945346744543,
      "loss": 0.4458,
      "step": 1242
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024239385474696787,
      "loss": 0.3084,
      "step": 1243
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024229819551585037,
      "loss": 0.3653,
      "step": 1244
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024220247583674206,
      "loss": 0.263,
      "step": 1245
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024210669577233183,
      "loss": 0.3514,
      "step": 1246
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024201085538534786,
      "loss": 0.3309,
      "step": 1247
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024191495473855815,
      "loss": 0.4376,
      "step": 1248
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024181899389476993,
      "loss": 0.4925,
      "step": 1249
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024172297291682999,
      "loss": 0.7311,
      "step": 1250
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024162689186762437,
      "loss": 0.2879,
      "step": 1251
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002415307508100786,
      "loss": 0.2659,
      "step": 1252
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024143454980715735,
      "loss": 0.5198,
      "step": 1253
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024133828892186474,
      "loss": 0.2921,
      "step": 1254
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024124196821724399,
      "loss": 0.4016,
      "step": 1255
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024114558775637743,
      "loss": 0.0565,
      "step": 1256
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002410491476023867,
      "loss": 0.4063,
      "step": 1257
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024095264781843248,
      "loss": 0.5591,
      "step": 1258
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024085608846771437,
      "loss": 0.3232,
      "step": 1259
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024075946961347108,
      "loss": 0.5452,
      "step": 1260
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00024066279131898033,
      "loss": 0.1777,
      "step": 1261
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00024056605364755875,
      "loss": 0.4445,
      "step": 1262
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002404692566625618,
      "loss": 0.5573,
      "step": 1263
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002403724004273838,
      "loss": 0.3589,
      "step": 1264
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.000240275485005458,
      "loss": 0.45,
      "step": 1265
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002401785104602561,
      "loss": 0.4352,
      "step": 1266
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.000240081476855289,
      "loss": 0.3541,
      "step": 1267
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023998438425410575,
      "loss": 0.3385,
      "step": 1268
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002398872327202945,
      "loss": 0.4435,
      "step": 1269
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002397900223174817,
      "loss": 0.5029,
      "step": 1270
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002396927531093324,
      "loss": 0.2222,
      "step": 1271
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023959542515955028,
      "loss": 0.4579,
      "step": 1272
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023949803853187739,
      "loss": 0.4173,
      "step": 1273
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023940059329009424,
      "loss": 0.6444,
      "step": 1274
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002393030894980198,
      "loss": 0.5842,
      "step": 1275
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023920552721951117,
      "loss": 0.6075,
      "step": 1276
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023910790651846404,
      "loss": 0.4788,
      "step": 1277
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023901022745881215,
      "loss": 0.2279,
      "step": 1278
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023891249010452755,
      "loss": 0.3379,
      "step": 1279
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023881469451962042,
      "loss": 0.2649,
      "step": 1280
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023871684076813912,
      "loss": 0.5249,
      "step": 1281
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023861892891417007,
      "loss": 0.5158,
      "step": 1282
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023852095902183784,
      "loss": 0.6219,
      "step": 1283
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023842293115530481,
      "loss": 0.4444,
      "step": 1284
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023832484537877153,
      "loss": 0.5689,
      "step": 1285
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023822670175647635,
      "loss": 0.4203,
      "step": 1286
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023812850035269566,
      "loss": 0.3482,
      "step": 1287
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023803024123174345,
      "loss": 0.3649,
      "step": 1288
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023793192445797174,
      "loss": 0.2666,
      "step": 1289
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002378335500957701,
      "loss": 0.2582,
      "step": 1290
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002377351182095661,
      "loss": 0.5452,
      "step": 1291
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023763662886382462,
      "loss": 0.2781,
      "step": 1292
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002375380821230485,
      "loss": 0.5404,
      "step": 1293
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023743947805177792,
      "loss": 0.5274,
      "step": 1294
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023734081671459084,
      "loss": 0.3333,
      "step": 1295
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023724209817610253,
      "loss": 0.3455,
      "step": 1296
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023714332250096578,
      "loss": 0.7642,
      "step": 1297
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023704448975387092,
      "loss": 0.3727,
      "step": 1298
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023694559999954545,
      "loss": 0.8526,
      "step": 1299
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023684665330275434,
      "loss": 0.3472,
      "step": 1300
    },
    {
      "epoch": 0.32,
      "eval_loss": 0.40911176800727844,
      "eval_runtime": 49.9255,
      "eval_samples_per_second": 0.661,
      "eval_steps_per_second": 0.661,
      "step": 1300
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00023674764972829982,
      "loss": 0.2547,
      "step": 1301
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023664858934102133,
      "loss": 0.5989,
      "step": 1302
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023654947220579567,
      "loss": 1.0535,
      "step": 1303
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002364502983875366,
      "loss": 0.8391,
      "step": 1304
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023635106795119511,
      "loss": 0.562,
      "step": 1305
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023625178096175928,
      "loss": 0.8478,
      "step": 1306
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002361524374842542,
      "loss": 0.4087,
      "step": 1307
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023605303758374196,
      "loss": 0.3507,
      "step": 1308
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002359535813253216,
      "loss": 0.3376,
      "step": 1309
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023585406877412907,
      "loss": 0.7603,
      "step": 1310
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023575449999533723,
      "loss": 0.1823,
      "step": 1311
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023565487505415567,
      "loss": 0.5729,
      "step": 1312
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023555519401583086,
      "loss": 0.2744,
      "step": 1313
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023545545694564593,
      "loss": 0.539,
      "step": 1314
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023535566390892079,
      "loss": 0.3945,
      "step": 1315
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002352558149710119,
      "loss": 0.5935,
      "step": 1316
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023515591019731237,
      "loss": 0.3972,
      "step": 1317
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.000235055949653252,
      "loss": 0.1965,
      "step": 1318
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023495593340429696,
      "loss": 0.2998,
      "step": 1319
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002348558615159499,
      "loss": 0.2971,
      "step": 1320
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023475573405375006,
      "loss": 0.7607,
      "step": 1321
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002346555510832729,
      "loss": 0.1805,
      "step": 1322
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023455531267013035,
      "loss": 0.2664,
      "step": 1323
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023445501887997062,
      "loss": 0.3286,
      "step": 1324
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002343546697784781,
      "loss": 0.4822,
      "step": 1325
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002342542654313736,
      "loss": 0.536,
      "step": 1326
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023415380590441391,
      "loss": 0.1786,
      "step": 1327
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023405329126339214,
      "loss": 0.6645,
      "step": 1328
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023395272157413727,
      "loss": 0.3816,
      "step": 1329
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023385209690251457,
      "loss": 0.5563,
      "step": 1330
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023375141731442518,
      "loss": 0.4219,
      "step": 1331
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023365068287580618,
      "loss": 0.5712,
      "step": 1332
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023354989365263073,
      "loss": 0.1375,
      "step": 1333
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002334490497109077,
      "loss": 0.2204,
      "step": 1334
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023334815111668191,
      "loss": 0.3932,
      "step": 1335
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023324719793603387,
      "loss": 0.2958,
      "step": 1336
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023314619023508,
      "loss": 0.2403,
      "step": 1337
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002330451280799722,
      "loss": 0.1713,
      "step": 1338
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023294401153689824,
      "loss": 0.5219,
      "step": 1339
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002328428406720814,
      "loss": 0.2712,
      "step": 1340
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00023274161555178057,
      "loss": 0.4026,
      "step": 1341
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023264033624229022,
      "loss": 0.7622,
      "step": 1342
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023253900280994016,
      "loss": 0.3258,
      "step": 1343
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023243761532109578,
      "loss": 0.6604,
      "step": 1344
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023233617384215788,
      "loss": 0.6454,
      "step": 1345
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023223467843956253,
      "loss": 0.4883,
      "step": 1346
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023213312917978117,
      "loss": 0.288,
      "step": 1347
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023203152612932053,
      "loss": 0.6047,
      "step": 1348
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023192986935472248,
      "loss": 0.4079,
      "step": 1349
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023182815892256419,
      "loss": 0.226,
      "step": 1350
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023172639489945794,
      "loss": 0.6881,
      "step": 1351
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.000231624577352051,
      "loss": 0.5078,
      "step": 1352
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023152270634702588,
      "loss": 0.1992,
      "step": 1353
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023142078195109993,
      "loss": 0.5064,
      "step": 1354
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023131880423102558,
      "loss": 0.23,
      "step": 1355
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023121677325359008,
      "loss": 0.7694,
      "step": 1356
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002311146890856157,
      "loss": 0.806,
      "step": 1357
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023101255179395943,
      "loss": 0.2244,
      "step": 1358
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023091036144551308,
      "loss": 0.4989,
      "step": 1359
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002308081181072032,
      "loss": 0.4586,
      "step": 1360
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002307058218459911,
      "loss": 0.5662,
      "step": 1361
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002306034727288727,
      "loss": 0.4541,
      "step": 1362
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023050107082287855,
      "loss": 0.3654,
      "step": 1363
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023039861619507374,
      "loss": 0.6479,
      "step": 1364
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023029610891255793,
      "loss": 0.4588,
      "step": 1365
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023019354904246526,
      "loss": 0.4785,
      "step": 1366
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00023009093665196434,
      "loss": 0.4554,
      "step": 1367
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00022998827180825804,
      "loss": 0.421,
      "step": 1368
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002298855545785838,
      "loss": 0.2856,
      "step": 1369
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00022978278503021313,
      "loss": 0.5274,
      "step": 1370
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00022967996323045202,
      "loss": 0.4052,
      "step": 1371
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00022957708924664056,
      "loss": 0.4023,
      "step": 1372
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00022947416314615305,
      "loss": 0.6348,
      "step": 1373
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00022937118499639786,
      "loss": 0.7367,
      "step": 1374
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002292681548648176,
      "loss": 0.4,
      "step": 1375
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00022916507281888873,
      "loss": 0.4822,
      "step": 1376
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00022906193892612185,
      "loss": 0.4004,
      "step": 1377
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00022895875325406145,
      "loss": 0.2319,
      "step": 1378
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.000228855515870286,
      "loss": 0.1947,
      "step": 1379
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002287522268424078,
      "loss": 0.3622,
      "step": 1380
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002286488862380729,
      "loss": 0.3479,
      "step": 1381
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022854549412496125,
      "loss": 0.321,
      "step": 1382
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022844205057078642,
      "loss": 0.6823,
      "step": 1383
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022833855564329575,
      "loss": 0.337,
      "step": 1384
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022823500941027022,
      "loss": 0.452,
      "step": 1385
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022813141193952434,
      "loss": 0.4786,
      "step": 1386
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022802776329890632,
      "loss": 0.6725,
      "step": 1387
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002279240635562977,
      "loss": 0.1967,
      "step": 1388
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022782031277961362,
      "loss": 0.3469,
      "step": 1389
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002277165110368026,
      "loss": 0.2622,
      "step": 1390
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022761265839584655,
      "loss": 0.769,
      "step": 1391
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022750875492476066,
      "loss": 0.6687,
      "step": 1392
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022740480069159355,
      "loss": 0.33,
      "step": 1393
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022730079576442687,
      "loss": 0.4473,
      "step": 1394
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002271967402113757,
      "loss": 0.3931,
      "step": 1395
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022709263410058806,
      "loss": 0.1735,
      "step": 1396
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022698847750024528,
      "loss": 0.5348,
      "step": 1397
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022688427047856162,
      "loss": 0.4162,
      "step": 1398
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022678001310378444,
      "loss": 0.4629,
      "step": 1399
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.000226675705444194,
      "loss": 0.4016,
      "step": 1400
    },
    {
      "epoch": 0.35,
      "eval_loss": 0.4050147235393524,
      "eval_runtime": 52.9067,
      "eval_samples_per_second": 0.624,
      "eval_steps_per_second": 0.624,
      "step": 1400
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022657134756810354,
      "loss": 0.2755,
      "step": 1401
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022646693954385923,
      "loss": 0.431,
      "step": 1402
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022636248143984,
      "loss": 0.6001,
      "step": 1403
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022625797332445762,
      "loss": 0.4453,
      "step": 1404
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002261534152661566,
      "loss": 0.3028,
      "step": 1405
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002260488073334142,
      "loss": 0.3447,
      "step": 1406
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002259441495947403,
      "loss": 0.3757,
      "step": 1407
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022583944211867734,
      "loss": 0.4806,
      "step": 1408
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022573468497380053,
      "loss": 0.2238,
      "step": 1409
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022562987822871737,
      "loss": 0.762,
      "step": 1410
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022552502195206806,
      "loss": 0.4641,
      "step": 1411
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022542011621252508,
      "loss": 0.2538,
      "step": 1412
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022531516107879334,
      "loss": 0.2885,
      "step": 1413
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022521015661961015,
      "loss": 0.3828,
      "step": 1414
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022510510290374506,
      "loss": 0.3871,
      "step": 1415
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.000225,
      "loss": 0.4624,
      "step": 1416
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022489484797720894,
      "loss": 0.5169,
      "step": 1417
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002247896469042381,
      "loss": 0.2059,
      "step": 1418
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022468439684998596,
      "loss": 0.2733,
      "step": 1419
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022457909788338277,
      "loss": 0.2675,
      "step": 1420
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00022447375007339116,
      "loss": 0.4849,
      "step": 1421
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022436835348900547,
      "loss": 0.4371,
      "step": 1422
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022426290819925213,
      "loss": 0.4432,
      "step": 1423
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022415741427318949,
      "loss": 0.538,
      "step": 1424
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022405187177990758,
      "loss": 0.8861,
      "step": 1425
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022394628078852843,
      "loss": 0.5886,
      "step": 1426
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022384064136820575,
      "loss": 0.6029,
      "step": 1427
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022373495358812498,
      "loss": 0.3104,
      "step": 1428
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002236292175175032,
      "loss": 0.4262,
      "step": 1429
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022352343322558916,
      "loss": 0.4119,
      "step": 1430
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002234176007816632,
      "loss": 0.2582,
      "step": 1431
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022331172025503708,
      "loss": 0.1369,
      "step": 1432
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022320579171505425,
      "loss": 0.6153,
      "step": 1433
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002230998152310894,
      "loss": 0.4798,
      "step": 1434
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022299379087254882,
      "loss": 0.6207,
      "step": 1435
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002228877187088699,
      "loss": 0.4888,
      "step": 1436
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002227815988095216,
      "loss": 0.3294,
      "step": 1437
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022267543124400405,
      "loss": 0.4108,
      "step": 1438
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022256921608184845,
      "loss": 0.3957,
      "step": 1439
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002224629533926174,
      "loss": 0.7353,
      "step": 1440
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002223566432459045,
      "loss": 0.393,
      "step": 1441
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022225028571133448,
      "loss": 0.7343,
      "step": 1442
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022214388085856307,
      "loss": 0.578,
      "step": 1443
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022203742875727696,
      "loss": 0.4176,
      "step": 1444
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022193092947719387,
      "loss": 0.3081,
      "step": 1445
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002218243830880624,
      "loss": 0.6192,
      "step": 1446
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022171778965966187,
      "loss": 0.7047,
      "step": 1447
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002216111492618026,
      "loss": 0.3958,
      "step": 1448
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022150446196432554,
      "loss": 0.4342,
      "step": 1449
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022139772783710249,
      "loss": 0.2081,
      "step": 1450
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002212909469500357,
      "loss": 0.3831,
      "step": 1451
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002211841193730583,
      "loss": 0.2974,
      "step": 1452
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002210772451761338,
      "loss": 0.351,
      "step": 1453
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002209703244292563,
      "loss": 0.3626,
      "step": 1454
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022086335720245041,
      "loss": 0.1581,
      "step": 1455
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022075634356577122,
      "loss": 0.3201,
      "step": 1456
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022064928358930412,
      "loss": 0.1673,
      "step": 1457
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022054217734316495,
      "loss": 0.2763,
      "step": 1458
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022043502489749967,
      "loss": 0.6419,
      "step": 1459
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022032782632248478,
      "loss": 0.354,
      "step": 1460
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022022058168832677,
      "loss": 0.2442,
      "step": 1461
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022011329106526236,
      "loss": 0.54,
      "step": 1462
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002200059545235584,
      "loss": 0.3079,
      "step": 1463
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021989857213351182,
      "loss": 0.2926,
      "step": 1464
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002197911439654495,
      "loss": 0.4951,
      "step": 1465
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021968367008972842,
      "loss": 0.3194,
      "step": 1466
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021957615057673538,
      "loss": 0.1976,
      "step": 1467
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021946858549688717,
      "loss": 0.1925,
      "step": 1468
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021936097492063035,
      "loss": 0.219,
      "step": 1469
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021925331891844136,
      "loss": 0.3719,
      "step": 1470
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002191456175608263,
      "loss": 0.4966,
      "step": 1471
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021903787091832102,
      "loss": 0.5521,
      "step": 1472
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021893007906149096,
      "loss": 0.6047,
      "step": 1473
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002188222420609313,
      "loss": 0.505,
      "step": 1474
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002187143599872667,
      "loss": 0.7962,
      "step": 1475
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021860643291115133,
      "loss": 0.3539,
      "step": 1476
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021849846090326886,
      "loss": 0.4019,
      "step": 1477
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021839044403433242,
      "loss": 0.3033,
      "step": 1478
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002182823823750844,
      "loss": 0.1361,
      "step": 1479
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021817427599629671,
      "loss": 0.2725,
      "step": 1480
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021806612496877038,
      "loss": 0.3234,
      "step": 1481
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002179579293633358,
      "loss": 0.6148,
      "step": 1482
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021784968925085246,
      "loss": 0.2003,
      "step": 1483
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021774140470220903,
      "loss": 0.362,
      "step": 1484
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021763307578832333,
      "loss": 0.3469,
      "step": 1485
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021752470258014218,
      "loss": 0.5133,
      "step": 1486
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021741628514864141,
      "loss": 0.4725,
      "step": 1487
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021730782356482587,
      "loss": 0.3447,
      "step": 1488
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021719931789972925,
      "loss": 0.5413,
      "step": 1489
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021709076822441417,
      "loss": 0.4936,
      "step": 1490
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021698217460997205,
      "loss": 0.6692,
      "step": 1491
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002168735371275231,
      "loss": 0.4624,
      "step": 1492
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002167648558482162,
      "loss": 0.3032,
      "step": 1493
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.000216656130843229,
      "loss": 0.439,
      "step": 1494
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021654736218376774,
      "loss": 0.3881,
      "step": 1495
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021643854994106727,
      "loss": 0.246,
      "step": 1496
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002163296941863909,
      "loss": 0.3874,
      "step": 1497
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021622079499103062,
      "loss": 0.343,
      "step": 1498
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021611185242630663,
      "loss": 0.4584,
      "step": 1499
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00021600286656356778,
      "loss": 0.5617,
      "step": 1500
    },
    {
      "epoch": 0.37,
      "eval_loss": 0.4020390510559082,
      "eval_runtime": 53.3265,
      "eval_samples_per_second": 0.619,
      "eval_steps_per_second": 0.619,
      "step": 1500
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002158938374741911,
      "loss": 0.2391,
      "step": 1501
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.000215784765229582,
      "loss": 0.4458,
      "step": 1502
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002156756499011741,
      "loss": 0.8816,
      "step": 1503
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021556649156042937,
      "loss": 0.5512,
      "step": 1504
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021545729027883779,
      "loss": 0.5807,
      "step": 1505
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002153480461279175,
      "loss": 0.2727,
      "step": 1506
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002152387591792148,
      "loss": 0.4453,
      "step": 1507
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.000215129429504304,
      "loss": 0.1281,
      "step": 1508
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021502005717478724,
      "loss": 0.3106,
      "step": 1509
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002149106422622948,
      "loss": 0.6643,
      "step": 1510
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021480118483848477,
      "loss": 0.6257,
      "step": 1511
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021469168497504303,
      "loss": 0.3666,
      "step": 1512
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021458214274368336,
      "loss": 0.4216,
      "step": 1513
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021447255821614714,
      "loss": 0.3219,
      "step": 1514
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002143629314642036,
      "loss": 0.4592,
      "step": 1515
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002142532625596495,
      "loss": 0.4418,
      "step": 1516
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021414355157430936,
      "loss": 0.6511,
      "step": 1517
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021403379858003508,
      "loss": 0.7556,
      "step": 1518
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021392400364870617,
      "loss": 0.3621,
      "step": 1519
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021381416685222965,
      "loss": 0.5222,
      "step": 1520
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021370428826253981,
      "loss": 0.2877,
      "step": 1521
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021359436795159853,
      "loss": 0.5898,
      "step": 1522
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021348440599139475,
      "loss": 0.2284,
      "step": 1523
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021337440245394493,
      "loss": 0.1306,
      "step": 1524
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002132643574112926,
      "loss": 0.4354,
      "step": 1525
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002131542709355085,
      "loss": 0.5278,
      "step": 1526
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021304414309869063,
      "loss": 0.2716,
      "step": 1527
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002129339739729639,
      "loss": 0.3736,
      "step": 1528
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021282376363048036,
      "loss": 0.4213,
      "step": 1529
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.000212713512143419,
      "loss": 0.2602,
      "step": 1530
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002126032195839859,
      "loss": 0.3499,
      "step": 1531
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002124928860244138,
      "loss": 0.4415,
      "step": 1532
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002123825115369625,
      "loss": 0.1903,
      "step": 1533
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021227209619391856,
      "loss": 0.3872,
      "step": 1534
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021216164006759516,
      "loss": 0.3448,
      "step": 1535
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021205114323033237,
      "loss": 0.3619,
      "step": 1536
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002119406057544968,
      "loss": 0.4945,
      "step": 1537
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021183002771248181,
      "loss": 0.2847,
      "step": 1538
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021171940917670713,
      "loss": 0.4929,
      "step": 1539
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021160875021961913,
      "loss": 0.2173,
      "step": 1540
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00021149805091369073,
      "loss": 0.3245,
      "step": 1541
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002113873113314211,
      "loss": 0.2138,
      "step": 1542
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00021127653154533595,
      "loss": 0.6255,
      "step": 1543
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00021116571162798717,
      "loss": 0.6234,
      "step": 1544
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00021105485165195305,
      "loss": 0.4338,
      "step": 1545
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00021094395168983805,
      "loss": 0.2999,
      "step": 1546
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002108330118142729,
      "loss": 0.2199,
      "step": 1547
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00021072203209791433,
      "loss": 0.5002,
      "step": 1548
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002106110126134453,
      "loss": 0.4593,
      "step": 1549
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002104999534335747,
      "loss": 0.285,
      "step": 1550
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00021038885463103756,
      "loss": 0.0914,
      "step": 1551
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00021027771627859468,
      "loss": 0.3999,
      "step": 1552
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.000210166538449033,
      "loss": 0.4933,
      "step": 1553
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.000210055321215165,
      "loss": 0.9619,
      "step": 1554
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002099440646498293,
      "loss": 0.1175,
      "step": 1555
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020983276882588996,
      "loss": 0.4933,
      "step": 1556
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020972143381623704,
      "loss": 0.4962,
      "step": 1557
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020961005969378607,
      "loss": 0.7436,
      "step": 1558
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020949864653147828,
      "loss": 0.2009,
      "step": 1559
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002093871944022804,
      "loss": 0.3327,
      "step": 1560
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002092757033791848,
      "loss": 0.4605,
      "step": 1561
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002091641735352092,
      "loss": 0.2949,
      "step": 1562
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002090526049433968,
      "loss": 0.402,
      "step": 1563
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020894099767681618,
      "loss": 0.4195,
      "step": 1564
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020882935180856124,
      "loss": 0.3608,
      "step": 1565
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002087176674117512,
      "loss": 0.1956,
      "step": 1566
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020860594455953034,
      "loss": 0.6551,
      "step": 1567
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002084941833250684,
      "loss": 0.4273,
      "step": 1568
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002083823837815601,
      "loss": 0.2955,
      "step": 1569
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020827054600222519,
      "loss": 0.6654,
      "step": 1570
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020815867006030852,
      "loss": 0.0887,
      "step": 1571
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020804675602908009,
      "loss": 0.2585,
      "step": 1572
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020793480398183462,
      "loss": 0.661,
      "step": 1573
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020782281399189182,
      "loss": 0.39,
      "step": 1574
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002077107861325962,
      "loss": 0.7523,
      "step": 1575
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020759872047731718,
      "loss": 0.3454,
      "step": 1576
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020748661709944885,
      "loss": 0.4313,
      "step": 1577
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020737447607241,
      "loss": 0.26,
      "step": 1578
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020726229746964413,
      "loss": 0.2341,
      "step": 1579
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020715008136461935,
      "loss": 0.7534,
      "step": 1580
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00020703782783082817,
      "loss": 0.2638,
      "step": 1581
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020692553694178788,
      "loss": 0.2968,
      "step": 1582
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020681320877104003,
      "loss": 0.2589,
      "step": 1583
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020670084339215064,
      "loss": 0.4865,
      "step": 1584
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002065884408787102,
      "loss": 0.2437,
      "step": 1585
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020647600130433333,
      "loss": 0.6637,
      "step": 1586
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020636352474265912,
      "loss": 0.5604,
      "step": 1587
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002062510112673507,
      "loss": 0.6519,
      "step": 1588
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020613846095209553,
      "loss": 0.3455,
      "step": 1589
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002060258738706051,
      "loss": 0.2965,
      "step": 1590
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.000205913250096615,
      "loss": 0.5008,
      "step": 1591
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020580058970388486,
      "loss": 0.2106,
      "step": 1592
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020568789276619831,
      "loss": 0.3357,
      "step": 1593
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020557515935736297,
      "loss": 0.0343,
      "step": 1594
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020546238955121014,
      "loss": 0.1104,
      "step": 1595
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020534958342159518,
      "loss": 0.5959,
      "step": 1596
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020523674104239712,
      "loss": 0.3605,
      "step": 1597
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002051238624875188,
      "loss": 0.4076,
      "step": 1598
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020501094783088663,
      "loss": 0.4256,
      "step": 1599
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020489799714645078,
      "loss": 0.1302,
      "step": 1600
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.38764214515686035,
      "eval_runtime": 52.8654,
      "eval_samples_per_second": 0.624,
      "eval_steps_per_second": 0.624,
      "step": 1600
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020478501050818498,
      "loss": 0.2686,
      "step": 1601
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020467198799008655,
      "loss": 0.1855,
      "step": 1602
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020455892966617618,
      "loss": 0.2337,
      "step": 1603
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002044458356104982,
      "loss": 0.5191,
      "step": 1604
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020433270589712024,
      "loss": 0.358,
      "step": 1605
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020421954060013319,
      "loss": 0.3252,
      "step": 1606
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020410633979365137,
      "loss": 0.5254,
      "step": 1607
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020399310355181233,
      "loss": 0.3897,
      "step": 1608
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020387983194877684,
      "loss": 0.7162,
      "step": 1609
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020376652505872875,
      "loss": 0.2626,
      "step": 1610
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020365318295587515,
      "loss": 0.5961,
      "step": 1611
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020353980571444604,
      "loss": 0.3344,
      "step": 1612
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020342639340869458,
      "loss": 0.1947,
      "step": 1613
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020331294611289686,
      "loss": 0.1346,
      "step": 1614
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002031994639013517,
      "loss": 0.3964,
      "step": 1615
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020308594684838112,
      "loss": 0.6373,
      "step": 1616
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020297239502832968,
      "loss": 0.3245,
      "step": 1617
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020285880851556482,
      "loss": 0.271,
      "step": 1618
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020274518738447667,
      "loss": 0.3797,
      "step": 1619
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020263153170947806,
      "loss": 0.501,
      "step": 1620
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00020251784156500444,
      "loss": 0.408,
      "step": 1621
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020240411702551384,
      "loss": 0.3329,
      "step": 1622
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020229035816548681,
      "loss": 0.5074,
      "step": 1623
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020217656505942634,
      "loss": 0.3817,
      "step": 1624
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020206273778185794,
      "loss": 0.4937,
      "step": 1625
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020194887640732938,
      "loss": 0.3041,
      "step": 1626
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020183498101041086,
      "loss": 0.3084,
      "step": 1627
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002017210516656948,
      "loss": 0.3087,
      "step": 1628
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002016070884477959,
      "loss": 0.4219,
      "step": 1629
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020149309143135095,
      "loss": 0.1783,
      "step": 1630
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020137906069101906,
      "loss": 0.4414,
      "step": 1631
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020126499630148122,
      "loss": 0.247,
      "step": 1632
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020115089833744055,
      "loss": 0.5667,
      "step": 1633
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020103676687362228,
      "loss": 0.6353,
      "step": 1634
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002009226019847733,
      "loss": 0.4863,
      "step": 1635
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020080840374566263,
      "loss": 0.2903,
      "step": 1636
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020069417223108105,
      "loss": 0.1929,
      "step": 1637
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002005799075158411,
      "loss": 0.2697,
      "step": 1638
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020046560967477715,
      "loss": 0.6273,
      "step": 1639
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020035127878274516,
      "loss": 0.4146,
      "step": 1640
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002002369149146228,
      "loss": 0.2991,
      "step": 1641
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00020012251814530933,
      "loss": 0.5959,
      "step": 1642
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002000080885497256,
      "loss": 0.2074,
      "step": 1643
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019989362620281384,
      "loss": 0.4515,
      "step": 1644
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019977913117953783,
      "loss": 0.4927,
      "step": 1645
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001996646035548827,
      "loss": 0.8928,
      "step": 1646
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.000199550043403855,
      "loss": 0.1985,
      "step": 1647
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019943545080148252,
      "loss": 0.3373,
      "step": 1648
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019932082582281425,
      "loss": 0.3768,
      "step": 1649
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019920616854292056,
      "loss": 0.4755,
      "step": 1650
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019909147903689274,
      "loss": 0.1804,
      "step": 1651
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001989767573798434,
      "loss": 0.5033,
      "step": 1652
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019886200364690616,
      "loss": 0.4345,
      "step": 1653
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001987472179132355,
      "loss": 0.3237,
      "step": 1654
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.000198632400254007,
      "loss": 0.3204,
      "step": 1655
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001985175507444171,
      "loss": 0.5049,
      "step": 1656
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001984026694596832,
      "loss": 0.614,
      "step": 1657
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001982877564750433,
      "loss": 0.5509,
      "step": 1658
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019817281186575633,
      "loss": 0.4043,
      "step": 1659
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019805783570710193,
      "loss": 0.3364,
      "step": 1660
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019794282807438028,
      "loss": 0.4562,
      "step": 1661
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019782778904291223,
      "loss": 0.3775,
      "step": 1662
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019771271868803934,
      "loss": 0.245,
      "step": 1663
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001975976170851235,
      "loss": 0.2359,
      "step": 1664
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019748248430954702,
      "loss": 0.3046,
      "step": 1665
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001973673204367129,
      "loss": 0.3034,
      "step": 1666
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001972521255420442,
      "loss": 0.2551,
      "step": 1667
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001971368997009845,
      "loss": 0.3397,
      "step": 1668
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019702164298899755,
      "loss": 0.0973,
      "step": 1669
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019690635548156736,
      "loss": 0.4279,
      "step": 1670
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019679103725419808,
      "loss": 0.3715,
      "step": 1671
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.000196675688382414,
      "loss": 0.4122,
      "step": 1672
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019656030894175952,
      "loss": 0.5655,
      "step": 1673
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019644489900779894,
      "loss": 0.2548,
      "step": 1674
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001963294586561166,
      "loss": 0.6353,
      "step": 1675
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019621398796231687,
      "loss": 0.3549,
      "step": 1676
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001960984870020238,
      "loss": 0.274,
      "step": 1677
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019598295585088134,
      "loss": 0.3806,
      "step": 1678
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019586739458455328,
      "loss": 0.2599,
      "step": 1679
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019575180327872303,
      "loss": 0.4013,
      "step": 1680
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019563618200909373,
      "loss": 0.3349,
      "step": 1681
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019552053085138807,
      "loss": 0.1509,
      "step": 1682
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019540484988134858,
      "loss": 0.7109,
      "step": 1683
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001952891391747369,
      "loss": 0.6271,
      "step": 1684
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019517339880733444,
      "loss": 0.3226,
      "step": 1685
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019505762885494194,
      "loss": 0.1961,
      "step": 1686
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019494182939337957,
      "loss": 0.0914,
      "step": 1687
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019482600049848674,
      "loss": 0.2007,
      "step": 1688
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019471014224612217,
      "loss": 0.4294,
      "step": 1689
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019459425471216385,
      "loss": 0.1039,
      "step": 1690
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001944783379725089,
      "loss": 0.4432,
      "step": 1691
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001943623921030736,
      "loss": 0.1881,
      "step": 1692
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019424641717979326,
      "loss": 0.2873,
      "step": 1693
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019413041327862225,
      "loss": 0.3552,
      "step": 1694
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019401438047553395,
      "loss": 0.4914,
      "step": 1695
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019389831884652058,
      "loss": 0.3066,
      "step": 1696
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019378222846759335,
      "loss": 0.4252,
      "step": 1697
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019366610941478222,
      "loss": 0.7134,
      "step": 1698
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019354996176413594,
      "loss": 0.5188,
      "step": 1699
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019343378559172203,
      "loss": 0.1614,
      "step": 1700
    },
    {
      "epoch": 0.42,
      "eval_loss": 0.383247971534729,
      "eval_runtime": 53.0312,
      "eval_samples_per_second": 0.622,
      "eval_steps_per_second": 0.622,
      "step": 1700
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001933175809736266,
      "loss": 0.4849,
      "step": 1701
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001932013479859545,
      "loss": 0.3526,
      "step": 1702
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019308508670482912,
      "loss": 0.1254,
      "step": 1703
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019296879720639233,
      "loss": 0.3755,
      "step": 1704
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019285247956680454,
      "loss": 0.3526,
      "step": 1705
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019273613386224458,
      "loss": 0.2913,
      "step": 1706
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019261976016890964,
      "loss": 0.2574,
      "step": 1707
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019250335856301523,
      "loss": 0.2226,
      "step": 1708
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019238692912079518,
      "loss": 0.1322,
      "step": 1709
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019227047191850154,
      "loss": 0.1524,
      "step": 1710
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019215398703240454,
      "loss": 0.7697,
      "step": 1711
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001920374745387925,
      "loss": 0.251,
      "step": 1712
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019192093451397187,
      "loss": 0.0396,
      "step": 1713
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001918043670342671,
      "loss": 0.3298,
      "step": 1714
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001916877721760207,
      "loss": 0.4496,
      "step": 1715
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019157115001559296,
      "loss": 0.5968,
      "step": 1716
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019145450062936218,
      "loss": 0.3935,
      "step": 1717
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019133782409372444,
      "loss": 0.7332,
      "step": 1718
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019122112048509363,
      "loss": 0.1237,
      "step": 1719
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019110438987990128,
      "loss": 0.2155,
      "step": 1720
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019098763235459674,
      "loss": 0.6298,
      "step": 1721
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001908708479856469,
      "loss": 0.7054,
      "step": 1722
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001907540368495362,
      "loss": 0.1746,
      "step": 1723
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019063719902276676,
      "loss": 0.2095,
      "step": 1724
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.000190520334581858,
      "loss": 0.3244,
      "step": 1725
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019040344360334694,
      "loss": 0.3562,
      "step": 1726
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001902865261637878,
      "loss": 0.4378,
      "step": 1727
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019016958233975223,
      "loss": 0.8404,
      "step": 1728
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019005261220782918,
      "loss": 0.4675,
      "step": 1729
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001899356158446248,
      "loss": 0.3494,
      "step": 1730
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00018981859332676244,
      "loss": 0.1958,
      "step": 1731
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00018970154473088254,
      "loss": 0.4893,
      "step": 1732
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001895844701336426,
      "loss": 0.2793,
      "step": 1733
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001894673696117172,
      "loss": 0.4733,
      "step": 1734
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00018935024324179795,
      "loss": 0.4698,
      "step": 1735
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00018923309110059324,
      "loss": 0.2807,
      "step": 1736
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00018911591326482846,
      "loss": 0.6594,
      "step": 1737
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00018899870981124584,
      "loss": 0.3443,
      "step": 1738
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00018888148081660423,
      "loss": 0.3299,
      "step": 1739
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001887642263576794,
      "loss": 0.2433,
      "step": 1740
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001886469465112637,
      "loss": 0.4946,
      "step": 1741
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018852964135416612,
      "loss": 0.5562,
      "step": 1742
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018841231096321215,
      "loss": 0.4137,
      "step": 1743
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018829495541524395,
      "loss": 0.2769,
      "step": 1744
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001881775747871201,
      "loss": 0.3922,
      "step": 1745
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018806016915571557,
      "loss": 0.5417,
      "step": 1746
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018794273859792173,
      "loss": 0.8971,
      "step": 1747
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018782528319064628,
      "loss": 0.4498,
      "step": 1748
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018770780301081316,
      "loss": 0.6661,
      "step": 1749
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018759029813536263,
      "loss": 0.3745,
      "step": 1750
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018747276864125099,
      "loss": 0.4789,
      "step": 1751
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001873552146054507,
      "loss": 0.2095,
      "step": 1752
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018723763610495038,
      "loss": 0.3017,
      "step": 1753
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018712003321675458,
      "loss": 0.7464,
      "step": 1754
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001870024060178839,
      "loss": 0.3366,
      "step": 1755
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001868847545853748,
      "loss": 0.1964,
      "step": 1756
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018676707899627963,
      "loss": 0.3461,
      "step": 1757
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018664937932766652,
      "loss": 0.5766,
      "step": 1758
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018653165565661947,
      "loss": 0.6553,
      "step": 1759
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018641390806023805,
      "loss": 0.4538,
      "step": 1760
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018629613661563771,
      "loss": 0.3396,
      "step": 1761
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001861783413999493,
      "loss": 0.1476,
      "step": 1762
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001860605224903194,
      "loss": 0.5166,
      "step": 1763
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018594267996390994,
      "loss": 0.2633,
      "step": 1764
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018582481389789853,
      "loss": 0.461,
      "step": 1765
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018570692436947804,
      "loss": 0.4119,
      "step": 1766
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018558901145585675,
      "loss": 0.4326,
      "step": 1767
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018547107523425834,
      "loss": 0.3179,
      "step": 1768
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001853531157819215,
      "loss": 0.2234,
      "step": 1769
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018523513317610045,
      "loss": 0.4883,
      "step": 1770
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001851171274940644,
      "loss": 0.4494,
      "step": 1771
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018499909881309766,
      "loss": 0.3473,
      "step": 1772
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018488104721049966,
      "loss": 0.4499,
      "step": 1773
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018476297276358484,
      "loss": 0.7145,
      "step": 1774
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018464487554968262,
      "loss": 0.8562,
      "step": 1775
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018452675564613726,
      "loss": 0.6474,
      "step": 1776
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001844086131303079,
      "loss": 0.552,
      "step": 1777
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018429044807956858,
      "loss": 0.6013,
      "step": 1778
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.000184172260571308,
      "loss": 0.3099,
      "step": 1779
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018405405068292954,
      "loss": 0.3136,
      "step": 1780
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018393581849185137,
      "loss": 0.2849,
      "step": 1781
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018381756407550613,
      "loss": 0.2918,
      "step": 1782
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018369928751134113,
      "loss": 0.4002,
      "step": 1783
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018358098887681805,
      "loss": 0.3782,
      "step": 1784
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018346266824941317,
      "loss": 0.5804,
      "step": 1785
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001833443257066171,
      "loss": 0.2552,
      "step": 1786
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018322596132593484,
      "loss": 0.4759,
      "step": 1787
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018310757518488557,
      "loss": 0.5138,
      "step": 1788
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001829891673610029,
      "loss": 0.3227,
      "step": 1789
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001828707379318345,
      "loss": 0.2596,
      "step": 1790
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001827522869749423,
      "loss": 0.2768,
      "step": 1791
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001826338145679022,
      "loss": 0.5163,
      "step": 1792
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001825153207883043,
      "loss": 0.2609,
      "step": 1793
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001823968057137525,
      "loss": 0.2243,
      "step": 1794
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018227826942186494,
      "loss": 0.431,
      "step": 1795
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018215971199027334,
      "loss": 0.4859,
      "step": 1796
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018204113349662345,
      "loss": 0.3921,
      "step": 1797
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001819225340185748,
      "loss": 0.6098,
      "step": 1798
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018180391363380058,
      "loss": 0.6177,
      "step": 1799
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018168527241998775,
      "loss": 0.5903,
      "step": 1800
    },
    {
      "epoch": 0.45,
      "eval_loss": 0.37952110171318054,
      "eval_runtime": 53.0353,
      "eval_samples_per_second": 0.622,
      "eval_steps_per_second": 0.622,
      "step": 1800
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018156661045483694,
      "loss": 0.1894,
      "step": 1801
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001814479278160622,
      "loss": 0.5884,
      "step": 1802
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018132922458139134,
      "loss": 0.112,
      "step": 1803
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018121050082856548,
      "loss": 0.3055,
      "step": 1804
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018109175663533932,
      "loss": 0.3877,
      "step": 1805
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018097299207948082,
      "loss": 0.941,
      "step": 1806
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018085420723877138,
      "loss": 0.3676,
      "step": 1807
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018073540219100563,
      "loss": 0.6736,
      "step": 1808
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018061657701399144,
      "loss": 0.3077,
      "step": 1809
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018049773178554986,
      "loss": 0.3827,
      "step": 1810
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001803788665835151,
      "loss": 0.2691,
      "step": 1811
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001802599814857344,
      "loss": 0.6347,
      "step": 1812
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018014107657006808,
      "loss": 0.4974,
      "step": 1813
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001800221519143894,
      "loss": 0.2156,
      "step": 1814
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001799032075965846,
      "loss": 0.2245,
      "step": 1815
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00017978424369455276,
      "loss": 0.2605,
      "step": 1816
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001796652602862058,
      "loss": 0.5792,
      "step": 1817
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00017954625744946837,
      "loss": 0.3411,
      "step": 1818
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001794272352622779,
      "loss": 0.2415,
      "step": 1819
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00017930819380258453,
      "loss": 0.2976,
      "step": 1820
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00017918913314835086,
      "loss": 0.8317,
      "step": 1821
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017907005337755222,
      "loss": 0.4741,
      "step": 1822
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017895095456817644,
      "loss": 0.162,
      "step": 1823
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001788318367982237,
      "loss": 0.9991,
      "step": 1824
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017871270014570677,
      "loss": 0.3996,
      "step": 1825
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017859354468865063,
      "loss": 0.0708,
      "step": 1826
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017847437050509272,
      "loss": 0.6669,
      "step": 1827
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001783551776730826,
      "loss": 0.5709,
      "step": 1828
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017823596627068215,
      "loss": 0.0724,
      "step": 1829
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017811673637596536,
      "loss": 0.2662,
      "step": 1830
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017799748806701836,
      "loss": 0.2488,
      "step": 1831
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001778782214219393,
      "loss": 0.3486,
      "step": 1832
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017775893651883836,
      "loss": 0.0957,
      "step": 1833
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017763963343583774,
      "loss": 0.6034,
      "step": 1834
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017752031225107143,
      "loss": 0.2886,
      "step": 1835
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017740097304268533,
      "loss": 0.5296,
      "step": 1836
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001772816158888372,
      "loss": 0.2639,
      "step": 1837
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017716224086769647,
      "loss": 0.9306,
      "step": 1838
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001770428480574443,
      "loss": 0.4149,
      "step": 1839
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017692343753627352,
      "loss": 0.2325,
      "step": 1840
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017680400938238853,
      "loss": 0.2294,
      "step": 1841
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017668456367400524,
      "loss": 0.2273,
      "step": 1842
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001765651004893512,
      "loss": 0.2443,
      "step": 1843
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017644561990666523,
      "loss": 0.3962,
      "step": 1844
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001763261220041977,
      "loss": 0.2115,
      "step": 1845
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001762066068602102,
      "loss": 0.1232,
      "step": 1846
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017608707455297567,
      "loss": 0.1561,
      "step": 1847
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001759675251607783,
      "loss": 0.1199,
      "step": 1848
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017584795876191342,
      "loss": 0.3873,
      "step": 1849
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017572837543468754,
      "loss": 0.2053,
      "step": 1850
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017560877525741827,
      "loss": 0.199,
      "step": 1851
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001754891583084342,
      "loss": 0.1631,
      "step": 1852
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017536952466607494,
      "loss": 0.2827,
      "step": 1853
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017524987440869106,
      "loss": 0.1573,
      "step": 1854
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017513020761464389,
      "loss": 0.472,
      "step": 1855
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001750105243623058,
      "loss": 0.735,
      "step": 1856
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001748908247300598,
      "loss": 0.232,
      "step": 1857
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017477110879629958,
      "loss": 0.2118,
      "step": 1858
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017465137663942958,
      "loss": 0.407,
      "step": 1859
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001745316283378649,
      "loss": 0.2908,
      "step": 1860
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017441186397003115,
      "loss": 0.1168,
      "step": 1861
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017429208361436447,
      "loss": 0.4992,
      "step": 1862
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017417228734931155,
      "loss": 0.827,
      "step": 1863
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017405247525332937,
      "loss": 0.6282,
      "step": 1864
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017393264740488533,
      "loss": 0.203,
      "step": 1865
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001738128038824572,
      "loss": 0.5777,
      "step": 1866
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017369294476453297,
      "loss": 0.1718,
      "step": 1867
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017357307012961083,
      "loss": 0.5423,
      "step": 1868
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017345318005619912,
      "loss": 0.1326,
      "step": 1869
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001733332746228164,
      "loss": 0.3932,
      "step": 1870
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017321335390799113,
      "loss": 0.3184,
      "step": 1871
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001730934179902619,
      "loss": 0.5672,
      "step": 1872
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017297346694817719,
      "loss": 0.311,
      "step": 1873
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017285350086029545,
      "loss": 0.1014,
      "step": 1874
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017273351980518487,
      "loss": 0.4045,
      "step": 1875
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017261352386142362,
      "loss": 0.6993,
      "step": 1876
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017249351310759942,
      "loss": 0.5287,
      "step": 1877
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017237348762230988,
      "loss": 0.4646,
      "step": 1878
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001722534474841621,
      "loss": 0.3302,
      "step": 1879
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017213339277177283,
      "loss": 0.1968,
      "step": 1880
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001720133235637684,
      "loss": 0.3885,
      "step": 1881
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017189323993878464,
      "loss": 0.3192,
      "step": 1882
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017177314197546677,
      "loss": 0.4297,
      "step": 1883
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017165302975246941,
      "loss": 0.2662,
      "step": 1884
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001715329033484565,
      "loss": 0.532,
      "step": 1885
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017141276284210133,
      "loss": 0.5907,
      "step": 1886
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017129260831208638,
      "loss": 0.2261,
      "step": 1887
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001711724398371034,
      "loss": 0.6047,
      "step": 1888
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001710522574958531,
      "loss": 0.3174,
      "step": 1889
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017093206136704545,
      "loss": 0.5594,
      "step": 1890
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017081185152939934,
      "loss": 0.4663,
      "step": 1891
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017069162806164265,
      "loss": 0.2667,
      "step": 1892
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017057139104251223,
      "loss": 0.6285,
      "step": 1893
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017045114055075377,
      "loss": 0.3518,
      "step": 1894
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017033087666512185,
      "loss": 0.5161,
      "step": 1895
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017021059946437962,
      "loss": 0.6742,
      "step": 1896
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017009030902729932,
      "loss": 0.4596,
      "step": 1897
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00016997000543266148,
      "loss": 0.1944,
      "step": 1898
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00016984968875925543,
      "loss": 0.2503,
      "step": 1899
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00016972935908587903,
      "loss": 0.1128,
      "step": 1900
    },
    {
      "epoch": 0.47,
      "eval_loss": 0.3773733377456665,
      "eval_runtime": 49.8368,
      "eval_samples_per_second": 0.662,
      "eval_steps_per_second": 0.662,
      "step": 1900
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00016960901649133874,
      "loss": 0.1711,
      "step": 1901
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016948866105444933,
      "loss": 0.5478,
      "step": 1902
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016936829285403408,
      "loss": 0.2326,
      "step": 1903
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001692479119689246,
      "loss": 0.1426,
      "step": 1904
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016912751847796082,
      "loss": 0.2836,
      "step": 1905
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016900711245999088,
      "loss": 0.3777,
      "step": 1906
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016888669399387125,
      "loss": 0.5997,
      "step": 1907
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016876626315846636,
      "loss": 0.6851,
      "step": 1908
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016864582003264894,
      "loss": 0.5082,
      "step": 1909
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016852536469529964,
      "loss": 0.1799,
      "step": 1910
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016840489722530713,
      "loss": 0.4348,
      "step": 1911
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016828441770156804,
      "loss": 0.2991,
      "step": 1912
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001681639262029869,
      "loss": 0.5405,
      "step": 1913
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016804342280847607,
      "loss": 0.6605,
      "step": 1914
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016792290759695574,
      "loss": 0.7856,
      "step": 1915
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016780238064735377,
      "loss": 0.1172,
      "step": 1916
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016768184203860576,
      "loss": 0.2089,
      "step": 1917
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001675612918496549,
      "loss": 0.498,
      "step": 1918
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016744073015945206,
      "loss": 0.5187,
      "step": 1919
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016732015704695556,
      "loss": 0.1837,
      "step": 1920
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016719957259113112,
      "loss": 0.2473,
      "step": 1921
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016707897687095208,
      "loss": 0.5534,
      "step": 1922
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016695836996539907,
      "loss": 0.1563,
      "step": 1923
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016683775195345998,
      "loss": 0.74,
      "step": 1924
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016671712291413006,
      "loss": 0.4199,
      "step": 1925
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016659648292641173,
      "loss": 0.1395,
      "step": 1926
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016647583206931465,
      "loss": 0.2541,
      "step": 1927
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016635517042185553,
      "loss": 0.4999,
      "step": 1928
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001662344980630582,
      "loss": 0.3288,
      "step": 1929
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016611381507195347,
      "loss": 0.192,
      "step": 1930
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016599312152757904,
      "loss": 0.3125,
      "step": 1931
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016587241750897966,
      "loss": 0.3396,
      "step": 1932
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001657517030952069,
      "loss": 0.1681,
      "step": 1933
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001656309783653191,
      "loss": 0.3051,
      "step": 1934
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016551024339838137,
      "loss": 0.4051,
      "step": 1935
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016538949827346548,
      "loss": 0.2681,
      "step": 1936
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016526874306964996,
      "loss": 0.3966,
      "step": 1937
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001651479778660199,
      "loss": 0.4514,
      "step": 1938
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016502720274166692,
      "loss": 0.8359,
      "step": 1939
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00016490641777568908,
      "loss": 0.278,
      "step": 1940
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.000164785623047191,
      "loss": 0.109,
      "step": 1941
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016466481863528355,
      "loss": 0.2944,
      "step": 1942
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016454400461908417,
      "loss": 0.4651,
      "step": 1943
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001644231810777163,
      "loss": 0.7331,
      "step": 1944
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016430234809030987,
      "loss": 0.4566,
      "step": 1945
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016418150573600085,
      "loss": 0.5858,
      "step": 1946
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001640606540939314,
      "loss": 0.4823,
      "step": 1947
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016393979324324972,
      "loss": 0.4245,
      "step": 1948
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016381892326311018,
      "loss": 0.5424,
      "step": 1949
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016369804423267288,
      "loss": 0.2064,
      "step": 1950
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016357715623110413,
      "loss": 0.4704,
      "step": 1951
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016345625933757583,
      "loss": 0.4227,
      "step": 1952
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016333535363126595,
      "loss": 0.2494,
      "step": 1953
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001632144391913581,
      "loss": 0.4958,
      "step": 1954
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016309351609704163,
      "loss": 0.52,
      "step": 1955
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016297258442751156,
      "loss": 0.349,
      "step": 1956
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016285164426196855,
      "loss": 0.5252,
      "step": 1957
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016273069567961882,
      "loss": 0.3483,
      "step": 1958
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016260973875967407,
      "loss": 0.0459,
      "step": 1959
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016248877358135148,
      "loss": 0.2361,
      "step": 1960
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016236780022387364,
      "loss": 0.1374,
      "step": 1961
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001622468187664685,
      "loss": 0.1813,
      "step": 1962
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016212582928836928,
      "loss": 0.1555,
      "step": 1963
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016200483186881454,
      "loss": 0.5191,
      "step": 1964
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016188382658704795,
      "loss": 0.479,
      "step": 1965
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016176281352231839,
      "loss": 0.3101,
      "step": 1966
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001616417927538797,
      "loss": 0.8164,
      "step": 1967
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016152076436099104,
      "loss": 0.2884,
      "step": 1968
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001613997284229163,
      "loss": 0.3501,
      "step": 1969
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016127868501892445,
      "loss": 0.4148,
      "step": 1970
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001611576342282893,
      "loss": 0.3691,
      "step": 1971
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016103657613028948,
      "loss": 0.6668,
      "step": 1972
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016091551080420847,
      "loss": 0.3423,
      "step": 1973
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016079443832933446,
      "loss": 0.2776,
      "step": 1974
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016067335878496025,
      "loss": 0.0357,
      "step": 1975
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016055227225038337,
      "loss": 0.317,
      "step": 1976
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016043117880490588,
      "loss": 0.3295,
      "step": 1977
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001603100785278344,
      "loss": 0.6408,
      "step": 1978
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001601889714984799,
      "loss": 0.1918,
      "step": 1979
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00016006785779615798,
      "loss": 0.3278,
      "step": 1980
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00015994673750018847,
      "loss": 0.3531,
      "step": 1981
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015982561068989557,
      "loss": 0.3281,
      "step": 1982
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015970447744460765,
      "loss": 0.6517,
      "step": 1983
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015958333784365743,
      "loss": 0.5414,
      "step": 1984
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015946219196638173,
      "loss": 0.6344,
      "step": 1985
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015934103989212144,
      "loss": 0.4613,
      "step": 1986
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015921988170022156,
      "loss": 0.3887,
      "step": 1987
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015909871747003113,
      "loss": 0.2931,
      "step": 1988
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001589775472809031,
      "loss": 0.3825,
      "step": 1989
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015885637121219426,
      "loss": 0.6134,
      "step": 1990
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001587351893432654,
      "loss": 0.1629,
      "step": 1991
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015861400175348097,
      "loss": 0.2529,
      "step": 1992
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001584928085222092,
      "loss": 0.3184,
      "step": 1993
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015837160972882212,
      "loss": 0.2343,
      "step": 1994
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015825040545269526,
      "loss": 0.2488,
      "step": 1995
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015812919577320777,
      "loss": 0.4047,
      "step": 1996
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001580079807697424,
      "loss": 0.4129,
      "step": 1997
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015788676052168538,
      "loss": 0.5673,
      "step": 1998
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015776553510842632,
      "loss": 0.3441,
      "step": 1999
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015764430460935827,
      "loss": 0.2917,
      "step": 2000
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.37665513157844543,
      "eval_runtime": 51.8167,
      "eval_samples_per_second": 0.637,
      "eval_steps_per_second": 0.637,
      "step": 2000
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015752306910387753,
      "loss": 0.3905,
      "step": 2001
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015740182867138376,
      "loss": 0.4305,
      "step": 2002
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015728058339127982,
      "loss": 0.1672,
      "step": 2003
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015715933334297175,
      "loss": 0.276,
      "step": 2004
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015703807860586872,
      "loss": 0.5642,
      "step": 2005
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001569168192593829,
      "loss": 0.1861,
      "step": 2006
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015679555538292958,
      "loss": 0.4395,
      "step": 2007
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015667428705592695,
      "loss": 0.2684,
      "step": 2008
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015655301435779618,
      "loss": 0.1428,
      "step": 2009
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001564317373679612,
      "loss": 0.4085,
      "step": 2010
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015631045616584888,
      "loss": 0.3486,
      "step": 2011
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015618917083088873,
      "loss": 0.3385,
      "step": 2012
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015606788144251302,
      "loss": 0.2059,
      "step": 2013
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001559465880801567,
      "loss": 0.6775,
      "step": 2014
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015582529082325722,
      "loss": 0.645,
      "step": 2015
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015570398975125473,
      "loss": 0.2167,
      "step": 2016
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015558268494359176,
      "loss": 0.2567,
      "step": 2017
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015546137647971337,
      "loss": 0.3246,
      "step": 2018
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015534006443906694,
      "loss": 0.4725,
      "step": 2019
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015521874890110223,
      "loss": 0.3501,
      "step": 2020
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001550974299452713,
      "loss": 0.4035,
      "step": 2021
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001549761076510284,
      "loss": 0.5689,
      "step": 2022
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015485478209783008,
      "loss": 0.5197,
      "step": 2023
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015473345336513485,
      "loss": 0.5577,
      "step": 2024
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015461212153240343,
      "loss": 0.5253,
      "step": 2025
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001544907866790986,
      "loss": 0.3906,
      "step": 2026
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015436944888468494,
      "loss": 0.1411,
      "step": 2027
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015424810822862916,
      "loss": 0.5907,
      "step": 2028
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015412676479039975,
      "loss": 0.4674,
      "step": 2029
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015400541864946704,
      "loss": 0.7692,
      "step": 2030
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015388406988530308,
      "loss": 0.1827,
      "step": 2031
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015376271857738165,
      "loss": 0.2144,
      "step": 2032
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001536413648051783,
      "loss": 0.8304,
      "step": 2033
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015352000864817008,
      "loss": 0.6906,
      "step": 2034
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001533986501858356,
      "loss": 0.3657,
      "step": 2035
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015327728949765513,
      "loss": 0.4305,
      "step": 2036
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015315592666311018,
      "loss": 0.1144,
      "step": 2037
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015303456176168375,
      "loss": 0.2637,
      "step": 2038
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015291319487286033,
      "loss": 0.3878,
      "step": 2039
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015279182607612552,
      "loss": 0.2142,
      "step": 2040
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015267045545096626,
      "loss": 0.3882,
      "step": 2041
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015254908307687072,
      "loss": 0.3685,
      "step": 2042
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015242770903332807,
      "loss": 0.5632,
      "step": 2043
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001523063333998288,
      "loss": 0.6861,
      "step": 2044
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001521849562558642,
      "loss": 0.0845,
      "step": 2045
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015206357768092672,
      "loss": 0.7327,
      "step": 2046
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015194219775450964,
      "loss": 0.4747,
      "step": 2047
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015182081655610726,
      "loss": 0.4397,
      "step": 2048
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015169943416521456,
      "loss": 0.2359,
      "step": 2049
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001515780506613274,
      "loss": 0.6112,
      "step": 2050
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015145666612394236,
      "loss": 0.3761,
      "step": 2051
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015133528063255664,
      "loss": 0.57,
      "step": 2052
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015121389426666813,
      "loss": 0.4737,
      "step": 2053
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015109250710577523,
      "loss": 0.362,
      "step": 2054
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001509711192293769,
      "loss": 0.1696,
      "step": 2055
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015084973071697254,
      "loss": 0.4496,
      "step": 2056
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015072834164806207,
      "loss": 0.515,
      "step": 2057
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001506069521021456,
      "loss": 0.4728,
      "step": 2058
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015048556215872366,
      "loss": 0.2667,
      "step": 2059
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001503641718972971,
      "loss": 0.3822,
      "step": 2060
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00015024278139736687,
      "loss": 0.3602,
      "step": 2061
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001501213907384341,
      "loss": 0.2177,
      "step": 2062
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00015,
      "loss": 0.4957,
      "step": 2063
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014987860926156588,
      "loss": 0.3027,
      "step": 2064
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014975721860263313,
      "loss": 0.2065,
      "step": 2065
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014963582810270282,
      "loss": 0.8752,
      "step": 2066
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014951443784127632,
      "loss": 0.1491,
      "step": 2067
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014939304789785438,
      "loss": 0.4968,
      "step": 2068
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014927165835193796,
      "loss": 0.496,
      "step": 2069
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014915026928302743,
      "loss": 0.5972,
      "step": 2070
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014902888077062313,
      "loss": 0.2721,
      "step": 2071
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014890749289422475,
      "loss": 0.2098,
      "step": 2072
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014878610573333187,
      "loss": 0.2822,
      "step": 2073
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014866471936744337,
      "loss": 0.4724,
      "step": 2074
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001485433338760576,
      "loss": 0.2437,
      "step": 2075
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014842194933867257,
      "loss": 0.1182,
      "step": 2076
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001483005658347854,
      "loss": 0.3774,
      "step": 2077
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014817918344389274,
      "loss": 0.3276,
      "step": 2078
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014805780224549033,
      "loss": 0.6734,
      "step": 2079
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001479364223190733,
      "loss": 0.4967,
      "step": 2080
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014781504374413582,
      "loss": 0.329,
      "step": 2081
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014769366660017126,
      "loss": 0.356,
      "step": 2082
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001475722909666719,
      "loss": 0.5213,
      "step": 2083
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014745091692312925,
      "loss": 0.3394,
      "step": 2084
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001473295445490337,
      "loss": 0.1735,
      "step": 2085
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014720817392387442,
      "loss": 0.5633,
      "step": 2086
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014708680512713965,
      "loss": 0.4252,
      "step": 2087
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001469654382383162,
      "loss": 0.4853,
      "step": 2088
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014684407333688984,
      "loss": 0.3869,
      "step": 2089
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014672271050234485,
      "loss": 0.4128,
      "step": 2090
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014660134981416437,
      "loss": 0.296,
      "step": 2091
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014647999135182992,
      "loss": 0.7265,
      "step": 2092
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001463586351948217,
      "loss": 0.5942,
      "step": 2093
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014623728142261835,
      "loss": 0.6765,
      "step": 2094
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001461159301146969,
      "loss": 0.3256,
      "step": 2095
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014599458135053297,
      "loss": 0.3509,
      "step": 2096
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001458732352096002,
      "loss": 0.3919,
      "step": 2097
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014575189177137084,
      "loss": 0.621,
      "step": 2098
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014563055111531503,
      "loss": 0.3726,
      "step": 2099
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014550921332090144,
      "loss": 0.2403,
      "step": 2100
    },
    {
      "epoch": 0.52,
      "eval_loss": 0.37003716826438904,
      "eval_runtime": 49.9759,
      "eval_samples_per_second": 0.66,
      "eval_steps_per_second": 0.66,
      "step": 2100
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014538787846759654,
      "loss": 0.2819,
      "step": 2101
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014526654663486518,
      "loss": 0.2955,
      "step": 2102
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014514521790216992,
      "loss": 0.647,
      "step": 2103
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014502389234897153,
      "loss": 0.4052,
      "step": 2104
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001449025700547287,
      "loss": 0.3127,
      "step": 2105
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014478125109889772,
      "loss": 0.2561,
      "step": 2106
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014465993556093306,
      "loss": 0.3505,
      "step": 2107
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001445386235202866,
      "loss": 0.2727,
      "step": 2108
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014441731505640822,
      "loss": 0.1966,
      "step": 2109
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014429601024874527,
      "loss": 0.3529,
      "step": 2110
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014417470917674278,
      "loss": 0.4242,
      "step": 2111
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014405341191984334,
      "loss": 0.3817,
      "step": 2112
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014393211855748698,
      "loss": 0.1471,
      "step": 2113
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014381082916911127,
      "loss": 0.3444,
      "step": 2114
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001436895438341511,
      "loss": 0.4558,
      "step": 2115
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001435682626320388,
      "loss": 0.5306,
      "step": 2116
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001434469856422038,
      "loss": 0.2398,
      "step": 2117
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014332571294407303,
      "loss": 0.4739,
      "step": 2118
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001432044446170704,
      "loss": 0.4564,
      "step": 2119
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014308318074061712,
      "loss": 0.7207,
      "step": 2120
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014296192139413128,
      "loss": 0.577,
      "step": 2121
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014284066665702825,
      "loss": 0.4949,
      "step": 2122
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014271941660872015,
      "loss": 0.3759,
      "step": 2123
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001425981713286162,
      "loss": 0.0614,
      "step": 2124
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014247693089612247,
      "loss": 0.5923,
      "step": 2125
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001423556953906417,
      "loss": 0.4894,
      "step": 2126
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014223446489157365,
      "loss": 0.4894,
      "step": 2127
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001421132394783146,
      "loss": 0.4411,
      "step": 2128
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014199201923025757,
      "loss": 0.3865,
      "step": 2129
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001418708042267922,
      "loss": 0.1979,
      "step": 2130
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014174959454730474,
      "loss": 0.6672,
      "step": 2131
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014162839027117788,
      "loss": 0.3923,
      "step": 2132
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014150719147779076,
      "loss": 0.423,
      "step": 2133
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.000141385998246519,
      "loss": 0.3076,
      "step": 2134
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014126481065673455,
      "loss": 0.4979,
      "step": 2135
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014114362878780572,
      "loss": 0.3191,
      "step": 2136
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014102245271909686,
      "loss": 0.3061,
      "step": 2137
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014090128252996885,
      "loss": 0.1339,
      "step": 2138
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014078011829977838,
      "loss": 0.2644,
      "step": 2139
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014065896010787858,
      "loss": 0.1201,
      "step": 2140
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014053780803361827,
      "loss": 0.3364,
      "step": 2141
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001404166621563426,
      "loss": 0.6111,
      "step": 2142
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00014029552255539232,
      "loss": 0.4671,
      "step": 2143
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00014017438931010446,
      "loss": 0.6022,
      "step": 2144
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001400532624998115,
      "loss": 0.1982,
      "step": 2145
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.000139932142203842,
      "loss": 0.3026,
      "step": 2146
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013981102850152007,
      "loss": 0.1238,
      "step": 2147
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013968992147216563,
      "loss": 0.2758,
      "step": 2148
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013956882119509412,
      "loss": 0.6104,
      "step": 2149
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013944772774961663,
      "loss": 0.2332,
      "step": 2150
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013932664121503975,
      "loss": 0.3354,
      "step": 2151
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013920556167066556,
      "loss": 0.3065,
      "step": 2152
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013908448919579155,
      "loss": 0.6547,
      "step": 2153
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001389634238697105,
      "loss": 0.3515,
      "step": 2154
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013884236577171066,
      "loss": 0.3179,
      "step": 2155
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013872131498107555,
      "loss": 0.3536,
      "step": 2156
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013860027157708364,
      "loss": 0.5085,
      "step": 2157
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013847923563900896,
      "loss": 0.419,
      "step": 2158
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013835820724612027,
      "loss": 0.4549,
      "step": 2159
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013823718647768167,
      "loss": 0.2148,
      "step": 2160
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013811617341295202,
      "loss": 0.2618,
      "step": 2161
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013799516813118548,
      "loss": 0.3594,
      "step": 2162
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001378741707116307,
      "loss": 0.2348,
      "step": 2163
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013775318123353155,
      "loss": 0.61,
      "step": 2164
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013763219977612636,
      "loss": 0.2488,
      "step": 2165
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013751122641864852,
      "loss": 0.351,
      "step": 2166
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001373902612403259,
      "loss": 0.5019,
      "step": 2167
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013726930432038115,
      "loss": 0.1938,
      "step": 2168
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013714835573803142,
      "loss": 0.3822,
      "step": 2169
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001370274155724884,
      "loss": 0.2037,
      "step": 2170
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001369064839029584,
      "loss": 0.5503,
      "step": 2171
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001367855608086419,
      "loss": 0.3271,
      "step": 2172
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013666464636873408,
      "loss": 0.3536,
      "step": 2173
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013654374066242414,
      "loss": 0.6839,
      "step": 2174
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013642284376889585,
      "loss": 0.233,
      "step": 2175
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001363019557673271,
      "loss": 0.2549,
      "step": 2176
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001361810767368898,
      "loss": 0.1805,
      "step": 2177
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013606020675675025,
      "loss": 0.147,
      "step": 2178
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001359393459060686,
      "loss": 0.3791,
      "step": 2179
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013581849426399917,
      "loss": 0.2786,
      "step": 2180
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00013569765190969013,
      "loss": 0.3435,
      "step": 2181
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001355768189222837,
      "loss": 0.2248,
      "step": 2182
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013545599538091586,
      "loss": 0.5481,
      "step": 2183
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013533518136471645,
      "loss": 0.299,
      "step": 2184
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.000135214376952809,
      "loss": 0.3144,
      "step": 2185
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001350935822243109,
      "loss": 0.1447,
      "step": 2186
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013497279725833305,
      "loss": 0.2931,
      "step": 2187
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013485202213398004,
      "loss": 0.5552,
      "step": 2188
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013473125693035,
      "loss": 0.4117,
      "step": 2189
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001346105017265345,
      "loss": 0.4487,
      "step": 2190
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013448975660161866,
      "loss": 0.4009,
      "step": 2191
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013436902163468088,
      "loss": 0.3698,
      "step": 2192
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001342482969047931,
      "loss": 0.5931,
      "step": 2193
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013412758249102031,
      "loss": 0.3102,
      "step": 2194
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.000134006878472421,
      "loss": 0.5198,
      "step": 2195
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013388618492804653,
      "loss": 0.3882,
      "step": 2196
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013376550193694176,
      "loss": 0.5749,
      "step": 2197
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013364482957814444,
      "loss": 0.3805,
      "step": 2198
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001335241679306853,
      "loss": 0.4326,
      "step": 2199
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013340351707358825,
      "loss": 0.1926,
      "step": 2200
    },
    {
      "epoch": 0.55,
      "eval_loss": 0.3643794357776642,
      "eval_runtime": 50.0132,
      "eval_samples_per_second": 0.66,
      "eval_steps_per_second": 0.66,
      "step": 2200
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013328287708586992,
      "loss": 0.2657,
      "step": 2201
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013316224804654005,
      "loss": 0.5095,
      "step": 2202
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001330416300346009,
      "loss": 0.3205,
      "step": 2203
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013292102312904792,
      "loss": 0.4493,
      "step": 2204
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013280042740886885,
      "loss": 0.2309,
      "step": 2205
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013267984295304444,
      "loss": 0.2935,
      "step": 2206
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013255926984054794,
      "loss": 0.665,
      "step": 2207
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013243870815034505,
      "loss": 0.6173,
      "step": 2208
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013231815796139424,
      "loss": 0.1904,
      "step": 2209
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001321976193526462,
      "loss": 0.5545,
      "step": 2210
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013207709240304426,
      "loss": 0.3994,
      "step": 2211
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001319565771915239,
      "loss": 0.3139,
      "step": 2212
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013183607379701313,
      "loss": 0.3859,
      "step": 2213
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013171558229843196,
      "loss": 0.1823,
      "step": 2214
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001315951027746929,
      "loss": 0.3171,
      "step": 2215
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013147463530470036,
      "loss": 0.5852,
      "step": 2216
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.000131354179967351,
      "loss": 0.6021,
      "step": 2217
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001312337368415336,
      "loss": 0.5139,
      "step": 2218
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013111330600612872,
      "loss": 0.3271,
      "step": 2219
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001309928875400091,
      "loss": 0.5363,
      "step": 2220
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00013087248152203915,
      "loss": 0.5837,
      "step": 2221
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00013075208803107541,
      "loss": 0.4969,
      "step": 2222
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001306317071459659,
      "loss": 0.6453,
      "step": 2223
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00013051133894555067,
      "loss": 0.5964,
      "step": 2224
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00013039098350866126,
      "loss": 0.3049,
      "step": 2225
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001302706409141209,
      "loss": 0.9911,
      "step": 2226
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00013015031124074457,
      "loss": 0.1668,
      "step": 2227
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001300299945673385,
      "loss": 0.2663,
      "step": 2228
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012990969097270068,
      "loss": 0.429,
      "step": 2229
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012978940053562032,
      "loss": 0.2666,
      "step": 2230
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001296691233348782,
      "loss": 0.3274,
      "step": 2231
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012954885944924623,
      "loss": 0.5225,
      "step": 2232
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012942860895748782,
      "loss": 0.6021,
      "step": 2233
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012930837193835735,
      "loss": 0.9828,
      "step": 2234
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001291881484706007,
      "loss": 0.1253,
      "step": 2235
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012906793863295455,
      "loss": 0.3511,
      "step": 2236
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012894774250414687,
      "loss": 0.3801,
      "step": 2237
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001288275601628966,
      "loss": 0.5901,
      "step": 2238
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012870739168791356,
      "loss": 0.1679,
      "step": 2239
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012858723715789867,
      "loss": 0.2619,
      "step": 2240
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001284670966515435,
      "loss": 0.4401,
      "step": 2241
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012834697024753064,
      "loss": 0.1767,
      "step": 2242
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012822685802453326,
      "loss": 0.1221,
      "step": 2243
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012810676006121536,
      "loss": 0.1918,
      "step": 2244
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001279866764362316,
      "loss": 0.2692,
      "step": 2245
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012786660722822714,
      "loss": 0.2422,
      "step": 2246
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001277465525158379,
      "loss": 0.6472,
      "step": 2247
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001276265123776901,
      "loss": 0.3742,
      "step": 2248
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012750648689240055,
      "loss": 0.259,
      "step": 2249
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012738647613857635,
      "loss": 0.4841,
      "step": 2250
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012726648019481513,
      "loss": 0.3486,
      "step": 2251
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012714649913970455,
      "loss": 0.4883,
      "step": 2252
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012702653305182284,
      "loss": 0.5445,
      "step": 2253
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001269065820097381,
      "loss": 0.4099,
      "step": 2254
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001267866460920089,
      "loss": 0.3585,
      "step": 2255
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001266667253771836,
      "loss": 0.2406,
      "step": 2256
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012654681994380085,
      "loss": 0.3364,
      "step": 2257
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012642692987038917,
      "loss": 0.6573,
      "step": 2258
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.000126307055235467,
      "loss": 0.5536,
      "step": 2259
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001261871961175428,
      "loss": 0.3695,
      "step": 2260
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00012606735259511464,
      "loss": 0.245,
      "step": 2261
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012594752474667063,
      "loss": 0.2059,
      "step": 2262
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012582771265068842,
      "loss": 0.0965,
      "step": 2263
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001257079163856355,
      "loss": 0.1863,
      "step": 2264
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012558813602996883,
      "loss": 0.1651,
      "step": 2265
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001254683716621351,
      "loss": 0.2723,
      "step": 2266
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012534862336057042,
      "loss": 0.1927,
      "step": 2267
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001252288912037004,
      "loss": 0.2314,
      "step": 2268
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012510917526994023,
      "loss": 0.3112,
      "step": 2269
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012498947563769415,
      "loss": 0.2292,
      "step": 2270
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001248697923853561,
      "loss": 0.41,
      "step": 2271
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012475012559130894,
      "loss": 0.5653,
      "step": 2272
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012463047533392508,
      "loss": 0.119,
      "step": 2273
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001245108416915658,
      "loss": 0.3333,
      "step": 2274
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012439122474258176,
      "loss": 0.2444,
      "step": 2275
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012427162456531246,
      "loss": 0.396,
      "step": 2276
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012415204123808655,
      "loss": 0.1893,
      "step": 2277
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001240324748392217,
      "loss": 0.1879,
      "step": 2278
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001239129254470243,
      "loss": 0.107,
      "step": 2279
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001237933931397898,
      "loss": 0.3754,
      "step": 2280
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012367387799580226,
      "loss": 0.2675,
      "step": 2281
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012355438009333474,
      "loss": 0.4817,
      "step": 2282
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012343489951064878,
      "loss": 0.6072,
      "step": 2283
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012331543632599474,
      "loss": 0.3022,
      "step": 2284
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012319599061761147,
      "loss": 0.4948,
      "step": 2285
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001230765624637265,
      "loss": 0.4644,
      "step": 2286
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001229571519425557,
      "loss": 0.4709,
      "step": 2287
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012283775913230348,
      "loss": 0.4535,
      "step": 2288
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012271838411116278,
      "loss": 0.3758,
      "step": 2289
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012259902695731465,
      "loss": 0.3717,
      "step": 2290
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001224796877489286,
      "loss": 0.2497,
      "step": 2291
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012236036656416226,
      "loss": 0.6613,
      "step": 2292
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012224106348116166,
      "loss": 0.4748,
      "step": 2293
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001221217785780607,
      "loss": 0.4188,
      "step": 2294
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012200251193298167,
      "loss": 0.6086,
      "step": 2295
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012188326362403464,
      "loss": 0.1765,
      "step": 2296
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012176403372931785,
      "loss": 0.3946,
      "step": 2297
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012164482232691738,
      "loss": 0.3385,
      "step": 2298
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012152562949490728,
      "loss": 0.2532,
      "step": 2299
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012140645531134934,
      "loss": 0.4032,
      "step": 2300
    },
    {
      "epoch": 0.57,
      "eval_loss": 0.3617573082447052,
      "eval_runtime": 51.0386,
      "eval_samples_per_second": 0.647,
      "eval_steps_per_second": 0.647,
      "step": 2300
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00012128729985429323,
      "loss": 0.4259,
      "step": 2301
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00012116816320177627,
      "loss": 0.2111,
      "step": 2302
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00012104904543182357,
      "loss": 0.1114,
      "step": 2303
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001209299466224478,
      "loss": 0.2711,
      "step": 2304
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00012081086685164914,
      "loss": 0.351,
      "step": 2305
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00012069180619741551,
      "loss": 0.6061,
      "step": 2306
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00012057276473772206,
      "loss": 0.2757,
      "step": 2307
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00012045374255053158,
      "loss": 0.4012,
      "step": 2308
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00012033473971379418,
      "loss": 0.4122,
      "step": 2309
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001202157563054472,
      "loss": 0.9633,
      "step": 2310
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00012009679240341538,
      "loss": 0.2963,
      "step": 2311
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011997784808561056,
      "loss": 0.3982,
      "step": 2312
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011985892342993193,
      "loss": 0.3034,
      "step": 2313
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011974001851426559,
      "loss": 0.2736,
      "step": 2314
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011962113341648494,
      "loss": 0.32,
      "step": 2315
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011950226821445013,
      "loss": 0.2869,
      "step": 2316
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011938342298600858,
      "loss": 0.4553,
      "step": 2317
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011926459780899434,
      "loss": 0.3319,
      "step": 2318
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011914579276122861,
      "loss": 0.3657,
      "step": 2319
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011902700792051916,
      "loss": 0.2219,
      "step": 2320
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011890824336466068,
      "loss": 0.1018,
      "step": 2321
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011878949917143454,
      "loss": 0.5951,
      "step": 2322
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011867077541860866,
      "loss": 0.5285,
      "step": 2323
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011855207218393781,
      "loss": 0.4431,
      "step": 2324
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011843338954516306,
      "loss": 0.3335,
      "step": 2325
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011831472758001225,
      "loss": 0.2022,
      "step": 2326
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001181960863661994,
      "loss": 0.0745,
      "step": 2327
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011807746598142517,
      "loss": 0.3661,
      "step": 2328
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011795886650337652,
      "loss": 0.1984,
      "step": 2329
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011784028800972663,
      "loss": 0.2898,
      "step": 2330
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011772173057813508,
      "loss": 0.2251,
      "step": 2331
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011760319428624744,
      "loss": 0.7147,
      "step": 2332
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011748467921169573,
      "loss": 0.5833,
      "step": 2333
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011736618543209779,
      "loss": 0.5246,
      "step": 2334
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011724771302505775,
      "loss": 0.261,
      "step": 2335
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011712926206816548,
      "loss": 0.5643,
      "step": 2336
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011701083263899713,
      "loss": 0.1899,
      "step": 2337
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001168924248151144,
      "loss": 0.3154,
      "step": 2338
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011677403867406516,
      "loss": 0.9979,
      "step": 2339
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00011665567429338289,
      "loss": 0.3786,
      "step": 2340
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001165373317505868,
      "loss": 0.2875,
      "step": 2341
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011641901112318196,
      "loss": 0.3107,
      "step": 2342
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011630071248865887,
      "loss": 0.4624,
      "step": 2343
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011618243592449388,
      "loss": 0.1852,
      "step": 2344
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011606418150814862,
      "loss": 0.2376,
      "step": 2345
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011594594931707046,
      "loss": 0.3472,
      "step": 2346
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.000115827739428692,
      "loss": 0.3036,
      "step": 2347
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011570955192043137,
      "loss": 0.5238,
      "step": 2348
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011559138686969207,
      "loss": 0.3344,
      "step": 2349
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011547324435386272,
      "loss": 0.3679,
      "step": 2350
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001153551244503174,
      "loss": 0.5979,
      "step": 2351
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011523702723641513,
      "loss": 0.3886,
      "step": 2352
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011511895278950035,
      "loss": 0.2912,
      "step": 2353
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011500090118690235,
      "loss": 0.414,
      "step": 2354
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011488287250593564,
      "loss": 0.2956,
      "step": 2355
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011476486682389953,
      "loss": 0.5523,
      "step": 2356
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001146468842180785,
      "loss": 0.3276,
      "step": 2357
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011452892476574167,
      "loss": 0.6139,
      "step": 2358
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011441098854414319,
      "loss": 0.3394,
      "step": 2359
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011429307563052196,
      "loss": 0.1067,
      "step": 2360
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011417518610210143,
      "loss": 0.4179,
      "step": 2361
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011405732003609006,
      "loss": 0.2273,
      "step": 2362
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011393947750968061,
      "loss": 0.3588,
      "step": 2363
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011382165860005071,
      "loss": 0.2487,
      "step": 2364
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011370386338436229,
      "loss": 0.3008,
      "step": 2365
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011358609193976195,
      "loss": 0.5172,
      "step": 2366
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011346834434338054,
      "loss": 0.5595,
      "step": 2367
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011335062067233344,
      "loss": 0.4434,
      "step": 2368
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011323292100372038,
      "loss": 0.5253,
      "step": 2369
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011311524541462516,
      "loss": 0.4381,
      "step": 2370
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011299759398211607,
      "loss": 0.2314,
      "step": 2371
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011287996678324537,
      "loss": 0.4378,
      "step": 2372
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011276236389504963,
      "loss": 0.332,
      "step": 2373
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001126447853945493,
      "loss": 0.5438,
      "step": 2374
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011252723135874906,
      "loss": 0.4797,
      "step": 2375
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011240970186463737,
      "loss": 0.4306,
      "step": 2376
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011229219698918682,
      "loss": 0.1611,
      "step": 2377
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011217471680935372,
      "loss": 0.273,
      "step": 2378
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011205726140207823,
      "loss": 0.433,
      "step": 2379
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011193983084428442,
      "loss": 0.1893,
      "step": 2380
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011182242521287985,
      "loss": 0.4912,
      "step": 2381
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00011170504458475604,
      "loss": 0.2221,
      "step": 2382
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00011158768903678783,
      "loss": 0.3944,
      "step": 2383
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00011147035864583394,
      "loss": 0.2925,
      "step": 2384
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00011135305348873628,
      "loss": 0.1925,
      "step": 2385
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001112357736423206,
      "loss": 0.4363,
      "step": 2386
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00011111851918339574,
      "loss": 0.5678,
      "step": 2387
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00011100129018875419,
      "loss": 0.3665,
      "step": 2388
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00011088408673517151,
      "loss": 0.2847,
      "step": 2389
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00011076690889940672,
      "loss": 0.2369,
      "step": 2390
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00011064975675820206,
      "loss": 0.531,
      "step": 2391
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00011053263038828276,
      "loss": 0.2767,
      "step": 2392
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00011041552986635741,
      "loss": 0.2995,
      "step": 2393
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00011029845526911746,
      "loss": 0.2148,
      "step": 2394
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00011018140667323756,
      "loss": 0.2046,
      "step": 2395
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00011006438415537518,
      "loss": 0.1899,
      "step": 2396
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001099473877921708,
      "loss": 0.3579,
      "step": 2397
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010983041766024777,
      "loss": 0.2972,
      "step": 2398
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010971347383621218,
      "loss": 0.1417,
      "step": 2399
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010959655639665307,
      "loss": 0.2621,
      "step": 2400
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.36211344599723816,
      "eval_runtime": 53.0319,
      "eval_samples_per_second": 0.622,
      "eval_steps_per_second": 0.622,
      "step": 2400
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010947966541814193,
      "loss": 0.3843,
      "step": 2401
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010936280097723324,
      "loss": 0.1558,
      "step": 2402
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010924596315046376,
      "loss": 0.5637,
      "step": 2403
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010912915201435312,
      "loss": 0.2278,
      "step": 2404
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010901236764540326,
      "loss": 0.2319,
      "step": 2405
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010889561012009876,
      "loss": 0.2835,
      "step": 2406
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010877887951490639,
      "loss": 0.4242,
      "step": 2407
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010866217590627559,
      "loss": 0.5728,
      "step": 2408
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010854549937063782,
      "loss": 0.3694,
      "step": 2409
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.000108428849984407,
      "loss": 0.5046,
      "step": 2410
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001083122278239793,
      "loss": 0.3778,
      "step": 2411
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010819563296573282,
      "loss": 0.6938,
      "step": 2412
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001080790654860281,
      "loss": 0.1809,
      "step": 2413
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010796252546120749,
      "loss": 0.2239,
      "step": 2414
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010784601296759545,
      "loss": 0.1315,
      "step": 2415
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010772952808149844,
      "loss": 0.2919,
      "step": 2416
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010761307087920481,
      "loss": 0.3927,
      "step": 2417
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010749664143698477,
      "loss": 0.1915,
      "step": 2418
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010738023983109034,
      "loss": 0.4278,
      "step": 2419
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010726386613775542,
      "loss": 0.6921,
      "step": 2420
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010714752043319544,
      "loss": 0.2486,
      "step": 2421
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00010703120279360767,
      "loss": 0.3389,
      "step": 2422
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010691491329517085,
      "loss": 0.1818,
      "step": 2423
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001067986520140455,
      "loss": 0.4171,
      "step": 2424
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010668241902637338,
      "loss": 0.4456,
      "step": 2425
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010656621440827802,
      "loss": 0.546,
      "step": 2426
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010645003823586405,
      "loss": 0.2166,
      "step": 2427
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010633389058521781,
      "loss": 0.2131,
      "step": 2428
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010621777153240664,
      "loss": 0.3298,
      "step": 2429
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010610168115347937,
      "loss": 0.3514,
      "step": 2430
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010598561952446607,
      "loss": 0.4615,
      "step": 2431
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010586958672137775,
      "loss": 0.4107,
      "step": 2432
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010575358282020674,
      "loss": 0.2065,
      "step": 2433
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001056376078969264,
      "loss": 0.5974,
      "step": 2434
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010552166202749109,
      "loss": 0.0516,
      "step": 2435
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010540574528783615,
      "loss": 0.4994,
      "step": 2436
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010528985775387783,
      "loss": 0.3778,
      "step": 2437
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010517399950151326,
      "loss": 0.2844,
      "step": 2438
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010505817060662046,
      "loss": 0.2488,
      "step": 2439
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010494237114505802,
      "loss": 0.1728,
      "step": 2440
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010482660119266553,
      "loss": 0.3116,
      "step": 2441
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001047108608252631,
      "loss": 0.2741,
      "step": 2442
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010459515011865141,
      "loss": 0.3347,
      "step": 2443
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001044794691486119,
      "loss": 0.3154,
      "step": 2444
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010436381799090628,
      "loss": 0.2094,
      "step": 2445
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010424819672127703,
      "loss": 0.0549,
      "step": 2446
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010413260541544674,
      "loss": 0.4644,
      "step": 2447
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001040170441491187,
      "loss": 0.3996,
      "step": 2448
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001039015129979762,
      "loss": 0.1362,
      "step": 2449
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010378601203768312,
      "loss": 0.2433,
      "step": 2450
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010367054134388335,
      "loss": 0.225,
      "step": 2451
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010355510099220105,
      "loss": 0.6325,
      "step": 2452
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010343969105824048,
      "loss": 0.5253,
      "step": 2453
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010332431161758596,
      "loss": 0.2272,
      "step": 2454
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010320896274580188,
      "loss": 0.3783,
      "step": 2455
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010309364451843263,
      "loss": 0.2436,
      "step": 2456
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010297835701100246,
      "loss": 0.0893,
      "step": 2457
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010286310029901548,
      "loss": 0.8412,
      "step": 2458
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001027478744579558,
      "loss": 0.1901,
      "step": 2459
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001026326795632871,
      "loss": 0.0371,
      "step": 2460
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010251751569045291,
      "loss": 0.4089,
      "step": 2461
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00010240238291487652,
      "loss": 0.1723,
      "step": 2462
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010228728131196063,
      "loss": 0.1415,
      "step": 2463
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010217221095708775,
      "loss": 0.2344,
      "step": 2464
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010205717192561973,
      "loss": 0.635,
      "step": 2465
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001019421642928981,
      "loss": 0.3515,
      "step": 2466
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010182718813424364,
      "loss": 0.2854,
      "step": 2467
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010171224352495673,
      "loss": 0.3349,
      "step": 2468
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001015973305403168,
      "loss": 0.1743,
      "step": 2469
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010148244925558285,
      "loss": 0.3275,
      "step": 2470
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.000101367599745993,
      "loss": 0.3226,
      "step": 2471
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010125278208676449,
      "loss": 0.6259,
      "step": 2472
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010113799635309385,
      "loss": 0.768,
      "step": 2473
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010102324262015656,
      "loss": 0.6521,
      "step": 2474
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010090852096310728,
      "loss": 0.2122,
      "step": 2475
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010079383145707945,
      "loss": 0.3505,
      "step": 2476
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010067917417718577,
      "loss": 0.1767,
      "step": 2477
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001005645491985175,
      "loss": 0.6048,
      "step": 2478
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010044995659614502,
      "loss": 0.6865,
      "step": 2479
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010033539644511726,
      "loss": 0.187,
      "step": 2480
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010022086882046213,
      "loss": 0.5134,
      "step": 2481
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00010010637379718615,
      "loss": 0.4152,
      "step": 2482
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.999191145027437e-05,
      "loss": 0.3805,
      "step": 2483
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.987748185469065e-05,
      "loss": 0.2853,
      "step": 2484
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.976308508537715e-05,
      "loss": 0.4959,
      "step": 2485
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.964872121725484e-05,
      "loss": 0.178,
      "step": 2486
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.953439032522283e-05,
      "loss": 0.1983,
      "step": 2487
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.94200924841589e-05,
      "loss": 0.3493,
      "step": 2488
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.930582776891893e-05,
      "loss": 0.0972,
      "step": 2489
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.919159625433735e-05,
      "loss": 0.4195,
      "step": 2490
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.90773980152267e-05,
      "loss": 0.1948,
      "step": 2491
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.896323312637772e-05,
      "loss": 0.184,
      "step": 2492
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.884910166255942e-05,
      "loss": 0.5226,
      "step": 2493
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.873500369851877e-05,
      "loss": 0.4602,
      "step": 2494
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.862093930898097e-05,
      "loss": 0.1207,
      "step": 2495
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.850690856864902e-05,
      "loss": 0.307,
      "step": 2496
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.839291155220415e-05,
      "loss": 0.4408,
      "step": 2497
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.827894833430519e-05,
      "loss": 0.1983,
      "step": 2498
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.816501898958915e-05,
      "loss": 0.5028,
      "step": 2499
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.805112359267059e-05,
      "loss": 0.4045,
      "step": 2500
    },
    {
      "epoch": 0.62,
      "eval_loss": 0.35865071415901184,
      "eval_runtime": 50.7679,
      "eval_samples_per_second": 0.65,
      "eval_steps_per_second": 0.65,
      "step": 2500
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.793726221814201e-05,
      "loss": 0.2938,
      "step": 2501
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.782343494057362e-05,
      "loss": 0.3881,
      "step": 2502
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.770964183451316e-05,
      "loss": 0.4172,
      "step": 2503
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.759588297448616e-05,
      "loss": 0.4163,
      "step": 2504
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.748215843499554e-05,
      "loss": 0.245,
      "step": 2505
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.736846829052195e-05,
      "loss": 0.5666,
      "step": 2506
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.725481261552333e-05,
      "loss": 0.191,
      "step": 2507
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.714119148443521e-05,
      "loss": 0.3555,
      "step": 2508
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.70276049716703e-05,
      "loss": 0.309,
      "step": 2509
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.691405315161888e-05,
      "loss": 0.22,
      "step": 2510
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.680053609864826e-05,
      "loss": 0.3954,
      "step": 2511
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.668705388710313e-05,
      "loss": 0.4286,
      "step": 2512
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.657360659130538e-05,
      "loss": 0.1586,
      "step": 2513
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.646019428555391e-05,
      "loss": 0.6813,
      "step": 2514
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.634681704412486e-05,
      "loss": 0.393,
      "step": 2515
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.623347494127122e-05,
      "loss": 0.4039,
      "step": 2516
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.61201680512232e-05,
      "loss": 0.4256,
      "step": 2517
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.600689644818767e-05,
      "loss": 0.169,
      "step": 2518
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.589366020634867e-05,
      "loss": 0.1448,
      "step": 2519
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.578045939986682e-05,
      "loss": 0.4271,
      "step": 2520
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.566729410287975e-05,
      "loss": 0.6961,
      "step": 2521
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.555416438950177e-05,
      "loss": 0.6398,
      "step": 2522
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.544107033382375e-05,
      "loss": 0.1739,
      "step": 2523
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.532801200991347e-05,
      "loss": 0.1497,
      "step": 2524
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.5214989491815e-05,
      "loss": 0.5515,
      "step": 2525
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.510200285354925e-05,
      "loss": 0.4066,
      "step": 2526
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.498905216911338e-05,
      "loss": 0.4774,
      "step": 2527
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.487613751248124e-05,
      "loss": 0.2824,
      "step": 2528
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.476325895760288e-05,
      "loss": 0.7531,
      "step": 2529
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.46504165784048e-05,
      "loss": 0.2724,
      "step": 2530
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.453761044878983e-05,
      "loss": 0.5,
      "step": 2531
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.442484064263699e-05,
      "loss": 0.3075,
      "step": 2532
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.431210723380165e-05,
      "loss": 0.3713,
      "step": 2533
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.419941029611509e-05,
      "loss": 0.3338,
      "step": 2534
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.408674990338501e-05,
      "loss": 0.1648,
      "step": 2535
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.397412612939488e-05,
      "loss": 0.4339,
      "step": 2536
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.38615390479045e-05,
      "loss": 0.2412,
      "step": 2537
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.374898873264929e-05,
      "loss": 0.2839,
      "step": 2538
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.363647525734091e-05,
      "loss": 0.3562,
      "step": 2539
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.352399869566663e-05,
      "loss": 0.7551,
      "step": 2540
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.341155912128977e-05,
      "loss": 0.34,
      "step": 2541
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.329915660784932e-05,
      "loss": 0.2662,
      "step": 2542
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.318679122895994e-05,
      "loss": 0.38,
      "step": 2543
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.307446305821213e-05,
      "loss": 0.1514,
      "step": 2544
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.296217216917178e-05,
      "loss": 0.595,
      "step": 2545
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.284991863538069e-05,
      "loss": 0.2616,
      "step": 2546
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.273770253035583e-05,
      "loss": 0.3657,
      "step": 2547
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.262552392758997e-05,
      "loss": 0.4408,
      "step": 2548
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.251338290055114e-05,
      "loss": 0.1019,
      "step": 2549
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.24012795226828e-05,
      "loss": 0.3804,
      "step": 2550
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.228921386740376e-05,
      "loss": 0.5453,
      "step": 2551
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.217718600810816e-05,
      "loss": 0.4193,
      "step": 2552
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.206519601816538e-05,
      "loss": 0.2478,
      "step": 2553
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.195324397091986e-05,
      "loss": 0.4661,
      "step": 2554
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.184132993969144e-05,
      "loss": 0.3336,
      "step": 2555
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.172945399777481e-05,
      "loss": 0.4816,
      "step": 2556
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.161761621843994e-05,
      "loss": 0.4228,
      "step": 2557
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.150581667493155e-05,
      "loss": 0.3053,
      "step": 2558
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.139405544046967e-05,
      "loss": 0.4985,
      "step": 2559
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.128233258824882e-05,
      "loss": 0.321,
      "step": 2560
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.11706481914387e-05,
      "loss": 0.5084,
      "step": 2561
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.10590023231838e-05,
      "loss": 0.7517,
      "step": 2562
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.094739505660315e-05,
      "loss": 0.2226,
      "step": 2563
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.083582646479079e-05,
      "loss": 0.4708,
      "step": 2564
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.07242966208152e-05,
      "loss": 0.456,
      "step": 2565
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.061280559771959e-05,
      "loss": 0.5026,
      "step": 2566
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.050135346852172e-05,
      "loss": 0.4059,
      "step": 2567
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.038994030621392e-05,
      "loss": 0.5108,
      "step": 2568
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.027856618376295e-05,
      "loss": 0.4996,
      "step": 2569
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.016723117411003e-05,
      "loss": 0.2691,
      "step": 2570
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.005593535017072e-05,
      "loss": 0.3969,
      "step": 2571
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.994467878483494e-05,
      "loss": 0.3399,
      "step": 2572
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.983346155096702e-05,
      "loss": 0.2622,
      "step": 2573
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.972228372140525e-05,
      "loss": 0.3667,
      "step": 2574
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.961114536896245e-05,
      "loss": 0.2055,
      "step": 2575
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.950004656642527e-05,
      "loss": 0.1418,
      "step": 2576
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.938898738655473e-05,
      "loss": 0.2727,
      "step": 2577
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.927796790208566e-05,
      "loss": 0.4459,
      "step": 2578
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.916698818572713e-05,
      "loss": 0.1869,
      "step": 2579
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.905604831016192e-05,
      "loss": 0.441,
      "step": 2580
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.894514834804696e-05,
      "loss": 0.3452,
      "step": 2581
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.883428837201283e-05,
      "loss": 0.3008,
      "step": 2582
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.872346845466405e-05,
      "loss": 0.2955,
      "step": 2583
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.861268866857887e-05,
      "loss": 0.5521,
      "step": 2584
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.850194908630926e-05,
      "loss": 0.2162,
      "step": 2585
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.839124978038083e-05,
      "loss": 0.2485,
      "step": 2586
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.828059082329288e-05,
      "loss": 0.2789,
      "step": 2587
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.81699722875182e-05,
      "loss": 0.3044,
      "step": 2588
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.805939424550315e-05,
      "loss": 0.345,
      "step": 2589
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.794885676966764e-05,
      "loss": 0.4423,
      "step": 2590
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.783835993240481e-05,
      "loss": 0.3547,
      "step": 2591
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.772790380608142e-05,
      "loss": 0.6588,
      "step": 2592
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.761748846303747e-05,
      "loss": 0.3779,
      "step": 2593
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.750711397558614e-05,
      "loss": 0.6003,
      "step": 2594
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.739678041601411e-05,
      "loss": 0.2617,
      "step": 2595
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.728648785658094e-05,
      "loss": 0.2763,
      "step": 2596
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.717623636951967e-05,
      "loss": 0.4156,
      "step": 2597
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.70660260270361e-05,
      "loss": 0.2314,
      "step": 2598
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.695585690130939e-05,
      "loss": 0.3591,
      "step": 2599
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.684572906449146e-05,
      "loss": 0.5619,
      "step": 2600
    },
    {
      "epoch": 0.65,
      "eval_loss": 0.352405309677124,
      "eval_runtime": 49.937,
      "eval_samples_per_second": 0.661,
      "eval_steps_per_second": 0.661,
      "step": 2600
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.673564258870744e-05,
      "loss": 0.315,
      "step": 2601
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.662559754605508e-05,
      "loss": 0.3861,
      "step": 2602
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.65155940086052e-05,
      "loss": 0.3392,
      "step": 2603
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.640563204840149e-05,
      "loss": 0.2844,
      "step": 2604
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.629571173746013e-05,
      "loss": 0.3547,
      "step": 2605
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.618583314777037e-05,
      "loss": 0.5509,
      "step": 2606
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.607599635129378e-05,
      "loss": 0.25,
      "step": 2607
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.596620141996493e-05,
      "loss": 0.5131,
      "step": 2608
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.585644842569062e-05,
      "loss": 0.2194,
      "step": 2609
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.574673744035047e-05,
      "loss": 0.481,
      "step": 2610
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.563706853579638e-05,
      "loss": 0.2963,
      "step": 2611
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.552744178385285e-05,
      "loss": 0.5477,
      "step": 2612
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.541785725631662e-05,
      "loss": 0.5378,
      "step": 2613
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.530831502495696e-05,
      "loss": 0.4539,
      "step": 2614
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.51988151615152e-05,
      "loss": 0.6642,
      "step": 2615
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.508935773770519e-05,
      "loss": 0.481,
      "step": 2616
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.49799428252128e-05,
      "loss": 0.2701,
      "step": 2617
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.487057049569606e-05,
      "loss": 0.2018,
      "step": 2618
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.476124082078524e-05,
      "loss": 0.4708,
      "step": 2619
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.465195387208252e-05,
      "loss": 0.4334,
      "step": 2620
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.45427097211623e-05,
      "loss": 0.2932,
      "step": 2621
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.443350843957065e-05,
      "loss": 0.3234,
      "step": 2622
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.432435009882585e-05,
      "loss": 0.2941,
      "step": 2623
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.421523477041801e-05,
      "loss": 0.1267,
      "step": 2624
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.410616252580888e-05,
      "loss": 0.3252,
      "step": 2625
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.399713343643222e-05,
      "loss": 0.3957,
      "step": 2626
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.388814757369333e-05,
      "loss": 0.1188,
      "step": 2627
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.377920500896942e-05,
      "loss": 0.3057,
      "step": 2628
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.367030581360907e-05,
      "loss": 0.2436,
      "step": 2629
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.356145005893276e-05,
      "loss": 0.5215,
      "step": 2630
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.345263781623224e-05,
      "loss": 0.3157,
      "step": 2631
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.334386915677101e-05,
      "loss": 0.1642,
      "step": 2632
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.323514415178378e-05,
      "loss": 0.4916,
      "step": 2633
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.312646287247686e-05,
      "loss": 0.4885,
      "step": 2634
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.301782539002793e-05,
      "loss": 0.6675,
      "step": 2635
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.290923177558577e-05,
      "loss": 0.718,
      "step": 2636
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.280068210027074e-05,
      "loss": 0.7176,
      "step": 2637
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.269217643517412e-05,
      "loss": 0.2575,
      "step": 2638
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.25837148513586e-05,
      "loss": 0.3357,
      "step": 2639
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.247529741985781e-05,
      "loss": 0.4354,
      "step": 2640
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.236692421167669e-05,
      "loss": 0.4282,
      "step": 2641
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.225859529779096e-05,
      "loss": 0.2523,
      "step": 2642
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.215031074914752e-05,
      "loss": 0.2786,
      "step": 2643
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.204207063666419e-05,
      "loss": 0.3111,
      "step": 2644
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.193387503122955e-05,
      "loss": 0.1468,
      "step": 2645
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.182572400370326e-05,
      "loss": 0.2534,
      "step": 2646
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.171761762491554e-05,
      "loss": 0.218,
      "step": 2647
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.16095559656676e-05,
      "loss": 0.2776,
      "step": 2648
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.15015390967311e-05,
      "loss": 0.6669,
      "step": 2649
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.139356708884866e-05,
      "loss": 0.3167,
      "step": 2650
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.128564001273327e-05,
      "loss": 0.2533,
      "step": 2651
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.117775793906868e-05,
      "loss": 0.5061,
      "step": 2652
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.1069920938509e-05,
      "loss": 0.1727,
      "step": 2653
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.0962129081679e-05,
      "loss": 0.2212,
      "step": 2654
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.085438243917371e-05,
      "loss": 0.4855,
      "step": 2655
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.074668108155861e-05,
      "loss": 0.2648,
      "step": 2656
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.063902507936965e-05,
      "loss": 0.4539,
      "step": 2657
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.053141450311283e-05,
      "loss": 0.5305,
      "step": 2658
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.042384942326468e-05,
      "loss": 0.5308,
      "step": 2659
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.031632991027161e-05,
      "loss": 0.464,
      "step": 2660
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.020885603455055e-05,
      "loss": 0.4029,
      "step": 2661
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.010142786648821e-05,
      "loss": 0.2918,
      "step": 2662
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.999404547644156e-05,
      "loss": 0.291,
      "step": 2663
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.988670893473763e-05,
      "loss": 0.8324,
      "step": 2664
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.977941831167319e-05,
      "loss": 0.4328,
      "step": 2665
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.967217367751519e-05,
      "loss": 0.3979,
      "step": 2666
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.956497510250026e-05,
      "loss": 0.3373,
      "step": 2667
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.945782265683506e-05,
      "loss": 0.1445,
      "step": 2668
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.935071641069585e-05,
      "loss": 0.5057,
      "step": 2669
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.924365643422879e-05,
      "loss": 0.0847,
      "step": 2670
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.913664279754956e-05,
      "loss": 0.6502,
      "step": 2671
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.902967557074372e-05,
      "loss": 0.355,
      "step": 2672
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.892275482386621e-05,
      "loss": 0.4264,
      "step": 2673
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.881588062694167e-05,
      "loss": 0.5212,
      "step": 2674
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.870905304996426e-05,
      "loss": 0.5083,
      "step": 2675
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.860227216289747e-05,
      "loss": 1.0383,
      "step": 2676
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.849553803567442e-05,
      "loss": 0.2518,
      "step": 2677
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.838885073819736e-05,
      "loss": 0.4572,
      "step": 2678
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.828221034033813e-05,
      "loss": 0.2327,
      "step": 2679
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.817561691193761e-05,
      "loss": 0.378,
      "step": 2680
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.806907052280612e-05,
      "loss": 0.5727,
      "step": 2681
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.7962571242723e-05,
      "loss": 0.2929,
      "step": 2682
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.785611914143686e-05,
      "loss": 0.2975,
      "step": 2683
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.774971428866546e-05,
      "loss": 0.4444,
      "step": 2684
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.764335675409542e-05,
      "loss": 0.2188,
      "step": 2685
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.753704660738256e-05,
      "loss": 0.3889,
      "step": 2686
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.743078391815148e-05,
      "loss": 0.3777,
      "step": 2687
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.732456875599596e-05,
      "loss": 0.4338,
      "step": 2688
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.721840119047835e-05,
      "loss": 0.7659,
      "step": 2689
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.711228129113008e-05,
      "loss": 0.5396,
      "step": 2690
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.700620912745124e-05,
      "loss": 0.2547,
      "step": 2691
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.690018476891057e-05,
      "loss": 0.4139,
      "step": 2692
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.679420828494578e-05,
      "loss": 0.4261,
      "step": 2693
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.66882797449629e-05,
      "loss": 0.2251,
      "step": 2694
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.658239921833682e-05,
      "loss": 0.3025,
      "step": 2695
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.64765667744108e-05,
      "loss": 0.5291,
      "step": 2696
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.637078248249681e-05,
      "loss": 0.2185,
      "step": 2697
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.6265046411875e-05,
      "loss": 0.2167,
      "step": 2698
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.615935863179425e-05,
      "loss": 0.1151,
      "step": 2699
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.605371921147155e-05,
      "loss": 0.4347,
      "step": 2700
    },
    {
      "epoch": 0.67,
      "eval_loss": 0.3495098352432251,
      "eval_runtime": 50.2833,
      "eval_samples_per_second": 0.656,
      "eval_steps_per_second": 0.656,
      "step": 2700
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.594812822009243e-05,
      "loss": 0.2346,
      "step": 2701
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.584258572681052e-05,
      "loss": 0.2686,
      "step": 2702
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.573709180074787e-05,
      "loss": 0.3596,
      "step": 2703
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.56316465109945e-05,
      "loss": 0.1207,
      "step": 2704
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.55262499266088e-05,
      "loss": 0.5589,
      "step": 2705
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.542090211661719e-05,
      "loss": 0.6401,
      "step": 2706
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.531560315001403e-05,
      "loss": 0.4527,
      "step": 2707
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.521035309576186e-05,
      "loss": 0.7128,
      "step": 2708
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.510515202279104e-05,
      "loss": 0.3836,
      "step": 2709
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.500000000000002e-05,
      "loss": 0.2949,
      "step": 2710
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.489489709625491e-05,
      "loss": 0.3762,
      "step": 2711
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.478984338038989e-05,
      "loss": 0.1845,
      "step": 2712
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.468483892120666e-05,
      "loss": 0.4817,
      "step": 2713
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.457988378747489e-05,
      "loss": 0.5591,
      "step": 2714
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.447497804793191e-05,
      "loss": 0.13,
      "step": 2715
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.437012177128256e-05,
      "loss": 0.4363,
      "step": 2716
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.426531502619946e-05,
      "loss": 0.3317,
      "step": 2717
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.41605578813226e-05,
      "loss": 0.4063,
      "step": 2718
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.405585040525972e-05,
      "loss": 0.4156,
      "step": 2719
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.395119266658577e-05,
      "loss": 0.353,
      "step": 2720
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.384658473384337e-05,
      "loss": 0.1432,
      "step": 2721
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.374202667554234e-05,
      "loss": 0.2976,
      "step": 2722
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.363751856015998e-05,
      "loss": 0.3048,
      "step": 2723
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.353306045614073e-05,
      "loss": 0.2642,
      "step": 2724
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.342865243189638e-05,
      "loss": 0.1577,
      "step": 2725
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.332429455580597e-05,
      "loss": 0.616,
      "step": 2726
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.321998689621557e-05,
      "loss": 0.3826,
      "step": 2727
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.311572952143834e-05,
      "loss": 0.3308,
      "step": 2728
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.301152249975472e-05,
      "loss": 0.332,
      "step": 2729
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.290736589941191e-05,
      "loss": 0.3218,
      "step": 2730
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.280325978862433e-05,
      "loss": 0.7182,
      "step": 2731
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.26992042355731e-05,
      "loss": 0.3459,
      "step": 2732
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.259519930840648e-05,
      "loss": 0.4745,
      "step": 2733
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.249124507523931e-05,
      "loss": 0.5397,
      "step": 2734
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.238734160415348e-05,
      "loss": 0.3348,
      "step": 2735
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.228348896319739e-05,
      "loss": 0.4047,
      "step": 2736
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.21796872203864e-05,
      "loss": 0.5199,
      "step": 2737
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.20759364437023e-05,
      "loss": 0.2114,
      "step": 2738
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.197223670109371e-05,
      "loss": 0.3586,
      "step": 2739
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.186858806047564e-05,
      "loss": 0.3306,
      "step": 2740
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.176499058972982e-05,
      "loss": 0.3314,
      "step": 2741
    },
    {
      "epoch": 0.68,
      "learning_rate": 7.166144435670425e-05,
      "loss": 0.3913,
      "step": 2742
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.155794942921361e-05,
      "loss": 0.2751,
      "step": 2743
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.145450587503876e-05,
      "loss": 0.3848,
      "step": 2744
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.135111376192706e-05,
      "loss": 0.2971,
      "step": 2745
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.124777315759219e-05,
      "loss": 0.1874,
      "step": 2746
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.114448412971393e-05,
      "loss": 0.3316,
      "step": 2747
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.104124674593851e-05,
      "loss": 0.3529,
      "step": 2748
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.093806107387814e-05,
      "loss": 0.37,
      "step": 2749
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.08349271811113e-05,
      "loss": 0.1376,
      "step": 2750
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.07318451351824e-05,
      "loss": 0.2498,
      "step": 2751
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.062881500360215e-05,
      "loss": 0.4008,
      "step": 2752
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.052583685384695e-05,
      "loss": 0.502,
      "step": 2753
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.042291075335946e-05,
      "loss": 0.3136,
      "step": 2754
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.032003676954797e-05,
      "loss": 0.3431,
      "step": 2755
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.021721496978683e-05,
      "loss": 0.1851,
      "step": 2756
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.011444542141621e-05,
      "loss": 0.1827,
      "step": 2757
    },
    {
      "epoch": 0.69,
      "learning_rate": 7.001172819174192e-05,
      "loss": 0.5294,
      "step": 2758
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.990906334803566e-05,
      "loss": 0.1861,
      "step": 2759
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.980645095753469e-05,
      "loss": 0.3975,
      "step": 2760
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.970389108744205e-05,
      "loss": 0.3579,
      "step": 2761
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.960138380492622e-05,
      "loss": 0.5619,
      "step": 2762
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.949892917712143e-05,
      "loss": 0.3679,
      "step": 2763
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.939652727112724e-05,
      "loss": 0.2277,
      "step": 2764
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.929417815400886e-05,
      "loss": 0.2074,
      "step": 2765
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.919188189279674e-05,
      "loss": 0.3879,
      "step": 2766
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.90896385544869e-05,
      "loss": 0.1318,
      "step": 2767
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.898744820604054e-05,
      "loss": 0.311,
      "step": 2768
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.888531091438429e-05,
      "loss": 0.3328,
      "step": 2769
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.878322674640993e-05,
      "loss": 0.3683,
      "step": 2770
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.868119576897442e-05,
      "loss": 0.2328,
      "step": 2771
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.85792180489001e-05,
      "loss": 0.2664,
      "step": 2772
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.847729365297413e-05,
      "loss": 0.1621,
      "step": 2773
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.837542264794901e-05,
      "loss": 0.6488,
      "step": 2774
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.827360510054208e-05,
      "loss": 0.2318,
      "step": 2775
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.817184107743579e-05,
      "loss": 0.2788,
      "step": 2776
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.807013064527753e-05,
      "loss": 0.1279,
      "step": 2777
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.796847387067948e-05,
      "loss": 0.5097,
      "step": 2778
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.786687082021885e-05,
      "loss": 0.3581,
      "step": 2779
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.776532156043747e-05,
      "loss": 0.4341,
      "step": 2780
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.766382615784215e-05,
      "loss": 0.6011,
      "step": 2781
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.756238467890421e-05,
      "loss": 0.3886,
      "step": 2782
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.746099719005986e-05,
      "loss": 0.1848,
      "step": 2783
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.73596637577098e-05,
      "loss": 0.5395,
      "step": 2784
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.725838444821938e-05,
      "loss": 0.4367,
      "step": 2785
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.715715932791858e-05,
      "loss": 0.4655,
      "step": 2786
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.705598846310172e-05,
      "loss": 0.4869,
      "step": 2787
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.695487192002779e-05,
      "loss": 0.1799,
      "step": 2788
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.685380976491999e-05,
      "loss": 0.2085,
      "step": 2789
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.67528020639661e-05,
      "loss": 0.6232,
      "step": 2790
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.665184888331806e-05,
      "loss": 0.1888,
      "step": 2791
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.65509502890923e-05,
      "loss": 0.4444,
      "step": 2792
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.645010634736923e-05,
      "loss": 0.1634,
      "step": 2793
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.634931712419379e-05,
      "loss": 0.1801,
      "step": 2794
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.624858268557482e-05,
      "loss": 0.3086,
      "step": 2795
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.614790309748537e-05,
      "loss": 0.6483,
      "step": 2796
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.604727842586269e-05,
      "loss": 0.2298,
      "step": 2797
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.594670873660784e-05,
      "loss": 0.184,
      "step": 2798
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.584619409558605e-05,
      "loss": 0.3284,
      "step": 2799
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.574573456862636e-05,
      "loss": 0.2572,
      "step": 2800
    },
    {
      "epoch": 0.7,
      "eval_loss": 0.3451196849346161,
      "eval_runtime": 52.7717,
      "eval_samples_per_second": 0.625,
      "eval_steps_per_second": 0.625,
      "step": 2800
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.564533022152187e-05,
      "loss": 0.4263,
      "step": 2801
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.554498112002937e-05,
      "loss": 0.4696,
      "step": 2802
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.544468732986963e-05,
      "loss": 0.3249,
      "step": 2803
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.534444891672704e-05,
      "loss": 0.2263,
      "step": 2804
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.524426594624991e-05,
      "loss": 0.1342,
      "step": 2805
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.514413848405009e-05,
      "loss": 0.3293,
      "step": 2806
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.504406659570304e-05,
      "loss": 0.3315,
      "step": 2807
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.4944050346748e-05,
      "loss": 0.3739,
      "step": 2808
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.484408980268762e-05,
      "loss": 0.3231,
      "step": 2809
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.474418502898815e-05,
      "loss": 0.251,
      "step": 2810
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.464433609107924e-05,
      "loss": 0.1552,
      "step": 2811
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.454454305435411e-05,
      "loss": 0.292,
      "step": 2812
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.444480598416916e-05,
      "loss": 0.4157,
      "step": 2813
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.434512494584436e-05,
      "loss": 0.2494,
      "step": 2814
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.424550000466279e-05,
      "loss": 0.1125,
      "step": 2815
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.414593122587089e-05,
      "loss": 0.5786,
      "step": 2816
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.40464186746784e-05,
      "loss": 0.5468,
      "step": 2817
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.394696241625803e-05,
      "loss": 0.1832,
      "step": 2818
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.384756251574582e-05,
      "loss": 0.4935,
      "step": 2819
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.374821903824071e-05,
      "loss": 0.4367,
      "step": 2820
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.36489320488049e-05,
      "loss": 0.4523,
      "step": 2821
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.354970161246341e-05,
      "loss": 0.4066,
      "step": 2822
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.345052779420436e-05,
      "loss": 0.557,
      "step": 2823
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.335141065897863e-05,
      "loss": 0.3877,
      "step": 2824
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.325235027170022e-05,
      "loss": 0.112,
      "step": 2825
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.315334669724567e-05,
      "loss": 0.4763,
      "step": 2826
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.305440000045451e-05,
      "loss": 0.2953,
      "step": 2827
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.295551024612908e-05,
      "loss": 0.5123,
      "step": 2828
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.285667749903415e-05,
      "loss": 0.6312,
      "step": 2829
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.275790182389747e-05,
      "loss": 0.3447,
      "step": 2830
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.265918328540913e-05,
      "loss": 0.5976,
      "step": 2831
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.256052194822208e-05,
      "loss": 0.3187,
      "step": 2832
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.246191787695151e-05,
      "loss": 0.6069,
      "step": 2833
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.23633711361754e-05,
      "loss": 0.4366,
      "step": 2834
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.226488179043391e-05,
      "loss": 0.4463,
      "step": 2835
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.216644990422984e-05,
      "loss": 0.3168,
      "step": 2836
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.206807554202827e-05,
      "loss": 0.4888,
      "step": 2837
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.19697587682565e-05,
      "loss": 0.3124,
      "step": 2838
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.187149964730433e-05,
      "loss": 0.424,
      "step": 2839
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.177329824352357e-05,
      "loss": 0.3605,
      "step": 2840
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.167515462122845e-05,
      "loss": 0.3888,
      "step": 2841
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.157706884469521e-05,
      "loss": 0.5026,
      "step": 2842
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.147904097816217e-05,
      "loss": 0.2509,
      "step": 2843
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.138107108582994e-05,
      "loss": 0.345,
      "step": 2844
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.128315923186087e-05,
      "loss": 0.1417,
      "step": 2845
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.11853054803796e-05,
      "loss": 0.1612,
      "step": 2846
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.108750989547245e-05,
      "loss": 0.1631,
      "step": 2847
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.0989772541187856e-05,
      "loss": 0.3651,
      "step": 2848
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.089209348153594e-05,
      "loss": 0.2931,
      "step": 2849
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.079447278048883e-05,
      "loss": 0.4685,
      "step": 2850
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.0696910501980204e-05,
      "loss": 0.2449,
      "step": 2851
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.0599406709905765e-05,
      "loss": 0.3639,
      "step": 2852
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.0501961468122615e-05,
      "loss": 0.2271,
      "step": 2853
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.040457484044977e-05,
      "loss": 0.5398,
      "step": 2854
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.030724689066762e-05,
      "loss": 0.2497,
      "step": 2855
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.020997768251831e-05,
      "loss": 0.3145,
      "step": 2856
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.011276727970552e-05,
      "loss": 0.3546,
      "step": 2857
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.00156157458942e-05,
      "loss": 0.4206,
      "step": 2858
    },
    {
      "epoch": 0.71,
      "learning_rate": 5.991852314471103e-05,
      "loss": 0.3885,
      "step": 2859
    },
    {
      "epoch": 0.71,
      "learning_rate": 5.982148953974385e-05,
      "loss": 0.3095,
      "step": 2860
    },
    {
      "epoch": 0.71,
      "learning_rate": 5.9724514994542047e-05,
      "loss": 0.2586,
      "step": 2861
    },
    {
      "epoch": 0.71,
      "learning_rate": 5.9627599572616175e-05,
      "loss": 0.2949,
      "step": 2862
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.953074333743822e-05,
      "loss": 0.4151,
      "step": 2863
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.943394635244124e-05,
      "loss": 0.1669,
      "step": 2864
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.933720868101967e-05,
      "loss": 0.4466,
      "step": 2865
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.9240530386528915e-05,
      "loss": 0.23,
      "step": 2866
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.914391153228561e-05,
      "loss": 0.4195,
      "step": 2867
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.9047352181567525e-05,
      "loss": 0.4519,
      "step": 2868
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.8950852397613224e-05,
      "loss": 0.3325,
      "step": 2869
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.885441224362254e-05,
      "loss": 0.278,
      "step": 2870
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.8758031782756e-05,
      "loss": 0.2584,
      "step": 2871
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.866171107813524e-05,
      "loss": 0.3504,
      "step": 2872
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.856545019284261e-05,
      "loss": 0.5918,
      "step": 2873
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.846924918992142e-05,
      "loss": 0.5698,
      "step": 2874
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.8373108132375604e-05,
      "loss": 0.3672,
      "step": 2875
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.8277027083169974e-05,
      "loss": 0.2853,
      "step": 2876
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.818100610523003e-05,
      "loss": 0.4514,
      "step": 2877
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.8085045261441835e-05,
      "loss": 0.5099,
      "step": 2878
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.7989144614652085e-05,
      "loss": 0.2228,
      "step": 2879
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.789330422766819e-05,
      "loss": 0.1896,
      "step": 2880
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.7797524163257905e-05,
      "loss": 0.4912,
      "step": 2881
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.770180448414964e-05,
      "loss": 0.3282,
      "step": 2882
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.7606145253032106e-05,
      "loss": 0.3905,
      "step": 2883
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.75105465325546e-05,
      "loss": 0.7096,
      "step": 2884
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.741500838532665e-05,
      "loss": 0.2518,
      "step": 2885
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.731953087391809e-05,
      "loss": 0.1942,
      "step": 2886
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.722411406085919e-05,
      "loss": 0.1405,
      "step": 2887
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.7128758008640406e-05,
      "loss": 0.5084,
      "step": 2888
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.703346277971231e-05,
      "loss": 0.2552,
      "step": 2889
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.69382284364858e-05,
      "loss": 0.3602,
      "step": 2890
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.6843055041331706e-05,
      "loss": 0.3632,
      "step": 2891
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.674794265658115e-05,
      "loss": 0.3854,
      "step": 2892
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.6652891344525105e-05,
      "loss": 0.0836,
      "step": 2893
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.65579011674147e-05,
      "loss": 0.5631,
      "step": 2894
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.64629721874609e-05,
      "loss": 0.6262,
      "step": 2895
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.636810446683473e-05,
      "loss": 0.379,
      "step": 2896
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.627329806766691e-05,
      "loss": 0.5156,
      "step": 2897
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.617855305204818e-05,
      "loss": 0.5273,
      "step": 2898
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.6083869482029044e-05,
      "loss": 0.4746,
      "step": 2899
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.598924741961959e-05,
      "loss": 0.0925,
      "step": 2900
    },
    {
      "epoch": 0.72,
      "eval_loss": 0.3423108458518982,
      "eval_runtime": 51.9119,
      "eval_samples_per_second": 0.636,
      "eval_steps_per_second": 0.636,
      "step": 2900
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.589468692678991e-05,
      "loss": 0.5282,
      "step": 2901
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.5800188065469476e-05,
      "loss": 0.3087,
      "step": 2902
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.570575089754768e-05,
      "loss": 0.4706,
      "step": 2903
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.5611375484873236e-05,
      "loss": 0.2056,
      "step": 2904
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.551706188925466e-05,
      "loss": 0.4807,
      "step": 2905
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.542281017245977e-05,
      "loss": 0.6144,
      "step": 2906
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.532862039621602e-05,
      "loss": 0.3112,
      "step": 2907
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.5234492622210256e-05,
      "loss": 0.2591,
      "step": 2908
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.514042691208863e-05,
      "loss": 0.7438,
      "step": 2909
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.504642332745679e-05,
      "loss": 0.2045,
      "step": 2910
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.495248192987954e-05,
      "loss": 0.5341,
      "step": 2911
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.4858602780881127e-05,
      "loss": 0.2325,
      "step": 2912
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.4764785941944846e-05,
      "loss": 0.5199,
      "step": 2913
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.467103147451337e-05,
      "loss": 0.5568,
      "step": 2914
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.457733943998834e-05,
      "loss": 0.5053,
      "step": 2915
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.4483709899730695e-05,
      "loss": 0.2579,
      "step": 2916
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.439014291506025e-05,
      "loss": 0.3331,
      "step": 2917
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.429663854725607e-05,
      "loss": 0.1383,
      "step": 2918
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.420319685755598e-05,
      "loss": 0.3012,
      "step": 2919
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.410981790715696e-05,
      "loss": 0.2321,
      "step": 2920
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.401650175721471e-05,
      "loss": 0.4334,
      "step": 2921
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.392324846884402e-05,
      "loss": 0.7976,
      "step": 2922
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.383005810311833e-05,
      "loss": 0.5114,
      "step": 2923
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.373693072106988e-05,
      "loss": 0.3937,
      "step": 2924
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.3643866383689794e-05,
      "loss": 0.4545,
      "step": 2925
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.355086515192776e-05,
      "loss": 0.2561,
      "step": 2926
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.345792708669221e-05,
      "loss": 0.6256,
      "step": 2927
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.336505224885031e-05,
      "loss": 0.3839,
      "step": 2928
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.327224069922755e-05,
      "loss": 0.107,
      "step": 2929
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.317949249860824e-05,
      "loss": 0.248,
      "step": 2930
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.3086807707735e-05,
      "loss": 0.5059,
      "step": 2931
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.29941863873091e-05,
      "loss": 0.4084,
      "step": 2932
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.290162859799007e-05,
      "loss": 0.1477,
      "step": 2933
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.280913440039598e-05,
      "loss": 0.2264,
      "step": 2934
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.271670385510313e-05,
      "loss": 0.5261,
      "step": 2935
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.2624337022646244e-05,
      "loss": 0.335,
      "step": 2936
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.25320339635182e-05,
      "loss": 0.3122,
      "step": 2937
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.24397947381702e-05,
      "loss": 1.0118,
      "step": 2938
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.234761940701169e-05,
      "loss": 0.3386,
      "step": 2939
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.225550803041008e-05,
      "loss": 0.3375,
      "step": 2940
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.2163460668691095e-05,
      "loss": 0.2811,
      "step": 2941
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.207147738213837e-05,
      "loss": 0.3481,
      "step": 2942
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.197955823099374e-05,
      "loss": 0.481,
      "step": 2943
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.188770327545686e-05,
      "loss": 0.1131,
      "step": 2944
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.179591257568554e-05,
      "loss": 0.1615,
      "step": 2945
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.170418619179529e-05,
      "loss": 0.606,
      "step": 2946
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.161252418385972e-05,
      "loss": 0.2319,
      "step": 2947
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.152092661191007e-05,
      "loss": 0.4837,
      "step": 2948
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.1429393535935506e-05,
      "loss": 0.1843,
      "step": 2949
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.133792501588301e-05,
      "loss": 0.2255,
      "step": 2950
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.1246521111657064e-05,
      "loss": 0.3612,
      "step": 2951
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.115518188312009e-05,
      "loss": 0.4808,
      "step": 2952
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.1063907390091933e-05,
      "loss": 0.32,
      "step": 2953
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.097269769235025e-05,
      "loss": 0.4489,
      "step": 2954
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.088155284963002e-05,
      "loss": 0.2593,
      "step": 2955
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.079047292162401e-05,
      "loss": 0.211,
      "step": 2956
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.069945796798222e-05,
      "loss": 0.1233,
      "step": 2957
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.0608508048312324e-05,
      "loss": 0.5068,
      "step": 2958
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.051762322217927e-05,
      "loss": 0.2409,
      "step": 2959
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.0426803549105335e-05,
      "loss": 0.2315,
      "step": 2960
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.033604908857028e-05,
      "loss": 0.6897,
      "step": 2961
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.0245359900011e-05,
      "loss": 0.2819,
      "step": 2962
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.015473604282182e-05,
      "loss": 0.2502,
      "step": 2963
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.006417757635404e-05,
      "loss": 0.2687,
      "step": 2964
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.997368455991638e-05,
      "loss": 0.2619,
      "step": 2965
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.988325705277448e-05,
      "loss": 0.3236,
      "step": 2966
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.9792895114151305e-05,
      "loss": 0.5207,
      "step": 2967
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.9702598803226614e-05,
      "loss": 0.1687,
      "step": 2968
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.9612368179137384e-05,
      "loss": 0.2517,
      "step": 2969
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.952220330097757e-05,
      "loss": 0.5477,
      "step": 2970
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.9432104227797905e-05,
      "loss": 0.3203,
      "step": 2971
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.9342071018606214e-05,
      "loss": 0.3189,
      "step": 2972
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.9252103732367025e-05,
      "loss": 0.3062,
      "step": 2973
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.9162202428001866e-05,
      "loss": 0.3903,
      "step": 2974
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.9072367164388845e-05,
      "loss": 0.3086,
      "step": 2975
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.8982598000363007e-05,
      "loss": 0.3078,
      "step": 2976
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.889289499471598e-05,
      "loss": 0.3854,
      "step": 2977
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.880325820619611e-05,
      "loss": 0.1968,
      "step": 2978
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.8713687693508455e-05,
      "loss": 0.4669,
      "step": 2979
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.862418351531449e-05,
      "loss": 0.5703,
      "step": 2980
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.853474573023242e-05,
      "loss": 0.1447,
      "step": 2981
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.844537439683684e-05,
      "loss": 0.3841,
      "step": 2982
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.835606957365894e-05,
      "loss": 0.3581,
      "step": 2983
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.826683131918621e-05,
      "loss": 0.4479,
      "step": 2984
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.8177659691862704e-05,
      "loss": 0.2699,
      "step": 2985
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.8088554750088685e-05,
      "loss": 0.2691,
      "step": 2986
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.799951655222088e-05,
      "loss": 0.5994,
      "step": 2987
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.791054515657217e-05,
      "loss": 0.1327,
      "step": 2988
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.78216406214118e-05,
      "loss": 0.3142,
      "step": 2989
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.773280300496524e-05,
      "loss": 0.246,
      "step": 2990
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.764403236541394e-05,
      "loss": 0.2781,
      "step": 2991
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.755532876089576e-05,
      "loss": 0.246,
      "step": 2992
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.7466692249504406e-05,
      "loss": 0.1799,
      "step": 2993
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.7378122889289855e-05,
      "loss": 0.6105,
      "step": 2994
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.728962073825797e-05,
      "loss": 0.3733,
      "step": 2995
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.720118585437056e-05,
      "loss": 0.4812,
      "step": 2996
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.7112818295545574e-05,
      "loss": 0.5222,
      "step": 2997
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.702451811965664e-05,
      "loss": 0.3306,
      "step": 2998
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.693628538453345e-05,
      "loss": 0.1448,
      "step": 2999
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.6848120147961366e-05,
      "loss": 0.4753,
      "step": 3000
    },
    {
      "epoch": 0.75,
      "eval_loss": 0.3383316993713379,
      "eval_runtime": 49.7291,
      "eval_samples_per_second": 0.664,
      "eval_steps_per_second": 0.664,
      "step": 3000
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.676002246768169e-05,
      "loss": 0.2572,
      "step": 3001
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.667199240139134e-05,
      "loss": 0.5582,
      "step": 3002
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.6584030006743096e-05,
      "loss": 0.5374,
      "step": 3003
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.649613534134526e-05,
      "loss": 0.2005,
      "step": 3004
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.6408308462761966e-05,
      "loss": 0.3433,
      "step": 3005
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.632054942851271e-05,
      "loss": 0.4311,
      "step": 3006
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.6232858296072846e-05,
      "loss": 0.3136,
      "step": 3007
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.6145235122872946e-05,
      "loss": 0.2368,
      "step": 3008
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.6057679966299334e-05,
      "loss": 0.3053,
      "step": 3009
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.597019288369371e-05,
      "loss": 0.2114,
      "step": 3010
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.588277393235307e-05,
      "loss": 0.4799,
      "step": 3011
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.579542316952998e-05,
      "loss": 0.6092,
      "step": 3012
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.570814065243218e-05,
      "loss": 0.0611,
      "step": 3013
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.562092643822287e-05,
      "loss": 0.4169,
      "step": 3014
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.553378058402033e-05,
      "loss": 0.2647,
      "step": 3015
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.54467031468983e-05,
      "loss": 0.6261,
      "step": 3016
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.535969418388548e-05,
      "loss": 0.373,
      "step": 3017
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.527275375196591e-05,
      "loss": 0.1869,
      "step": 3018
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.518588190807862e-05,
      "loss": 0.2941,
      "step": 3019
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.509907870911776e-05,
      "loss": 0.7492,
      "step": 3020
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.501234421193262e-05,
      "loss": 0.3582,
      "step": 3021
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.492567847332733e-05,
      "loss": 0.1832,
      "step": 3022
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.483908155006112e-05,
      "loss": 0.6226,
      "step": 3023
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.4752553498848034e-05,
      "loss": 0.1957,
      "step": 3024
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.466609437635716e-05,
      "loss": 0.313,
      "step": 3025
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.457970423921227e-05,
      "loss": 0.4329,
      "step": 3026
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.449338314399214e-05,
      "loss": 0.3308,
      "step": 3027
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.440713114723016e-05,
      "loss": 0.4831,
      "step": 3028
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.432094830541454e-05,
      "loss": 0.5301,
      "step": 3029
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.4234834674988275e-05,
      "loss": 0.5913,
      "step": 3030
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.4148790312348914e-05,
      "loss": 0.3747,
      "step": 3031
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.4062815273848634e-05,
      "loss": 0.2612,
      "step": 3032
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.3976909615794356e-05,
      "loss": 0.2981,
      "step": 3033
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.389107339444735e-05,
      "loss": 0.4127,
      "step": 3034
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.380530666602366e-05,
      "loss": 0.4514,
      "step": 3035
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.371960948669355e-05,
      "loss": 0.4528,
      "step": 3036
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.3633981912581985e-05,
      "loss": 0.5975,
      "step": 3037
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.354842399976818e-05,
      "loss": 0.4762,
      "step": 3038
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.346293580428573e-05,
      "loss": 0.5761,
      "step": 3039
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.3377517382122654e-05,
      "loss": 0.2514,
      "step": 3040
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.3292168789221286e-05,
      "loss": 0.512,
      "step": 3041
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.32068900814781e-05,
      "loss": 0.3365,
      "step": 3042
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.312168131474396e-05,
      "loss": 0.4224,
      "step": 3043
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.3036542544823743e-05,
      "loss": 0.4799,
      "step": 3044
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.295147382747669e-05,
      "loss": 0.2171,
      "step": 3045
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.286647521841596e-05,
      "loss": 0.308,
      "step": 3046
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.2781546773309e-05,
      "loss": 0.338,
      "step": 3047
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.2696688547777076e-05,
      "loss": 0.1131,
      "step": 3048
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.2611900597395666e-05,
      "loss": 0.31,
      "step": 3049
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.252718297769418e-05,
      "loss": 0.3782,
      "step": 3050
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.244253574415585e-05,
      "loss": 0.1435,
      "step": 3051
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.2357958952217975e-05,
      "loss": 0.2966,
      "step": 3052
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.227345265727154e-05,
      "loss": 0.374,
      "step": 3053
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.218901691466158e-05,
      "loss": 0.2672,
      "step": 3054
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.210465177968669e-05,
      "loss": 0.5219,
      "step": 3055
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.202035730759942e-05,
      "loss": 0.2878,
      "step": 3056
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.193613355360587e-05,
      "loss": 0.3826,
      "step": 3057
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.185198057286603e-05,
      "loss": 0.16,
      "step": 3058
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.176789842049331e-05,
      "loss": 0.4018,
      "step": 3059
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.1683887151554876e-05,
      "loss": 0.2029,
      "step": 3060
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.1599946821071517e-05,
      "loss": 0.2261,
      "step": 3061
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.151607748401738e-05,
      "loss": 0.264,
      "step": 3062
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.1432279195320304e-05,
      "loss": 0.2892,
      "step": 3063
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.134855200986143e-05,
      "loss": 0.4166,
      "step": 3064
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.126489598247552e-05,
      "loss": 0.1358,
      "step": 3065
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.118131116795052e-05,
      "loss": 0.567,
      "step": 3066
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.109779762102797e-05,
      "loss": 0.5017,
      "step": 3067
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.101435539640249e-05,
      "loss": 0.537,
      "step": 3068
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.0930984548722215e-05,
      "loss": 0.7064,
      "step": 3069
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.0847685132588355e-05,
      "loss": 0.5181,
      "step": 3070
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.0764457202555476e-05,
      "loss": 0.5646,
      "step": 3071
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.068130081313119e-05,
      "loss": 0.123,
      "step": 3072
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.059821601877644e-05,
      "loss": 0.582,
      "step": 3073
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.051520287390509e-05,
      "loss": 0.384,
      "step": 3074
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.0432261432884146e-05,
      "loss": 0.3685,
      "step": 3075
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.0349391750033735e-05,
      "loss": 0.3906,
      "step": 3076
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.026659387962684e-05,
      "loss": 0.3702,
      "step": 3077
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.0183867875889594e-05,
      "loss": 0.612,
      "step": 3078
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.0101213793000835e-05,
      "loss": 0.3282,
      "step": 3079
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.001863168509252e-05,
      "loss": 0.1258,
      "step": 3080
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.993612160624939e-05,
      "loss": 0.2659,
      "step": 3081
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.985368361050889e-05,
      "loss": 0.7373,
      "step": 3082
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.9771317751861494e-05,
      "loss": 0.4659,
      "step": 3083
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.968902408425019e-05,
      "loss": 0.4283,
      "step": 3084
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.960680266157088e-05,
      "loss": 0.2476,
      "step": 3085
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.952465353767198e-05,
      "loss": 0.2909,
      "step": 3086
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.9442576766354755e-05,
      "loss": 0.4622,
      "step": 3087
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.936057240137286e-05,
      "loss": 0.4661,
      "step": 3088
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.9278640496432746e-05,
      "loss": 0.3959,
      "step": 3089
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.9196781105193224e-05,
      "loss": 0.27,
      "step": 3090
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.911499428126572e-05,
      "loss": 0.414,
      "step": 3091
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.903328007821417e-05,
      "loss": 0.4068,
      "step": 3092
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.895163854955479e-05,
      "loss": 0.206,
      "step": 3093
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.887006974875642e-05,
      "loss": 0.3217,
      "step": 3094
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.878857372924002e-05,
      "loss": 0.3924,
      "step": 3095
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.870715054437912e-05,
      "loss": 0.6042,
      "step": 3096
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.862580024749935e-05,
      "loss": 0.1783,
      "step": 3097
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.8544522891878805e-05,
      "loss": 0.3076,
      "step": 3098
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.846331853074757e-05,
      "loss": 0.4327,
      "step": 3099
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.838218721728813e-05,
      "loss": 0.4655,
      "step": 3100
    },
    {
      "epoch": 0.77,
      "eval_loss": 0.3357388377189636,
      "eval_runtime": 49.9385,
      "eval_samples_per_second": 0.661,
      "eval_steps_per_second": 0.661,
      "step": 3100
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.8301129004635104e-05,
      "loss": 0.3176,
      "step": 3101
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.8220143945875087e-05,
      "loss": 0.5328,
      "step": 3102
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.813923209404693e-05,
      "loss": 0.4568,
      "step": 3103
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.805839350214143e-05,
      "loss": 0.5161,
      "step": 3104
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.7977628223101496e-05,
      "loss": 0.2771,
      "step": 3105
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.7896936309821916e-05,
      "loss": 0.4723,
      "step": 3106
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.781631781514955e-05,
      "loss": 0.0926,
      "step": 3107
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.773577279188302e-05,
      "loss": 0.2167,
      "step": 3108
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.765530129277304e-05,
      "loss": 0.3558,
      "step": 3109
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.7574903370522e-05,
      "loss": 0.328,
      "step": 3110
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.74945790777841e-05,
      "loss": 0.352,
      "step": 3111
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.741432846716545e-05,
      "loss": 0.4774,
      "step": 3112
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.7334151591223785e-05,
      "loss": 0.3252,
      "step": 3113
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.725404850246864e-05,
      "loss": 0.5194,
      "step": 3114
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.717401925336111e-05,
      "loss": 0.2655,
      "step": 3115
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.709406389631409e-05,
      "loss": 0.3198,
      "step": 3116
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.70141824836919e-05,
      "loss": 0.1851,
      "step": 3117
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.693437506781061e-05,
      "loss": 0.2107,
      "step": 3118
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.685464170093766e-05,
      "loss": 0.2252,
      "step": 3119
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.6774982435292105e-05,
      "loss": 0.327,
      "step": 3120
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.66953973230445e-05,
      "loss": 0.1264,
      "step": 3121
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.661588641631668e-05,
      "loss": 0.4189,
      "step": 3122
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.653644976718206e-05,
      "loss": 0.5705,
      "step": 3123
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.6457087427665266e-05,
      "loss": 0.4468,
      "step": 3124
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.637779944974241e-05,
      "loss": 0.508,
      "step": 3125
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.629858588534071e-05,
      "loss": 0.2955,
      "step": 3126
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.621944678633889e-05,
      "loss": 0.786,
      "step": 3127
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.614038220456668e-05,
      "loss": 0.3363,
      "step": 3128
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.6061392191805174e-05,
      "loss": 0.2987,
      "step": 3129
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.5982476799786484e-05,
      "loss": 0.2537,
      "step": 3130
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.5903636080193974e-05,
      "loss": 0.3505,
      "step": 3131
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.5824870084662096e-05,
      "loss": 0.5119,
      "step": 3132
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.574617886477622e-05,
      "loss": 0.3272,
      "step": 3133
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.566756247207296e-05,
      "loss": 0.1651,
      "step": 3134
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.558902095803974e-05,
      "loss": 0.3614,
      "step": 3135
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.5510554374115045e-05,
      "loss": 0.342,
      "step": 3136
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.5432162771688234e-05,
      "loss": 0.7226,
      "step": 3137
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.535384620209963e-05,
      "loss": 0.3131,
      "step": 3138
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.5275604716640305e-05,
      "loss": 0.519,
      "step": 3139
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.5197438366552275e-05,
      "loss": 0.1478,
      "step": 3140
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.511934720302825e-05,
      "loss": 0.0917,
      "step": 3141
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.504133127721176e-05,
      "loss": 0.2108,
      "step": 3142
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.496339064019709e-05,
      "loss": 0.6875,
      "step": 3143
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.488552534302908e-05,
      "loss": 0.2197,
      "step": 3144
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.480773543670341e-05,
      "loss": 0.1915,
      "step": 3145
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.473002097216626e-05,
      "loss": 0.5731,
      "step": 3146
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.465238200031437e-05,
      "loss": 0.1505,
      "step": 3147
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.45748185719952e-05,
      "loss": 0.1607,
      "step": 3148
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.4497330738006554e-05,
      "loss": 0.4177,
      "step": 3149
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.4419918549096895e-05,
      "loss": 0.5001,
      "step": 3150
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.434258205596499e-05,
      "loss": 0.3215,
      "step": 3151
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.4265321309260155e-05,
      "loss": 0.2173,
      "step": 3152
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.418813635958198e-05,
      "loss": 0.3349,
      "step": 3153
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.411102725748057e-05,
      "loss": 0.366,
      "step": 3154
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.4033994053456185e-05,
      "loss": 0.5837,
      "step": 3155
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.3957036797959516e-05,
      "loss": 0.1353,
      "step": 3156
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.388015554139139e-05,
      "loss": 0.6227,
      "step": 3157
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.380335033410302e-05,
      "loss": 0.3019,
      "step": 3158
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.372662122639561e-05,
      "loss": 0.3265,
      "step": 3159
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.3649968268520744e-05,
      "loss": 0.1962,
      "step": 3160
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.3573391510679906e-05,
      "loss": 0.1215,
      "step": 3161
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.3496891003024887e-05,
      "loss": 0.1954,
      "step": 3162
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.342046679565744e-05,
      "loss": 0.3644,
      "step": 3163
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.3344118938629295e-05,
      "loss": 0.5943,
      "step": 3164
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.3267847481942315e-05,
      "loss": 0.3326,
      "step": 3165
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.319165247554817e-05,
      "loss": 0.3607,
      "step": 3166
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.311553396934862e-05,
      "loss": 0.1539,
      "step": 3167
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.303949201319515e-05,
      "loss": 0.2509,
      "step": 3168
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.29635266568893e-05,
      "loss": 0.3607,
      "step": 3169
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.288763795018227e-05,
      "loss": 0.244,
      "step": 3170
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.281182594277516e-05,
      "loss": 0.6146,
      "step": 3171
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.2736090684318886e-05,
      "loss": 0.3062,
      "step": 3172
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.266043222441392e-05,
      "loss": 0.3289,
      "step": 3173
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.258485061261065e-05,
      "loss": 0.7695,
      "step": 3174
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.2509345898408925e-05,
      "loss": 0.414,
      "step": 3175
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.243391813125843e-05,
      "loss": 0.4848,
      "step": 3176
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.23585673605583e-05,
      "loss": 0.2936,
      "step": 3177
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.2283293635657384e-05,
      "loss": 0.4685,
      "step": 3178
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.22080970058539e-05,
      "loss": 0.2477,
      "step": 3179
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.213297752039576e-05,
      "loss": 0.25,
      "step": 3180
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.2057935228480184e-05,
      "loss": 0.2361,
      "step": 3181
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.198297017925395e-05,
      "loss": 0.3789,
      "step": 3182
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.190808242181326e-05,
      "loss": 0.4109,
      "step": 3183
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.183327200520362e-05,
      "loss": 0.3164,
      "step": 3184
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.175853897841985e-05,
      "loss": 0.5256,
      "step": 3185
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.168388339040625e-05,
      "loss": 0.2801,
      "step": 3186
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.160930529005621e-05,
      "loss": 0.1581,
      "step": 3187
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.153480472621257e-05,
      "loss": 0.1813,
      "step": 3188
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.1460381747667156e-05,
      "loss": 0.1664,
      "step": 3189
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.138603640316123e-05,
      "loss": 0.1933,
      "step": 3190
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.131176874138505e-05,
      "loss": 0.169,
      "step": 3191
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.123757881097798e-05,
      "loss": 0.3349,
      "step": 3192
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.11634666605286e-05,
      "loss": 0.5318,
      "step": 3193
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.108943233857451e-05,
      "loss": 0.295,
      "step": 3194
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.101547589360222e-05,
      "loss": 0.2328,
      "step": 3195
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.094159737404742e-05,
      "loss": 0.3658,
      "step": 3196
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.0867796828294605e-05,
      "loss": 0.2871,
      "step": 3197
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.079407430467736e-05,
      "loss": 0.5672,
      "step": 3198
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.072042985147797e-05,
      "loss": 0.5073,
      "step": 3199
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.064686351692782e-05,
      "loss": 0.4977,
      "step": 3200
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.3337692320346832,
      "eval_runtime": 53.0306,
      "eval_samples_per_second": 0.622,
      "eval_steps_per_second": 0.622,
      "step": 3200
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.057337534920693e-05,
      "loss": 0.0604,
      "step": 3201
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.0499965396444247e-05,
      "loss": 0.2337,
      "step": 3202
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.0426633706717502e-05,
      "loss": 0.2148,
      "step": 3203
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.0353380328053074e-05,
      "loss": 0.4353,
      "step": 3204
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.0280205308426152e-05,
      "loss": 0.4081,
      "step": 3205
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.0207108695760523e-05,
      "loss": 0.3308,
      "step": 3206
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.0134090537928723e-05,
      "loss": 0.2884,
      "step": 3207
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.0061150882751767e-05,
      "loss": 0.143,
      "step": 3208
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.9988289777999442e-05,
      "loss": 0.3953,
      "step": 3209
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.9915507271389876e-05,
      "loss": 0.5071,
      "step": 3210
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.9842803410589927e-05,
      "loss": 0.1973,
      "step": 3211
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.977017824321479e-05,
      "loss": 0.6918,
      "step": 3212
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.9697631816828176e-05,
      "loss": 0.117,
      "step": 3213
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.962516417894234e-05,
      "loss": 0.4661,
      "step": 3214
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.9552775377017696e-05,
      "loss": 0.5275,
      "step": 3215
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.9480465458463272e-05,
      "loss": 0.2782,
      "step": 3216
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.9408234470636233e-05,
      "loss": 0.4819,
      "step": 3217
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.9336082460842227e-05,
      "loss": 0.43,
      "step": 3218
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.9264009476335017e-05,
      "loss": 0.1345,
      "step": 3219
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.9192015564316747e-05,
      "loss": 0.6688,
      "step": 3220
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.9120100771937648e-05,
      "loss": 0.5227,
      "step": 3221
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.9048265146296283e-05,
      "loss": 0.3937,
      "step": 3222
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.8976508734439204e-05,
      "loss": 0.4346,
      "step": 3223
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.8904831583361226e-05,
      "loss": 0.2445,
      "step": 3224
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.883323374000514e-05,
      "loss": 0.3852,
      "step": 3225
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.876171525126192e-05,
      "loss": 0.1967,
      "step": 3226
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.8690276163970444e-05,
      "loss": 0.2244,
      "step": 3227
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.8618916524917622e-05,
      "loss": 0.1962,
      "step": 3228
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.8547636380838423e-05,
      "loss": 0.2785,
      "step": 3229
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.8476435778415584e-05,
      "loss": 0.3128,
      "step": 3230
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.8405314764279975e-05,
      "loss": 0.5072,
      "step": 3231
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.83342733850101e-05,
      "loss": 0.3714,
      "step": 3232
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.8263311687132457e-05,
      "loss": 0.3358,
      "step": 3233
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.8192429717121374e-05,
      "loss": 0.5146,
      "step": 3234
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.8121627521398848e-05,
      "loss": 0.3942,
      "step": 3235
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.8050905146334757e-05,
      "loss": 0.1729,
      "step": 3236
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.798026263824658e-05,
      "loss": 0.1818,
      "step": 3237
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.790970004339961e-05,
      "loss": 0.4255,
      "step": 3238
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.7839217408006686e-05,
      "loss": 0.2543,
      "step": 3239
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.776881477822841e-05,
      "loss": 0.3008,
      "step": 3240
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.7698492200172833e-05,
      "loss": 0.239,
      "step": 3241
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.7628249719895702e-05,
      "loss": 0.4595,
      "step": 3242
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.7558087383400296e-05,
      "loss": 0.3457,
      "step": 3243
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.748800523663731e-05,
      "loss": 0.4799,
      "step": 3244
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.7418003325505067e-05,
      "loss": 0.2366,
      "step": 3245
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.734808169584919e-05,
      "loss": 0.242,
      "step": 3246
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.7278240393462845e-05,
      "loss": 0.437,
      "step": 3247
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.7208479464086487e-05,
      "loss": 0.2475,
      "step": 3248
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.713879895340804e-05,
      "loss": 0.3229,
      "step": 3249
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.7069198907062663e-05,
      "loss": 0.3693,
      "step": 3250
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.6999679370632914e-05,
      "loss": 0.2698,
      "step": 3251
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.69302403896485e-05,
      "loss": 0.4854,
      "step": 3252
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.686088200958647e-05,
      "loss": 0.3925,
      "step": 3253
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.6791604275871127e-05,
      "loss": 0.6191,
      "step": 3254
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.672240723387377e-05,
      "loss": 0.4181,
      "step": 3255
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.6653290928913075e-05,
      "loss": 0.1483,
      "step": 3256
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.6584255406254622e-05,
      "loss": 0.1364,
      "step": 3257
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.6515300711111322e-05,
      "loss": 0.5482,
      "step": 3258
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.6446426888642925e-05,
      "loss": 0.536,
      "step": 3259
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.6377633983956386e-05,
      "loss": 0.1937,
      "step": 3260
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.6308922042105518e-05,
      "loss": 0.69,
      "step": 3261
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.624029110809129e-05,
      "loss": 0.2357,
      "step": 3262
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.617174122686146e-05,
      "loss": 0.3204,
      "step": 3263
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.610327244331072e-05,
      "loss": 0.4284,
      "step": 3264
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.6034884802280764e-05,
      "loss": 0.6887,
      "step": 3265
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.5966578348560003e-05,
      "loss": 0.4017,
      "step": 3266
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.5898353126883798e-05,
      "loss": 0.4768,
      "step": 3267
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.5830209181934193e-05,
      "loss": 0.1398,
      "step": 3268
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.576214655834012e-05,
      "loss": 0.2048,
      "step": 3269
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.569416530067714e-05,
      "loss": 0.2723,
      "step": 3270
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.5626265453467653e-05,
      "loss": 0.4336,
      "step": 3271
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.5558447061180574e-05,
      "loss": 0.3265,
      "step": 3272
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.5490710168231626e-05,
      "loss": 0.252,
      "step": 3273
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.5423054818983117e-05,
      "loss": 0.3565,
      "step": 3274
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.535548105774386e-05,
      "loss": 0.4254,
      "step": 3275
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.528798892876938e-05,
      "loss": 0.2332,
      "step": 3276
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.5220578476261595e-05,
      "loss": 0.2385,
      "step": 3277
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.5153249744369063e-05,
      "loss": 0.3286,
      "step": 3278
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.5086002777186704e-05,
      "loss": 0.3437,
      "step": 3279
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.5018837618755982e-05,
      "loss": 0.2372,
      "step": 3280
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.4951754313064714e-05,
      "loss": 0.221,
      "step": 3281
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.4884752904047173e-05,
      "loss": 0.2796,
      "step": 3282
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.4817833435583906e-05,
      "loss": 0.4308,
      "step": 3283
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.4750995951501905e-05,
      "loss": 0.5191,
      "step": 3284
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.4684240495574426e-05,
      "loss": 0.2396,
      "step": 3285
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.4617567111520926e-05,
      "loss": 0.3612,
      "step": 3286
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.4550975843007258e-05,
      "loss": 0.1271,
      "step": 3287
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.448446673364536e-05,
      "loss": 0.4395,
      "step": 3288
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.4418039826993448e-05,
      "loss": 0.2323,
      "step": 3289
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.435169516655584e-05,
      "loss": 0.0437,
      "step": 3290
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.4285432795783084e-05,
      "loss": 0.2874,
      "step": 3291
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.4219252758071695e-05,
      "loss": 0.2301,
      "step": 3292
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.415315509676437e-05,
      "loss": 0.3582,
      "step": 3293
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.4087139855149883e-05,
      "loss": 0.3788,
      "step": 3294
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.40212070764629e-05,
      "loss": 0.1092,
      "step": 3295
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.3955356803884212e-05,
      "loss": 0.2581,
      "step": 3296
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.388958908054049e-05,
      "loss": 0.2823,
      "step": 3297
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.382390394950439e-05,
      "loss": 0.2846,
      "step": 3298
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.3758301453794458e-05,
      "loss": 0.3618,
      "step": 3299
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.369278163637509e-05,
      "loss": 0.3554,
      "step": 3300
    },
    {
      "epoch": 0.82,
      "eval_loss": 0.3325382471084595,
      "eval_runtime": 49.7614,
      "eval_samples_per_second": 0.663,
      "eval_steps_per_second": 0.663,
      "step": 3300
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.3627344540156594e-05,
      "loss": 0.3568,
      "step": 3301
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.3561990207995058e-05,
      "loss": 0.1618,
      "step": 3302
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.3496718682692405e-05,
      "loss": 0.2618,
      "step": 3303
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.343153000699626e-05,
      "loss": 0.0464,
      "step": 3304
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.3366424223600096e-05,
      "loss": 0.395,
      "step": 3305
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.3301401375142943e-05,
      "loss": 0.7565,
      "step": 3306
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.3236461504209708e-05,
      "loss": 0.3862,
      "step": 3307
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.3171604653330754e-05,
      "loss": 0.5239,
      "step": 3308
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.310683086498224e-05,
      "loss": 0.5261,
      "step": 3309
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.3042140181585812e-05,
      "loss": 0.2427,
      "step": 3310
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.297753264550876e-05,
      "loss": 0.2557,
      "step": 3311
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.291300829906386e-05,
      "loss": 0.319,
      "step": 3312
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.2848567184509458e-05,
      "loss": 0.5522,
      "step": 3313
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.278420934404941e-05,
      "loss": 0.2851,
      "step": 3314
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.2719934819832895e-05,
      "loss": 0.1355,
      "step": 3315
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.2655743653954716e-05,
      "loss": 0.2442,
      "step": 3316
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.2591635888454923e-05,
      "loss": 0.4892,
      "step": 3317
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.252761156531908e-05,
      "loss": 0.4763,
      "step": 3318
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.2463670726477968e-05,
      "loss": 0.4324,
      "step": 3319
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.2399813413807817e-05,
      "loss": 0.3291,
      "step": 3320
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.233603966913004e-05,
      "loss": 0.3889,
      "step": 3321
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.2272349534211433e-05,
      "loss": 0.2483,
      "step": 3322
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.2208743050763922e-05,
      "loss": 0.4917,
      "step": 3323
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.214522026044474e-05,
      "loss": 0.4527,
      "step": 3324
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.2081781204856308e-05,
      "loss": 0.3026,
      "step": 3325
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.201842592554609e-05,
      "loss": 0.251,
      "step": 3326
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.195515446400683e-05,
      "loss": 0.2346,
      "step": 3327
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.1891966861676263e-05,
      "loss": 0.2818,
      "step": 3328
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.1828863159937306e-05,
      "loss": 0.3651,
      "step": 3329
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.176584340011781e-05,
      "loss": 0.3394,
      "step": 3330
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.17029076234908e-05,
      "loss": 0.3102,
      "step": 3331
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.1640055871274102e-05,
      "loss": 0.32,
      "step": 3332
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.157728818463074e-05,
      "loss": 0.1974,
      "step": 3333
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.1514604604668466e-05,
      "loss": 0.1618,
      "step": 3334
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.1452005172440124e-05,
      "loss": 0.4581,
      "step": 3335
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.1389489928943316e-05,
      "loss": 0.3575,
      "step": 3336
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.1327058915120626e-05,
      "loss": 0.2512,
      "step": 3337
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.1264712171859332e-05,
      "loss": 0.6523,
      "step": 3338
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.1202449739991672e-05,
      "loss": 0.5096,
      "step": 3339
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.1140271660294545e-05,
      "loss": 0.665,
      "step": 3340
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.1078177973489723e-05,
      "loss": 0.1037,
      "step": 3341
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.101616872024358e-05,
      "loss": 0.1821,
      "step": 3342
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.0954243941167247e-05,
      "loss": 0.4779,
      "step": 3343
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.0892403676816565e-05,
      "loss": 1.0328,
      "step": 3344
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.0830647967692037e-05,
      "loss": 0.2062,
      "step": 3345
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.076897685423866e-05,
      "loss": 0.439,
      "step": 3346
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.0707390376846212e-05,
      "loss": 0.3362,
      "step": 3347
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.0645888575848873e-05,
      "loss": 0.6683,
      "step": 3348
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.0584471491525494e-05,
      "loss": 0.5089,
      "step": 3349
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.0523139164099335e-05,
      "loss": 0.3456,
      "step": 3350
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.0461891633738257e-05,
      "loss": 0.305,
      "step": 3351
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.0400728940554484e-05,
      "loss": 0.2465,
      "step": 3352
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.0339651124604768e-05,
      "loss": 0.3548,
      "step": 3353
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.0278658225890176e-05,
      "loss": 0.3186,
      "step": 3354
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.0217750284356243e-05,
      "loss": 0.7048,
      "step": 3355
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.0156927339892875e-05,
      "loss": 0.3057,
      "step": 3356
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.009618943233419e-05,
      "loss": 0.3978,
      "step": 3357
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.003553660145879e-05,
      "loss": 0.2341,
      "step": 3358
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.997496888698938e-05,
      "loss": 0.0742,
      "step": 3359
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.991448632859306e-05,
      "loss": 0.3103,
      "step": 3360
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.985408896588105e-05,
      "loss": 0.0306,
      "step": 3361
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.9793776838408903e-05,
      "loss": 0.2988,
      "step": 3362
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.9733549985676195e-05,
      "loss": 0.2791,
      "step": 3363
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.967340844712677e-05,
      "loss": 0.2147,
      "step": 3364
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.961335226214859e-05,
      "loss": 0.3814,
      "step": 3365
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.9553381470073642e-05,
      "loss": 0.0701,
      "step": 3366
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.9493496110178076e-05,
      "loss": 0.3044,
      "step": 3367
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.9433696221681982e-05,
      "loss": 0.4691,
      "step": 3368
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.9373981843749637e-05,
      "loss": 0.4336,
      "step": 3369
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.931435301548912e-05,
      "loss": 0.3383,
      "step": 3370
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.9254809775952645e-05,
      "loss": 0.5109,
      "step": 3371
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.9195352164136267e-05,
      "loss": 0.341,
      "step": 3372
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.9135980218980056e-05,
      "loss": 0.5674,
      "step": 3373
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.907669397936784e-05,
      "loss": 0.5345,
      "step": 3374
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.9017493484127495e-05,
      "loss": 0.1842,
      "step": 3375
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.8958378772030553e-05,
      "loss": 0.2201,
      "step": 3376
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.889934988179253e-05,
      "loss": 0.0802,
      "step": 3377
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.8840406852072637e-05,
      "loss": 0.3695,
      "step": 3378
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.8781549721473842e-05,
      "loss": 0.5718,
      "step": 3379
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.8722778528542965e-05,
      "loss": 0.3232,
      "step": 3380
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.8664093311770413e-05,
      "loss": 0.2504,
      "step": 3381
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.8605494109590373e-05,
      "loss": 0.5234,
      "step": 3382
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.8546980960380652e-05,
      "loss": 0.2232,
      "step": 3383
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.8488553902462745e-05,
      "loss": 0.4762,
      "step": 3384
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.8430212974101687e-05,
      "loss": 0.3844,
      "step": 3385
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.8371958213506206e-05,
      "loss": 0.4645,
      "step": 3386
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.8313789658828542e-05,
      "loss": 0.2331,
      "step": 3387
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.8255707348164427e-05,
      "loss": 0.2455,
      "step": 3388
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.819771131955325e-05,
      "loss": 0.3202,
      "step": 3389
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.813980161097772e-05,
      "loss": 0.182,
      "step": 3390
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.808197826036416e-05,
      "loss": 0.3723,
      "step": 3391
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.802424130558222e-05,
      "loss": 0.7165,
      "step": 3392
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.796659078444509e-05,
      "loss": 0.2793,
      "step": 3393
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.7909026734709215e-05,
      "loss": 0.4634,
      "step": 3394
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.7851549194074504e-05,
      "loss": 0.1488,
      "step": 3395
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.7794158200184235e-05,
      "loss": 0.7743,
      "step": 3396
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.7736853790624867e-05,
      "loss": 0.1612,
      "step": 3397
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.76796360029263e-05,
      "loss": 0.3586,
      "step": 3398
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.7622504874561593e-05,
      "loss": 0.2532,
      "step": 3399
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.7565460442947166e-05,
      "loss": 0.1407,
      "step": 3400
    },
    {
      "epoch": 0.85,
      "eval_loss": 0.33149999380111694,
      "eval_runtime": 52.9057,
      "eval_samples_per_second": 0.624,
      "eval_steps_per_second": 0.624,
      "step": 3400
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.7508502745442515e-05,
      "loss": 0.2607,
      "step": 3401
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.7451631819350486e-05,
      "loss": 0.244,
      "step": 3402
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.739484770191697e-05,
      "loss": 0.4562,
      "step": 3403
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.7338150430331083e-05,
      "loss": 0.5194,
      "step": 3404
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.728154004172499e-05,
      "loss": 0.1859,
      "step": 3405
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.722501657317405e-05,
      "loss": 0.074,
      "step": 3406
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.7168580061696652e-05,
      "loss": 0.4921,
      "step": 3407
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.7112230544254197e-05,
      "loss": 0.651,
      "step": 3408
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.705596805775118e-05,
      "loss": 0.3674,
      "step": 3409
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.699979263903501e-05,
      "loss": 0.6503,
      "step": 3410
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.694370432489619e-05,
      "loss": 0.262,
      "step": 3411
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.688770315206807e-05,
      "loss": 0.241,
      "step": 3412
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.6831789157226998e-05,
      "loss": 0.6483,
      "step": 3413
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.6775962376992196e-05,
      "loss": 0.33,
      "step": 3414
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.672022284792575e-05,
      "loss": 0.3964,
      "step": 3415
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.666457060653266e-05,
      "loss": 0.5965,
      "step": 3416
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.6609005689260674e-05,
      "loss": 0.7059,
      "step": 3417
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.6553528132500476e-05,
      "loss": 0.1382,
      "step": 3418
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.6498137972585407e-05,
      "loss": 0.1619,
      "step": 3419
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.6442835245791648e-05,
      "loss": 0.2227,
      "step": 3420
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.6387619988338074e-05,
      "loss": 0.2261,
      "step": 3421
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.633249223638633e-05,
      "loss": 0.7005,
      "step": 3422
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.627745202604065e-05,
      "loss": 0.3693,
      "step": 3423
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.6222499393348097e-05,
      "loss": 0.3705,
      "step": 3424
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.6167634374298205e-05,
      "loss": 0.2082,
      "step": 3425
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.6112857004823228e-05,
      "loss": 0.1525,
      "step": 3426
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.6058167320798033e-05,
      "loss": 0.3064,
      "step": 3427
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.6003565358039954e-05,
      "loss": 0.3597,
      "step": 3428
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5949051152309005e-05,
      "loss": 0.2552,
      "step": 3429
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.589462473930761e-05,
      "loss": 0.3964,
      "step": 3430
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5840286154680803e-05,
      "loss": 0.1437,
      "step": 3431
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5786035434015986e-05,
      "loss": 0.306,
      "step": 3432
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5731872612843138e-05,
      "loss": 0.2949,
      "step": 3433
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5677797726634535e-05,
      "loss": 0.2414,
      "step": 3434
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5623810810804977e-05,
      "loss": 0.1833,
      "step": 3435
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.556991190071162e-05,
      "loss": 0.5011,
      "step": 3436
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5516101031653938e-05,
      "loss": 0.2933,
      "step": 3437
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5462378238873814e-05,
      "loss": 0.4396,
      "step": 3438
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5408743557555363e-05,
      "loss": 0.2279,
      "step": 3439
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5355197022825095e-05,
      "loss": 0.3028,
      "step": 3440
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.53017386697517e-05,
      "loss": 0.4959,
      "step": 3441
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5248368533346195e-05,
      "loss": 0.3695,
      "step": 3442
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5195086648561733e-05,
      "loss": 0.3842,
      "step": 3443
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5141893050293768e-05,
      "loss": 0.1795,
      "step": 3444
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5088787773379834e-05,
      "loss": 0.5019,
      "step": 3445
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.503577085259971e-05,
      "loss": 0.2531,
      "step": 3446
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.498284232267527e-05,
      "loss": 0.3589,
      "step": 3447
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.493000221827047e-05,
      "loss": 0.2116,
      "step": 3448
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4877250573991428e-05,
      "loss": 0.2491,
      "step": 3449
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4824587424386225e-05,
      "loss": 0.364,
      "step": 3450
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4772012803945104e-05,
      "loss": 0.5614,
      "step": 3451
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.471952674710024e-05,
      "loss": 0.8843,
      "step": 3452
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.466712928822582e-05,
      "loss": 0.529,
      "step": 3453
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4614820461638044e-05,
      "loss": 0.2241,
      "step": 3454
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.456260030159499e-05,
      "loss": 0.51,
      "step": 3455
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4510468842296802e-05,
      "loss": 0.2177,
      "step": 3456
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4458426117885368e-05,
      "loss": 0.2294,
      "step": 3457
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.440647216244461e-05,
      "loss": 0.3382,
      "step": 3458
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4354607010000175e-05,
      "loss": 0.2138,
      "step": 3459
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4302830694519673e-05,
      "loss": 0.4561,
      "step": 3460
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4251143249912434e-05,
      "loss": 0.2834,
      "step": 3461
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4199544710029692e-05,
      "loss": 0.1713,
      "step": 3462
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4148035108664335e-05,
      "loss": 0.4089,
      "step": 3463
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.409661447955111e-05,
      "loss": 0.3403,
      "step": 3464
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4045282856366413e-05,
      "loss": 0.2422,
      "step": 3465
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.3994040272728385e-05,
      "loss": 0.4285,
      "step": 3466
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.394288676219688e-05,
      "loss": 0.2686,
      "step": 3467
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.3891822358273352e-05,
      "loss": 0.4094,
      "step": 3468
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.3840847094400965e-05,
      "loss": 0.0959,
      "step": 3469
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.378996100396441e-05,
      "loss": 0.1554,
      "step": 3470
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.3739164120290119e-05,
      "loss": 0.1188,
      "step": 3471
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.3688456476645959e-05,
      "loss": 0.3195,
      "step": 3472
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.363783810624145e-05,
      "loss": 0.3972,
      "step": 3473
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.358730904222754e-05,
      "loss": 0.6379,
      "step": 3474
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.3536869317696842e-05,
      "loss": 0.4849,
      "step": 3475
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.3486518965683313e-05,
      "loss": 0.2196,
      "step": 3476
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.3436258019162427e-05,
      "loss": 0.2136,
      "step": 3477
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.3386086511051187e-05,
      "loss": 0.1934,
      "step": 3478
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.3336004474207857e-05,
      "loss": 0.4555,
      "step": 3479
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.3286011941432267e-05,
      "loss": 0.3846,
      "step": 3480
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.3236108945465511e-05,
      "loss": 0.5809,
      "step": 3481
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.318629551899013e-05,
      "loss": 0.4459,
      "step": 3482
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.313657169462991e-05,
      "loss": 0.4947,
      "step": 3483
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.3086937504950068e-05,
      "loss": 0.6935,
      "step": 3484
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.3037392982457018e-05,
      "loss": 0.1698,
      "step": 3485
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.2987938159598472e-05,
      "loss": 0.447,
      "step": 3486
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.2938573068763486e-05,
      "loss": 0.3143,
      "step": 3487
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.2889297742282235e-05,
      "loss": 0.2042,
      "step": 3488
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.28401122124261e-05,
      "loss": 0.1934,
      "step": 3489
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.279101651140777e-05,
      "loss": 0.4746,
      "step": 3490
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.2742010671380987e-05,
      "loss": 0.2651,
      "step": 3491
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.269309472444071e-05,
      "loss": 0.2619,
      "step": 3492
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.2644268702622984e-05,
      "loss": 0.5248,
      "step": 3493
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.2595532637904993e-05,
      "loss": 0.3432,
      "step": 3494
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.2546886562205005e-05,
      "loss": 0.196,
      "step": 3495
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.249833050738227e-05,
      "loss": 0.2659,
      "step": 3496
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.24498645052372e-05,
      "loss": 0.6346,
      "step": 3497
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.2401488587511189e-05,
      "loss": 0.2863,
      "step": 3498
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.235320278588659e-05,
      "loss": 0.5078,
      "step": 3499
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.230500713198681e-05,
      "loss": 0.5542,
      "step": 3500
    },
    {
      "epoch": 0.87,
      "eval_loss": 0.3312348425388336,
      "eval_runtime": 50.0779,
      "eval_samples_per_second": 0.659,
      "eval_steps_per_second": 0.659,
      "step": 3500
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.2256901657376133e-05,
      "loss": 0.3568,
      "step": 3501
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.2208886393559858e-05,
      "loss": 0.3706,
      "step": 3502
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.2160961371984145e-05,
      "loss": 0.2396,
      "step": 3503
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.2113126624036107e-05,
      "loss": 0.3954,
      "step": 3504
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.2065382181043664e-05,
      "loss": 0.171,
      "step": 3505
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.2017728074275706e-05,
      "loss": 0.3023,
      "step": 3506
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.19701643349418e-05,
      "loss": 0.1191,
      "step": 3507
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.192269099419248e-05,
      "loss": 0.3744,
      "step": 3508
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1875308083119023e-05,
      "loss": 0.1785,
      "step": 3509
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1828015632753424e-05,
      "loss": 0.1485,
      "step": 3510
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1780813674068556e-05,
      "loss": 0.2604,
      "step": 3511
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.173370223797786e-05,
      "loss": 0.4555,
      "step": 3512
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1686681355335686e-05,
      "loss": 0.1281,
      "step": 3513
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1639751056936886e-05,
      "loss": 0.4963,
      "step": 3514
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1592911373517156e-05,
      "loss": 0.1335,
      "step": 3515
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1546162335752723e-05,
      "loss": 0.2165,
      "step": 3516
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1499503974260498e-05,
      "loss": 0.2579,
      "step": 3517
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1452936319598055e-05,
      "loss": 0.2472,
      "step": 3518
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1406459402263446e-05,
      "loss": 0.1662,
      "step": 3519
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1360073252695406e-05,
      "loss": 0.4757,
      "step": 3520
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1313777901273151e-05,
      "loss": 0.1901,
      "step": 3521
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.126757337831648e-05,
      "loss": 0.1068,
      "step": 3522
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1221459714085668e-05,
      "loss": 0.5846,
      "step": 3523
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.117543693878154e-05,
      "loss": 0.4412,
      "step": 3524
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.112950508254532e-05,
      "loss": 0.5112,
      "step": 3525
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1083664175458762e-05,
      "loss": 0.2348,
      "step": 3526
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1037914247543983e-05,
      "loss": 0.4148,
      "step": 3527
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.0992255328763599e-05,
      "loss": 0.6367,
      "step": 3528
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.0946687449020525e-05,
      "loss": 0.5007,
      "step": 3529
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.090121063815817e-05,
      "loss": 0.1848,
      "step": 3530
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.0855824925960176e-05,
      "loss": 0.2541,
      "step": 3531
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.0810530342150604e-05,
      "loss": 0.5249,
      "step": 3532
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.0765326916393819e-05,
      "loss": 0.596,
      "step": 3533
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.0720214678294443e-05,
      "loss": 0.2023,
      "step": 3534
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.0675193657397458e-05,
      "loss": 0.2685,
      "step": 3535
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.0630263883188011e-05,
      "loss": 0.5063,
      "step": 3536
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.0585425385091566e-05,
      "loss": 0.1679,
      "step": 3537
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.0540678192473784e-05,
      "loss": 0.2195,
      "step": 3538
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.0496022334640502e-05,
      "loss": 0.4619,
      "step": 3539
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.0451457840837813e-05,
      "loss": 0.3562,
      "step": 3540
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.0406984740251834e-05,
      "loss": 0.2483,
      "step": 3541
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.0362603062009001e-05,
      "loss": 0.4674,
      "step": 3542
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.031831283517573e-05,
      "loss": 0.5902,
      "step": 3543
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.027411408875865e-05,
      "loss": 0.1692,
      "step": 3544
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.0230006851704392e-05,
      "loss": 0.2863,
      "step": 3545
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.018599115289973e-05,
      "loss": 0.556,
      "step": 3546
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.0142067021171425e-05,
      "loss": 0.1306,
      "step": 3547
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.0098234485286277e-05,
      "loss": 0.2308,
      "step": 3548
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.0054493573951195e-05,
      "loss": 0.2934,
      "step": 3549
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.001084431581291e-05,
      "loss": 0.235,
      "step": 3550
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.967286739458308e-06,
      "loss": 0.1479,
      "step": 3551
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.923820873414067e-06,
      "loss": 0.3342,
      "step": 3552
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.880446746146936e-06,
      "loss": 0.4275,
      "step": 3553
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.837164386063473e-06,
      "loss": 0.338,
      "step": 3554
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.793973821510253e-06,
      "loss": 0.3899,
      "step": 3555
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.7508750807736e-06,
      "loss": 0.594,
      "step": 3556
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.707868192079788e-06,
      "loss": 0.2256,
      "step": 3557
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.664953183594965e-06,
      "loss": 0.2091,
      "step": 3558
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.622130083424984e-06,
      "loss": 0.1222,
      "step": 3559
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.579398919615627e-06,
      "loss": 0.3995,
      "step": 3560
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.536759720152376e-06,
      "loss": 0.5522,
      "step": 3561
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.494212512960558e-06,
      "loss": 0.2381,
      "step": 3562
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.451757325905168e-06,
      "loss": 0.2601,
      "step": 3563
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.409394186791047e-06,
      "loss": 0.1409,
      "step": 3564
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.367123123362619e-06,
      "loss": 0.114,
      "step": 3565
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.324944163304137e-06,
      "loss": 0.2589,
      "step": 3566
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.282857334239441e-06,
      "loss": 0.1398,
      "step": 3567
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.240862663732045e-06,
      "loss": 0.369,
      "step": 3568
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.198960179285169e-06,
      "loss": 0.3685,
      "step": 3569
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.15714990834156e-06,
      "loss": 0.5194,
      "step": 3570
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.115431878283697e-06,
      "loss": 0.3686,
      "step": 3571
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.07380611643349e-06,
      "loss": 0.133,
      "step": 3572
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.032272650052586e-06,
      "loss": 0.2574,
      "step": 3573
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.990831506342056e-06,
      "loss": 0.2431,
      "step": 3574
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.949482712442607e-06,
      "loss": 0.2066,
      "step": 3575
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.908226295434362e-06,
      "loss": 0.4775,
      "step": 3576
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.867062282337057e-06,
      "loss": 0.4618,
      "step": 3577
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.825990700109791e-06,
      "loss": 0.1864,
      "step": 3578
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.78501157565122e-06,
      "loss": 0.5064,
      "step": 3579
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.744124935799451e-06,
      "loss": 0.2268,
      "step": 3580
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.70333080733192e-06,
      "loss": 0.4476,
      "step": 3581
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.662629216965606e-06,
      "loss": 0.2597,
      "step": 3582
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.622020191356765e-06,
      "loss": 0.3066,
      "step": 3583
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.581503757101144e-06,
      "loss": 0.1851,
      "step": 3584
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.54107994073373e-06,
      "loss": 0.7258,
      "step": 3585
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.500748768728966e-06,
      "loss": 0.5541,
      "step": 3586
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.460510267500503e-06,
      "loss": 0.3825,
      "step": 3587
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.42036446340143e-06,
      "loss": 0.1609,
      "step": 3588
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.380311382724048e-06,
      "loss": 0.1968,
      "step": 3589
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.340351051699906e-06,
      "loss": 0.4144,
      "step": 3590
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.300483496499905e-06,
      "loss": 0.1729,
      "step": 3591
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.260708743234073e-06,
      "loss": 0.5918,
      "step": 3592
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.221026817951776e-06,
      "loss": 0.2581,
      "step": 3593
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.181437746641456e-06,
      "loss": 0.2054,
      "step": 3594
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.141941555230891e-06,
      "loss": 0.2753,
      "step": 3595
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.102538269586872e-06,
      "loss": 0.34,
      "step": 3596
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.063227915515474e-06,
      "loss": 0.3108,
      "step": 3597
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.02401051876182e-06,
      "loss": 0.0621,
      "step": 3598
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.984886105010229e-06,
      "loss": 0.2123,
      "step": 3599
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.945854699884075e-06,
      "loss": 0.4346,
      "step": 3600
    },
    {
      "epoch": 0.9,
      "eval_loss": 0.3306220471858978,
      "eval_runtime": 52.8726,
      "eval_samples_per_second": 0.624,
      "eval_steps_per_second": 0.624,
      "step": 3600
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.906916328945822e-06,
      "loss": 0.207,
      "step": 3601
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.868071017697025e-06,
      "loss": 0.3336,
      "step": 3602
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.829318791578282e-06,
      "loss": 0.4357,
      "step": 3603
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.790659675969174e-06,
      "loss": 0.3954,
      "step": 3604
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.752093696188433e-06,
      "loss": 0.1304,
      "step": 3605
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.713620877493638e-06,
      "loss": 0.2459,
      "step": 3606
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.675241245081481e-06,
      "loss": 0.378,
      "step": 3607
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.636954824087544e-06,
      "loss": 0.3614,
      "step": 3608
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.598761639586432e-06,
      "loss": 0.4425,
      "step": 3609
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.560661716591604e-06,
      "loss": 0.3345,
      "step": 3610
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.522655080055518e-06,
      "loss": 0.5744,
      "step": 3611
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.484741754869478e-06,
      "loss": 0.1323,
      "step": 3612
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.446921765863739e-06,
      "loss": 0.4484,
      "step": 3613
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.409195137807339e-06,
      "loss": 0.6978,
      "step": 3614
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.371561895408279e-06,
      "loss": 0.1818,
      "step": 3615
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.334022063313294e-06,
      "loss": 0.9165,
      "step": 3616
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.296575666108051e-06,
      "loss": 0.1987,
      "step": 3617
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.259222728316899e-06,
      "loss": 0.1649,
      "step": 3618
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.221963274403086e-06,
      "loss": 0.3416,
      "step": 3619
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.184797328768626e-06,
      "loss": 0.5291,
      "step": 3620
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.147724915754216e-06,
      "loss": 0.5904,
      "step": 3621
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.1107460596393666e-06,
      "loss": 0.0884,
      "step": 3622
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.0738607846422545e-06,
      "loss": 0.314,
      "step": 3623
    },
    {
      "epoch": 0.91,
      "learning_rate": 7.037069114919875e-06,
      "loss": 0.2879,
      "step": 3624
    },
    {
      "epoch": 0.91,
      "learning_rate": 7.000371074567785e-06,
      "loss": 0.2902,
      "step": 3625
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.9637666876203425e-06,
      "loss": 0.403,
      "step": 3626
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.927255978050455e-06,
      "loss": 0.2765,
      "step": 3627
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.89083896976978e-06,
      "loss": 0.3904,
      "step": 3628
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.85451568662857e-06,
      "loss": 0.3234,
      "step": 3629
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.818286152415664e-06,
      "loss": 0.1789,
      "step": 3630
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.782150390858582e-06,
      "loss": 0.1897,
      "step": 3631
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.746108425623309e-06,
      "loss": 0.2337,
      "step": 3632
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.71016028031453e-06,
      "loss": 0.4115,
      "step": 3633
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.67430597847538e-06,
      "loss": 0.7782,
      "step": 3634
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.6385455435876255e-06,
      "loss": 0.2299,
      "step": 3635
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.602878999071465e-06,
      "loss": 0.281,
      "step": 3636
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.567306368285713e-06,
      "loss": 0.3633,
      "step": 3637
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.531827674527551e-06,
      "loss": 0.528,
      "step": 3638
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.496442941032759e-06,
      "loss": 0.539,
      "step": 3639
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.4611521909754995e-06,
      "loss": 0.3084,
      "step": 3640
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.425955447468434e-06,
      "loss": 0.2109,
      "step": 3641
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.390852733562607e-06,
      "loss": 0.3862,
      "step": 3642
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.355844072247545e-06,
      "loss": 0.404,
      "step": 3643
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.320929486451093e-06,
      "loss": 0.2421,
      "step": 3644
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.286108999039591e-06,
      "loss": 0.3386,
      "step": 3645
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.251382632817681e-06,
      "loss": 0.4716,
      "step": 3646
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.2167504105283205e-06,
      "loss": 0.1823,
      "step": 3647
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.182212354852934e-06,
      "loss": 0.3301,
      "step": 3648
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.147768488411159e-06,
      "loss": 0.3589,
      "step": 3649
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.113418833761002e-06,
      "loss": 0.2589,
      "step": 3650
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.079163413398802e-06,
      "loss": 0.3261,
      "step": 3651
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.0450022497590785e-06,
      "loss": 0.4535,
      "step": 3652
    },
    {
      "epoch": 0.91,
      "learning_rate": 6.010935365214736e-06,
      "loss": 0.1825,
      "step": 3653
    },
    {
      "epoch": 0.91,
      "learning_rate": 5.976962782076844e-06,
      "loss": 0.4871,
      "step": 3654
    },
    {
      "epoch": 0.91,
      "learning_rate": 5.943084522594771e-06,
      "loss": 0.4621,
      "step": 3655
    },
    {
      "epoch": 0.91,
      "learning_rate": 5.909300608956052e-06,
      "loss": 0.2143,
      "step": 3656
    },
    {
      "epoch": 0.91,
      "learning_rate": 5.8756110632865065e-06,
      "loss": 0.3928,
      "step": 3657
    },
    {
      "epoch": 0.91,
      "learning_rate": 5.8420159076501e-06,
      "loss": 0.379,
      "step": 3658
    },
    {
      "epoch": 0.91,
      "learning_rate": 5.808515164048966e-06,
      "loss": 0.3044,
      "step": 3659
    },
    {
      "epoch": 0.91,
      "learning_rate": 5.775108854423488e-06,
      "loss": 0.5211,
      "step": 3660
    },
    {
      "epoch": 0.91,
      "learning_rate": 5.741797000652082e-06,
      "loss": 0.3178,
      "step": 3661
    },
    {
      "epoch": 0.91,
      "learning_rate": 5.708579624551429e-06,
      "loss": 0.7243,
      "step": 3662
    },
    {
      "epoch": 0.91,
      "learning_rate": 5.675456747876228e-06,
      "loss": 0.499,
      "step": 3663
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.6424283923193755e-06,
      "loss": 0.1566,
      "step": 3664
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.60949457951177e-06,
      "loss": 0.3311,
      "step": 3665
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.576655331022489e-06,
      "loss": 0.5994,
      "step": 3666
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.543910668358598e-06,
      "loss": 0.3252,
      "step": 3667
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.511260612965274e-06,
      "loss": 0.3218,
      "step": 3668
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.478705186225713e-06,
      "loss": 0.0679,
      "step": 3669
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.446244409461109e-06,
      "loss": 0.2667,
      "step": 3670
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.413878303930741e-06,
      "loss": 0.272,
      "step": 3671
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.381606890831785e-06,
      "loss": 0.4734,
      "step": 3672
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.34943019129952e-06,
      "loss": 0.6293,
      "step": 3673
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.317348226407054e-06,
      "loss": 0.0422,
      "step": 3674
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.2853610171655795e-06,
      "loss": 0.3364,
      "step": 3675
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.253468584524157e-06,
      "loss": 0.2477,
      "step": 3676
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.22167094936981e-06,
      "loss": 0.4282,
      "step": 3677
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.18996813252745e-06,
      "loss": 0.5912,
      "step": 3678
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.158360154759933e-06,
      "loss": 0.191,
      "step": 3679
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.126847036767933e-06,
      "loss": 0.4242,
      "step": 3680
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.095428799190093e-06,
      "loss": 0.1441,
      "step": 3681
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.064105462602819e-06,
      "loss": 0.1536,
      "step": 3682
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.032877047520417e-06,
      "loss": 0.3035,
      "step": 3683
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.001743574395045e-06,
      "loss": 0.4461,
      "step": 3684
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.9707050636166255e-06,
      "loss": 0.058,
      "step": 3685
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.939761535512965e-06,
      "loss": 0.3269,
      "step": 3686
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.908913010349552e-06,
      "loss": 0.3554,
      "step": 3687
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.878159508329776e-06,
      "loss": 0.0718,
      "step": 3688
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.84750104959471e-06,
      "loss": 0.3052,
      "step": 3689
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.816937654223224e-06,
      "loss": 0.2325,
      "step": 3690
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.786469342231908e-06,
      "loss": 0.0828,
      "step": 3691
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.756096133575099e-06,
      "loss": 0.3217,
      "step": 3692
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.725818048144836e-06,
      "loss": 0.1292,
      "step": 3693
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.69563510577084e-06,
      "loss": 0.2425,
      "step": 3694
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.665547326220564e-06,
      "loss": 0.5195,
      "step": 3695
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.635554729199114e-06,
      "loss": 0.3331,
      "step": 3696
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.605657334349261e-06,
      "loss": 0.2253,
      "step": 3697
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.575855161251391e-06,
      "loss": 0.2497,
      "step": 3698
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.546148229423624e-06,
      "loss": 0.4141,
      "step": 3699
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.516536558321582e-06,
      "loss": 0.4976,
      "step": 3700
    },
    {
      "epoch": 0.92,
      "eval_loss": 0.3302282392978668,
      "eval_runtime": 52.5398,
      "eval_samples_per_second": 0.628,
      "eval_steps_per_second": 0.628,
      "step": 3700
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.487020167338584e-06,
      "loss": 0.4427,
      "step": 3701
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.457599075805535e-06,
      "loss": 0.207,
      "step": 3702
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.428273302990887e-06,
      "loss": 0.3593,
      "step": 3703
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.399042868100727e-06,
      "loss": 0.1305,
      "step": 3704
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.3699077902786425e-06,
      "loss": 0.475,
      "step": 3705
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.3408680886058036e-06,
      "loss": 0.3593,
      "step": 3706
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.311923782100912e-06,
      "loss": 0.4276,
      "step": 3707
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.283074889720206e-06,
      "loss": 0.2669,
      "step": 3708
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.254321430357372e-06,
      "loss": 0.0605,
      "step": 3709
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.225663422843678e-06,
      "loss": 0.2713,
      "step": 3710
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.197100885947863e-06,
      "loss": 0.12,
      "step": 3711
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.168633838376079e-06,
      "loss": 0.3198,
      "step": 3712
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.140262298772012e-06,
      "loss": 0.4582,
      "step": 3713
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.111986285716751e-06,
      "loss": 0.1677,
      "step": 3714
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.083805817728863e-06,
      "loss": 0.395,
      "step": 3715
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.055720913264288e-06,
      "loss": 0.3845,
      "step": 3716
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.027731590716426e-06,
      "loss": 0.6937,
      "step": 3717
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.999837868416051e-06,
      "loss": 0.3997,
      "step": 3718
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.972039764631346e-06,
      "loss": 0.4431,
      "step": 3719
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.9443372975678676e-06,
      "loss": 0.6093,
      "step": 3720
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.916730485368502e-06,
      "loss": 0.1746,
      "step": 3721
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.88921934611357e-06,
      "loss": 0.358,
      "step": 3722
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.861803897820626e-06,
      "loss": 0.6143,
      "step": 3723
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.8344841584446565e-06,
      "loss": 0.229,
      "step": 3724
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.8072601458778994e-06,
      "loss": 0.3734,
      "step": 3725
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.7801318779499436e-06,
      "loss": 0.265,
      "step": 3726
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.753099372427626e-06,
      "loss": 0.3531,
      "step": 3727
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.726162647015102e-06,
      "loss": 0.3296,
      "step": 3728
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.6993217193537925e-06,
      "loss": 0.4529,
      "step": 3729
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.6725766070223538e-06,
      "loss": 0.2241,
      "step": 3730
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.6459273275367574e-06,
      "loss": 0.1276,
      "step": 3731
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.6193738983501088e-06,
      "loss": 0.1105,
      "step": 3732
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.592916336852847e-06,
      "loss": 0.5228,
      "step": 3733
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.566554660372528e-06,
      "loss": 0.4841,
      "step": 3734
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.540288886173992e-06,
      "loss": 0.4841,
      "step": 3735
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.5141190314592118e-06,
      "loss": 0.1861,
      "step": 3736
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.4880451133673937e-06,
      "loss": 0.3704,
      "step": 3737
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.4620671489748618e-06,
      "loss": 0.4318,
      "step": 3738
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.4361851552951235e-06,
      "loss": 0.3173,
      "step": 3739
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.4103991492788362e-06,
      "loss": 0.0869,
      "step": 3740
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.3847091478137744e-06,
      "loss": 0.3677,
      "step": 3741
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.359115167724863e-06,
      "loss": 0.307,
      "step": 3742
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.3336172257741266e-06,
      "loss": 0.3248,
      "step": 3743
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.3082153386606915e-06,
      "loss": 0.1937,
      "step": 3744
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.282909523020749e-06,
      "loss": 0.1215,
      "step": 3745
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.2576997954276597e-06,
      "loss": 0.3815,
      "step": 3746
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.2325861723917324e-06,
      "loss": 0.3803,
      "step": 3747
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.2075686703604444e-06,
      "loss": 0.3466,
      "step": 3748
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.1826473057182223e-06,
      "loss": 0.4366,
      "step": 3749
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.157822094786627e-06,
      "loss": 0.4728,
      "step": 3750
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.133093053824204e-06,
      "loss": 0.2707,
      "step": 3751
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.1084601990264635e-06,
      "loss": 0.6432,
      "step": 3752
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.083923546526018e-06,
      "loss": 0.2907,
      "step": 3753
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.0594831123923967e-06,
      "loss": 0.4854,
      "step": 3754
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.0351389126321624e-06,
      "loss": 0.0981,
      "step": 3755
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.0108909631888445e-06,
      "loss": 0.0749,
      "step": 3756
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.9867392799428735e-06,
      "loss": 0.3223,
      "step": 3757
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.9626838787117302e-06,
      "loss": 0.1294,
      "step": 3758
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.9387247752497787e-06,
      "loss": 0.5292,
      "step": 3759
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.9148619852483347e-06,
      "loss": 0.2929,
      "step": 3760
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.891095524335629e-06,
      "loss": 0.4531,
      "step": 3761
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.867425408076812e-06,
      "loss": 0.1614,
      "step": 3762
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.8438516519738995e-06,
      "loss": 0.6529,
      "step": 3763
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.8203742714658915e-06,
      "loss": 0.5962,
      "step": 3764
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.796993281928539e-06,
      "loss": 0.1981,
      "step": 3765
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.7737086986745925e-06,
      "loss": 0.3816,
      "step": 3766
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.7505205369535542e-06,
      "loss": 0.4042,
      "step": 3767
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.727428811951876e-06,
      "loss": 0.6413,
      "step": 3768
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.704433538792761e-06,
      "loss": 0.2977,
      "step": 3769
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.6815347325363445e-06,
      "loss": 0.4233,
      "step": 3770
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.6587324081794646e-06,
      "loss": 0.2196,
      "step": 3771
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.636026580655859e-06,
      "loss": 0.6237,
      "step": 3772
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.6134172648360653e-06,
      "loss": 0.263,
      "step": 3773
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.5909044755273554e-06,
      "loss": 0.2998,
      "step": 3774
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.568488227473836e-06,
      "loss": 0.2601,
      "step": 3775
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.546168535356363e-06,
      "loss": 0.4548,
      "step": 3776
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.523945413792594e-06,
      "loss": 0.5518,
      "step": 3777
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.5018188773368532e-06,
      "loss": 0.1159,
      "step": 3778
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.4797889404802982e-06,
      "loss": 0.3147,
      "step": 3779
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.457855617650789e-06,
      "loss": 0.2439,
      "step": 3780
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.436018923212901e-06,
      "loss": 0.1459,
      "step": 3781
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.4142788714679616e-06,
      "loss": 0.5119,
      "step": 3782
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.392635476653948e-06,
      "loss": 0.1685,
      "step": 3783
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.371088752945571e-06,
      "loss": 0.2809,
      "step": 3784
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.349638714454244e-06,
      "loss": 0.1873,
      "step": 3785
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.3282853752280117e-06,
      "loss": 0.3786,
      "step": 3786
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.3070287492516383e-06,
      "loss": 0.2257,
      "step": 3787
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.2858688504465205e-06,
      "loss": 0.4534,
      "step": 3788
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.2648056926706902e-06,
      "loss": 0.414,
      "step": 3789
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.2438392897188795e-06,
      "loss": 0.4652,
      "step": 3790
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.2229696553223874e-06,
      "loss": 0.2144,
      "step": 3791
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.202196803149198e-06,
      "loss": 0.4017,
      "step": 3792
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.1815207468038443e-06,
      "loss": 0.1314,
      "step": 3793
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.1609414998275455e-06,
      "loss": 0.3586,
      "step": 3794
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.140459075698037e-06,
      "loss": 0.1275,
      "step": 3795
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.1200734878296877e-06,
      "loss": 0.2002,
      "step": 3796
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.099784749573452e-06,
      "loss": 0.4057,
      "step": 3797
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.0795928742168342e-06,
      "loss": 0.7897,
      "step": 3798
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.059497874983923e-06,
      "loss": 0.5639,
      "step": 3799
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.039499765035324e-06,
      "loss": 0.4025,
      "step": 3800
    },
    {
      "epoch": 0.95,
      "eval_loss": 0.32995572686195374,
      "eval_runtime": 52.975,
      "eval_samples_per_second": 0.623,
      "eval_steps_per_second": 0.623,
      "step": 3800
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.0195985574682104e-06,
      "loss": 0.4397,
      "step": 3801
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.9997942653163403e-06,
      "loss": 0.3195,
      "step": 3802
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.9800869015498876e-06,
      "loss": 0.4228,
      "step": 3803
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.960476479075662e-06,
      "loss": 0.539,
      "step": 3804
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.940963010736907e-06,
      "loss": 0.3701,
      "step": 3805
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.921546509313415e-06,
      "loss": 0.2059,
      "step": 3806
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.9022269875214158e-06,
      "loss": 0.2291,
      "step": 3807
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.8830044580137038e-06,
      "loss": 0.4507,
      "step": 3808
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.8638789333794923e-06,
      "loss": 0.2845,
      "step": 3809
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.8448504261444785e-06,
      "loss": 0.3561,
      "step": 3810
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.825918948770827e-06,
      "loss": 0.2658,
      "step": 3811
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.8070845136571532e-06,
      "loss": 0.3067,
      "step": 3812
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.7883471331385234e-06,
      "loss": 0.3668,
      "step": 3813
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.7697068194864216e-06,
      "loss": 0.3236,
      "step": 3814
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.7511635849087824e-06,
      "loss": 0.5399,
      "step": 3815
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.7327174415499577e-06,
      "loss": 0.3679,
      "step": 3816
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.7143684014906844e-06,
      "loss": 0.3575,
      "step": 3817
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.6961164767481495e-06,
      "loss": 0.2714,
      "step": 3818
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.6779616792759077e-06,
      "loss": 0.4665,
      "step": 3819
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.6599040209639148e-06,
      "loss": 0.1229,
      "step": 3820
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.641943513638494e-06,
      "loss": 0.2632,
      "step": 3821
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.6240801690623528e-06,
      "loss": 0.3501,
      "step": 3822
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.6063139989345663e-06,
      "loss": 0.4084,
      "step": 3823
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.5886450148905604e-06,
      "loss": 0.6466,
      "step": 3824
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.5710732285021287e-06,
      "loss": 0.2535,
      "step": 3825
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.553598651277399e-06,
      "loss": 0.7585,
      "step": 3826
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.5362212946608165e-06,
      "loss": 0.3271,
      "step": 3827
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.5189411700331777e-06,
      "loss": 0.2496,
      "step": 3828
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.5017582887115797e-06,
      "loss": 0.4543,
      "step": 3829
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.4846726619494875e-06,
      "loss": 0.0957,
      "step": 3830
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.4676843009365668e-06,
      "loss": 0.0751,
      "step": 3831
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.450793216798901e-06,
      "loss": 0.293,
      "step": 3832
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.4339994205987748e-06,
      "loss": 0.5423,
      "step": 3833
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.417302923334823e-06,
      "loss": 0.6455,
      "step": 3834
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.4007037359418992e-06,
      "loss": 0.3719,
      "step": 3835
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3842018692911571e-06,
      "loss": 0.3354,
      "step": 3836
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3677973341900184e-06,
      "loss": 0.564,
      "step": 3837
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.351490141382139e-06,
      "loss": 0.367,
      "step": 3838
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3352803015474422e-06,
      "loss": 0.1032,
      "step": 3839
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3191678253020865e-06,
      "loss": 0.7238,
      "step": 3840
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.303152723198464e-06,
      "loss": 0.4667,
      "step": 3841
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.2872350057252012e-06,
      "loss": 0.411,
      "step": 3842
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.271414683307126e-06,
      "loss": 0.1205,
      "step": 3843
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.2556917663053167e-06,
      "loss": 0.1004,
      "step": 3844
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.2400662650170202e-06,
      "loss": 0.2227,
      "step": 3845
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.2245381896757e-06,
      "loss": 0.3531,
      "step": 3846
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.209107550451005e-06,
      "loss": 0.4261,
      "step": 3847
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.1937743574487845e-06,
      "loss": 0.1352,
      "step": 3848
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.1785386207110724e-06,
      "loss": 0.2739,
      "step": 3849
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.163400350216054e-06,
      "loss": 0.2681,
      "step": 3850
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.148359555878098e-06,
      "loss": 0.6574,
      "step": 3851
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.1334162475477415e-06,
      "loss": 0.3355,
      "step": 3852
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.1185704350116564e-06,
      "loss": 0.3591,
      "step": 3853
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.103822127992665e-06,
      "loss": 0.6417,
      "step": 3854
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.0891713361497578e-06,
      "loss": 0.2541,
      "step": 3855
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.0746180690780093e-06,
      "loss": 0.3683,
      "step": 3856
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.0601623363086787e-06,
      "loss": 0.0452,
      "step": 3857
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.0458041473091093e-06,
      "loss": 0.2757,
      "step": 3858
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.0315435114827962e-06,
      "loss": 0.5181,
      "step": 3859
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.0173804381693017e-06,
      "loss": 0.3749,
      "step": 3860
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.0033149366443227e-06,
      "loss": 0.1694,
      "step": 3861
    },
    {
      "epoch": 0.96,
      "learning_rate": 9.893470161196403e-07,
      "loss": 0.5238,
      "step": 3862
    },
    {
      "epoch": 0.96,
      "learning_rate": 9.75476685743154e-07,
      "loss": 0.5429,
      "step": 3863
    },
    {
      "epoch": 0.97,
      "learning_rate": 9.61703954598797e-07,
      "loss": 0.1106,
      "step": 3864
    },
    {
      "epoch": 0.97,
      "learning_rate": 9.480288317066376e-07,
      "loss": 0.357,
      "step": 3865
    },
    {
      "epoch": 0.97,
      "learning_rate": 9.344513260227949e-07,
      "loss": 0.3664,
      "step": 3866
    },
    {
      "epoch": 0.97,
      "learning_rate": 9.209714464394391e-07,
      "loss": 0.3388,
      "step": 3867
    },
    {
      "epoch": 0.97,
      "learning_rate": 9.075892017848419e-07,
      "loss": 0.7511,
      "step": 3868
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.943046008233091e-07,
      "loss": 0.3094,
      "step": 3869
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.811176522551977e-07,
      "loss": 0.3819,
      "step": 3870
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.680283647168995e-07,
      "loss": 0.3716,
      "step": 3871
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.550367467808572e-07,
      "loss": 0.1599,
      "step": 3872
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.421428069555813e-07,
      "loss": 0.2756,
      "step": 3873
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.293465536855337e-07,
      "loss": 0.2574,
      "step": 3874
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.166479953512939e-07,
      "loss": 0.3646,
      "step": 3875
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.040471402693594e-07,
      "loss": 0.3512,
      "step": 3876
    },
    {
      "epoch": 0.97,
      "learning_rate": 7.915439966923287e-07,
      "loss": 0.2001,
      "step": 3877
    },
    {
      "epoch": 0.97,
      "learning_rate": 7.791385728087518e-07,
      "loss": 0.4071,
      "step": 3878
    },
    {
      "epoch": 0.97,
      "learning_rate": 7.668308767431963e-07,
      "loss": 0.4752,
      "step": 3879
    },
    {
      "epoch": 0.97,
      "learning_rate": 7.546209165562312e-07,
      "loss": 0.2503,
      "step": 3880
    },
    {
      "epoch": 0.97,
      "learning_rate": 7.42508700244393e-07,
      "loss": 0.2234,
      "step": 3881
    },
    {
      "epoch": 0.97,
      "learning_rate": 7.304942357402532e-07,
      "loss": 0.3033,
      "step": 3882
    },
    {
      "epoch": 0.97,
      "learning_rate": 7.18577530912301e-07,
      "loss": 0.2804,
      "step": 3883
    },
    {
      "epoch": 0.97,
      "learning_rate": 7.067585935650599e-07,
      "loss": 0.6261,
      "step": 3884
    },
    {
      "epoch": 0.97,
      "learning_rate": 6.950374314389883e-07,
      "loss": 0.2447,
      "step": 3885
    },
    {
      "epoch": 0.97,
      "learning_rate": 6.834140522104959e-07,
      "loss": 0.3629,
      "step": 3886
    },
    {
      "epoch": 0.97,
      "learning_rate": 6.71888463491993e-07,
      "loss": 0.2987,
      "step": 3887
    },
    {
      "epoch": 0.97,
      "learning_rate": 6.604606728318251e-07,
      "loss": 0.2506,
      "step": 3888
    },
    {
      "epoch": 0.97,
      "learning_rate": 6.491306877142722e-07,
      "loss": 0.3787,
      "step": 3889
    },
    {
      "epoch": 0.97,
      "learning_rate": 6.378985155595983e-07,
      "loss": 0.3006,
      "step": 3890
    },
    {
      "epoch": 0.97,
      "learning_rate": 6.267641637239695e-07,
      "loss": 0.1306,
      "step": 3891
    },
    {
      "epoch": 0.97,
      "learning_rate": 6.157276394995192e-07,
      "loss": 0.2016,
      "step": 3892
    },
    {
      "epoch": 0.97,
      "learning_rate": 6.047889501142655e-07,
      "loss": 0.3583,
      "step": 3893
    },
    {
      "epoch": 0.97,
      "learning_rate": 5.939481027321946e-07,
      "loss": 0.4345,
      "step": 3894
    },
    {
      "epoch": 0.97,
      "learning_rate": 5.832051044532105e-07,
      "loss": 0.3144,
      "step": 3895
    },
    {
      "epoch": 0.97,
      "learning_rate": 5.725599623131017e-07,
      "loss": 0.3369,
      "step": 3896
    },
    {
      "epoch": 0.97,
      "learning_rate": 5.620126832836247e-07,
      "loss": 0.3251,
      "step": 3897
    },
    {
      "epoch": 0.97,
      "learning_rate": 5.515632742723875e-07,
      "loss": 0.316,
      "step": 3898
    },
    {
      "epoch": 0.97,
      "learning_rate": 5.412117421229323e-07,
      "loss": 0.3019,
      "step": 3899
    },
    {
      "epoch": 0.97,
      "learning_rate": 5.309580936146696e-07,
      "loss": 0.0857,
      "step": 3900
    },
    {
      "epoch": 0.97,
      "eval_loss": 0.32989436388015747,
      "eval_runtime": 50.6354,
      "eval_samples_per_second": 0.652,
      "eval_steps_per_second": 0.652,
      "step": 3900
    },
    {
      "epoch": 0.97,
      "learning_rate": 5.20802335462961e-07,
      "loss": 0.3665,
      "step": 3901
    },
    {
      "epoch": 0.97,
      "learning_rate": 5.107444743189859e-07,
      "loss": 0.3519,
      "step": 3902
    },
    {
      "epoch": 0.97,
      "learning_rate": 5.007845167698588e-07,
      "loss": 0.286,
      "step": 3903
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.909224693385782e-07,
      "loss": 0.1775,
      "step": 3904
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.811583384839945e-07,
      "loss": 0.1482,
      "step": 3905
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.714921306008257e-07,
      "loss": 0.2805,
      "step": 3906
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.619238520196744e-07,
      "loss": 0.3049,
      "step": 3907
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.524535090069947e-07,
      "loss": 0.4106,
      "step": 3908
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.430811077651419e-07,
      "loss": 0.2437,
      "step": 3909
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.3380665443225603e-07,
      "loss": 0.3406,
      "step": 3910
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.2463015508239496e-07,
      "loss": 0.0688,
      "step": 3911
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.1555161572541794e-07,
      "loss": 0.325,
      "step": 3912
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.065710423070689e-07,
      "loss": 0.2804,
      "step": 3913
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.9768844070889316e-07,
      "loss": 0.4491,
      "step": 3914
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.889038167482872e-07,
      "loss": 0.6525,
      "step": 3915
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.802171761784989e-07,
      "loss": 0.2392,
      "step": 3916
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.7162852468857753e-07,
      "loss": 0.3381,
      "step": 3917
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.631378679034069e-07,
      "loss": 0.3082,
      "step": 3918
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.5474521138370573e-07,
      "loss": 0.2646,
      "step": 3919
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.464505606259771e-07,
      "loss": 0.7028,
      "step": 3920
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.3825392106257565e-07,
      "loss": 0.2747,
      "step": 3921
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.3015529806164063e-07,
      "loss": 0.3408,
      "step": 3922
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.22154696927146e-07,
      "loss": 0.3429,
      "step": 3923
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.1425212289881707e-07,
      "loss": 0.5602,
      "step": 3924
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.0644758115223046e-07,
      "loss": 0.1956,
      "step": 3925
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.987410767987475e-07,
      "loss": 0.3929,
      "step": 3926
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.9113261488551423e-07,
      "loss": 0.1841,
      "step": 3927
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.8362220039546135e-07,
      "loss": 0.5273,
      "step": 3928
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.76209838247321e-07,
      "loss": 0.6876,
      "step": 3929
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.688955332955933e-07,
      "loss": 0.5008,
      "step": 3930
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.6167929033057977e-07,
      "loss": 0.262,
      "step": 3931
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.545611140783332e-07,
      "loss": 0.2239,
      "step": 3932
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.475410092006913e-07,
      "loss": 0.3559,
      "step": 3933
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.4061898029529294e-07,
      "loss": 0.2571,
      "step": 3934
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.3379503189549508e-07,
      "loss": 0.2684,
      "step": 3935
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.2706916847043954e-07,
      "loss": 0.1954,
      "step": 3936
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.204413944250527e-07,
      "loss": 0.6028,
      "step": 3937
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.139117140999791e-07,
      "loss": 0.1125,
      "step": 3938
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.0748013177164792e-07,
      "loss": 0.2954,
      "step": 3939
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.0114665165223975e-07,
      "loss": 0.5173,
      "step": 3940
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.9491127788968663e-07,
      "loss": 0.515,
      "step": 3941
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.8877401456763864e-07,
      "loss": 0.3571,
      "step": 3942
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.8273486570554717e-07,
      "loss": 0.245,
      "step": 3943
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.7679383525854852e-07,
      "loss": 0.1331,
      "step": 3944
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.7095092711756352e-07,
      "loss": 0.5101,
      "step": 3945
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.6520614510923125e-07,
      "loss": 0.308,
      "step": 3946
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.595594929959254e-07,
      "loss": 0.1653,
      "step": 3947
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.5401097447573785e-07,
      "loss": 0.2384,
      "step": 3948
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4856059318254508e-07,
      "loss": 0.5334,
      "step": 3949
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4320835268587516e-07,
      "loss": 0.3773,
      "step": 3950
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3795425649104074e-07,
      "loss": 0.3481,
      "step": 3951
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.32798308039056e-07,
      "loss": 0.23,
      "step": 3952
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.277405107066698e-07,
      "loss": 0.4099,
      "step": 3953
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.227808678063158e-07,
      "loss": 0.0958,
      "step": 3954
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.1791938258617906e-07,
      "loss": 0.081,
      "step": 3955
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.1315605823012941e-07,
      "loss": 0.4467,
      "step": 3956
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.0849089785778808e-07,
      "loss": 0.1146,
      "step": 3957
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.0392390452446109e-07,
      "loss": 0.518,
      "step": 3958
    },
    {
      "epoch": 0.99,
      "learning_rate": 9.945508122115586e-08,
      "loss": 0.4893,
      "step": 3959
    },
    {
      "epoch": 0.99,
      "learning_rate": 9.50844308746146e-08,
      "loss": 0.3373,
      "step": 3960
    },
    {
      "epoch": 0.99,
      "learning_rate": 9.081195634724758e-08,
      "loss": 0.2175,
      "step": 3961
    },
    {
      "epoch": 0.99,
      "learning_rate": 8.663766043718323e-08,
      "loss": 0.5233,
      "step": 3962
    },
    {
      "epoch": 0.99,
      "learning_rate": 8.256154587826801e-08,
      "loss": 0.5394,
      "step": 3963
    },
    {
      "epoch": 0.99,
      "learning_rate": 7.858361534003321e-08,
      "loss": 0.2644,
      "step": 3964
    },
    {
      "epoch": 0.99,
      "learning_rate": 7.47038714277115e-08,
      "loss": 0.2881,
      "step": 3965
    },
    {
      "epoch": 0.99,
      "learning_rate": 7.092231668220372e-08,
      "loss": 0.3568,
      "step": 3966
    },
    {
      "epoch": 0.99,
      "learning_rate": 6.723895358016207e-08,
      "loss": 0.1805,
      "step": 3967
    },
    {
      "epoch": 0.99,
      "learning_rate": 6.36537845338736e-08,
      "loss": 0.5037,
      "step": 3968
    },
    {
      "epoch": 0.99,
      "learning_rate": 6.016681189132678e-08,
      "loss": 0.4577,
      "step": 3969
    },
    {
      "epoch": 0.99,
      "learning_rate": 5.67780379362448e-08,
      "loss": 0.4725,
      "step": 3970
    },
    {
      "epoch": 0.99,
      "learning_rate": 5.348746488800237e-08,
      "loss": 0.3706,
      "step": 3971
    },
    {
      "epoch": 0.99,
      "learning_rate": 5.029509490162564e-08,
      "loss": 0.3998,
      "step": 3972
    },
    {
      "epoch": 0.99,
      "learning_rate": 4.720093006790881e-08,
      "loss": 0.4557,
      "step": 3973
    },
    {
      "epoch": 0.99,
      "learning_rate": 4.420497241326426e-08,
      "loss": 0.7032,
      "step": 3974
    },
    {
      "epoch": 0.99,
      "learning_rate": 4.130722389978913e-08,
      "loss": 0.1815,
      "step": 3975
    },
    {
      "epoch": 0.99,
      "learning_rate": 3.850768642531532e-08,
      "loss": 0.3411,
      "step": 3976
    },
    {
      "epoch": 0.99,
      "learning_rate": 3.5806361823292886e-08,
      "loss": 0.2294,
      "step": 3977
    },
    {
      "epoch": 0.99,
      "learning_rate": 3.320325186287332e-08,
      "loss": 0.2459,
      "step": 3978
    },
    {
      "epoch": 0.99,
      "learning_rate": 3.0698358248909535e-08,
      "loss": 0.1792,
      "step": 3979
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.82916826218893e-08,
      "loss": 0.3147,
      "step": 3980
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.5983226557985126e-08,
      "loss": 0.2406,
      "step": 3981
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.3772991569070977e-08,
      "loss": 0.2682,
      "step": 3982
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.1660979102688935e-08,
      "loss": 0.3181,
      "step": 3983
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.964719054199926e-08,
      "loss": 0.2485,
      "step": 3984
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.7731627205896937e-08,
      "loss": 0.4617,
      "step": 3985
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.591429034892844e-08,
      "loss": 0.3077,
      "step": 3986
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.4195181161291702e-08,
      "loss": 0.3873,
      "step": 3987
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.257430076886945e-08,
      "loss": 0.3703,
      "step": 3988
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.1051650233229182e-08,
      "loss": 0.5948,
      "step": 3989
    },
    {
      "epoch": 1.0,
      "learning_rate": 9.627230551556564e-09,
      "loss": 0.3265,
      "step": 3990
    },
    {
      "epoch": 1.0,
      "learning_rate": 8.301042656755353e-09,
      "loss": 0.151,
      "step": 3991
    },
    {
      "epoch": 1.0,
      "learning_rate": 7.0730874173641204e-09,
      "loss": 0.3299,
      "step": 3992
    },
    {
      "epoch": 1.0,
      "learning_rate": 5.94336563758957e-09,
      "loss": 0.4671,
      "step": 3993
    },
    {
      "epoch": 1.0,
      "learning_rate": 4.911878057323182e-09,
      "loss": 0.2807,
      "step": 3994
    },
    {
      "epoch": 1.0,
      "learning_rate": 3.978625352091258e-09,
      "loss": 0.3064,
      "step": 3995
    },
    {
      "epoch": 1.0,
      "learning_rate": 3.1436081331215333e-09,
      "loss": 0.2956,
      "step": 3996
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.406826947276563e-09,
      "loss": 0.4377,
      "step": 3997
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.7682822770870297e-09,
      "loss": 0.1302,
      "step": 3998
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.2279745407350883e-09,
      "loss": 0.1975,
      "step": 3999
    },
    {
      "epoch": 1.0,
      "learning_rate": 7.85904092104328e-10,
      "loss": 0.2589,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.3298940062522888,
      "eval_runtime": 49.678,
      "eval_samples_per_second": 0.664,
      "eval_steps_per_second": 0.664,
      "step": 4000
    }
  ],
  "logging_steps": 1,
  "max_steps": 4003,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 1.0784325637450383e+18,
  "trial_name": null,
  "trial_params": null
}