{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 20,
  "global_step": 1125,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 5e-06,
      "loss": 1.0235,
      "step": 1
    },
    {
      "epoch": 0.01,
      "learning_rate": 1e-05,
      "loss": 1.0135,
      "step": 2
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.5e-05,
      "loss": 1.0056,
      "step": 3
    },
    {
      "epoch": 0.02,
      "learning_rate": 2e-05,
      "loss": 1.073,
      "step": 4
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.5e-05,
      "loss": 1.052,
      "step": 5
    },
    {
      "epoch": 0.03,
      "learning_rate": 3e-05,
      "loss": 0.9612,
      "step": 6
    },
    {
      "epoch": 0.03,
      "learning_rate": 3.5e-05,
      "loss": 1.0725,
      "step": 7
    },
    {
      "epoch": 0.04,
      "learning_rate": 4e-05,
      "loss": 1.045,
      "step": 8
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.5e-05,
      "loss": 1.0201,
      "step": 9
    },
    {
      "epoch": 0.04,
      "learning_rate": 5e-05,
      "loss": 0.9867,
      "step": 10
    },
    {
      "epoch": 0.05,
      "learning_rate": 5.500000000000001e-05,
      "loss": 1.0,
      "step": 11
    },
    {
      "epoch": 0.05,
      "learning_rate": 6e-05,
      "loss": 1.0433,
      "step": 12
    },
    {
      "epoch": 0.06,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.9577,
      "step": 13
    },
    {
      "epoch": 0.06,
      "learning_rate": 7e-05,
      "loss": 0.9348,
      "step": 14
    },
    {
      "epoch": 0.07,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.0122,
      "step": 15
    },
    {
      "epoch": 0.07,
      "learning_rate": 8e-05,
      "loss": 1.0221,
      "step": 16
    },
    {
      "epoch": 0.08,
      "learning_rate": 8.5e-05,
      "loss": 0.911,
      "step": 17
    },
    {
      "epoch": 0.08,
      "learning_rate": 9e-05,
      "loss": 0.8725,
      "step": 18
    },
    {
      "epoch": 0.08,
      "learning_rate": 9.5e-05,
      "loss": 0.9493,
      "step": 19
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001,
      "loss": 0.9235,
      "step": 20
    },
    {
      "epoch": 0.09,
      "eval_loss": 0.9229408502578735,
      "eval_runtime": 16.01,
      "eval_samples_per_second": 5.059,
      "eval_steps_per_second": 1.312,
      "step": 20
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.999979792392145e-05,
      "loss": 0.8717,
      "step": 21
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.999919169731915e-05,
      "loss": 0.929,
      "step": 22
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.999818132509326e-05,
      "loss": 0.9449,
      "step": 23
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.99967668154107e-05,
      "loss": 0.911,
      "step": 24
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.999494817970498e-05,
      "loss": 0.9356,
      "step": 25
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.999272543267621e-05,
      "loss": 0.9319,
      "step": 26
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.999009859229096e-05,
      "loss": 0.8534,
      "step": 27
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.998706767978208e-05,
      "loss": 0.8563,
      "step": 28
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.998363271964859e-05,
      "loss": 0.9176,
      "step": 29
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.997979373965541e-05,
      "loss": 0.8752,
      "step": 30
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.997555077083319e-05,
      "loss": 0.9263,
      "step": 31
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.9970903847478e-05,
      "loss": 0.8348,
      "step": 32
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.996585300715116e-05,
      "loss": 0.9279,
      "step": 33
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.99603982906788e-05,
      "loss": 0.8997,
      "step": 34
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.995453974215163e-05,
      "loss": 0.9019,
      "step": 35
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.994827740892458e-05,
      "loss": 0.865,
      "step": 36
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.994161134161634e-05,
      "loss": 0.8226,
      "step": 37
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.9934541594109e-05,
      "loss": 0.8094,
      "step": 38
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.992706822354768e-05,
      "loss": 0.8706,
      "step": 39
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.991919129033994e-05,
      "loss": 0.8544,
      "step": 40
    },
    {
      "epoch": 0.18,
      "eval_loss": 0.8436371684074402,
      "eval_runtime": 16.1045,
      "eval_samples_per_second": 5.03,
      "eval_steps_per_second": 1.304,
      "step": 40
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.991091085815532e-05,
      "loss": 0.8041,
      "step": 41
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.990222699392498e-05,
      "loss": 0.8563,
      "step": 42
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.989313976784094e-05,
      "loss": 0.8284,
      "step": 43
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.988364925335565e-05,
      "loss": 0.8608,
      "step": 44
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.987375552718133e-05,
      "loss": 0.8893,
      "step": 45
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.986345866928941e-05,
      "loss": 0.8383,
      "step": 46
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.985275876290982e-05,
      "loss": 0.7732,
      "step": 47
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.98416558945304e-05,
      "loss": 0.8397,
      "step": 48
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.983015015389607e-05,
      "loss": 0.8497,
      "step": 49
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.981824163400827e-05,
      "loss": 0.8805,
      "step": 50
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.980593043112404e-05,
      "loss": 0.8751,
      "step": 51
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.979321664475541e-05,
      "loss": 0.849,
      "step": 52
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.978010037766841e-05,
      "loss": 0.8635,
      "step": 53
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.976658173588244e-05,
      "loss": 0.7918,
      "step": 54
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.975266082866923e-05,
      "loss": 0.8353,
      "step": 55
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.97383377685521e-05,
      "loss": 0.8522,
      "step": 56
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.972361267130495e-05,
      "loss": 0.8983,
      "step": 57
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.970848565595137e-05,
      "loss": 0.8749,
      "step": 58
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.96929568447637e-05,
      "loss": 0.8603,
      "step": 59
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.967702636326194e-05,
      "loss": 0.8139,
      "step": 60
    },
    {
      "epoch": 0.27,
      "eval_loss": 0.8146741986274719,
      "eval_runtime": 16.1449,
      "eval_samples_per_second": 5.017,
      "eval_steps_per_second": 1.301,
      "step": 60
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.966069434021293e-05,
      "loss": 0.7807,
      "step": 61
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.964396090762908e-05,
      "loss": 0.7711,
      "step": 62
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.962682620076744e-05,
      "loss": 0.8169,
      "step": 63
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.96092903581286e-05,
      "loss": 0.8259,
      "step": 64
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.959135352145552e-05,
      "loss": 0.8244,
      "step": 65
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.957301583573244e-05,
      "loss": 0.7796,
      "step": 66
    },
    {
      "epoch": 0.3,
      "learning_rate": 9.955427744918367e-05,
      "loss": 0.8131,
      "step": 67
    },
    {
      "epoch": 0.3,
      "learning_rate": 9.953513851327236e-05,
      "loss": 0.7746,
      "step": 68
    },
    {
      "epoch": 0.31,
      "learning_rate": 9.951559918269939e-05,
      "loss": 0.7514,
      "step": 69
    },
    {
      "epoch": 0.31,
      "learning_rate": 9.9495659615402e-05,
      "loss": 0.7851,
      "step": 70
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.947531997255256e-05,
      "loss": 0.8309,
      "step": 71
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.94545804185573e-05,
      "loss": 0.7847,
      "step": 72
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.943344112105492e-05,
      "loss": 0.844,
      "step": 73
    },
    {
      "epoch": 0.33,
      "learning_rate": 9.94119022509153e-05,
      "loss": 0.8984,
      "step": 74
    },
    {
      "epoch": 0.33,
      "learning_rate": 9.938996398223801e-05,
      "loss": 0.7758,
      "step": 75
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.936762649235105e-05,
      "loss": 0.827,
      "step": 76
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.934488996180931e-05,
      "loss": 0.8177,
      "step": 77
    },
    {
      "epoch": 0.35,
      "learning_rate": 9.932175457439317e-05,
      "loss": 0.8236,
      "step": 78
    },
    {
      "epoch": 0.35,
      "learning_rate": 9.92982205171069e-05,
      "loss": 0.7999,
      "step": 79
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.927428798017738e-05,
      "loss": 0.8399,
      "step": 80
    },
    {
      "epoch": 0.36,
      "eval_loss": 0.7993953227996826,
      "eval_runtime": 16.1588,
      "eval_samples_per_second": 5.013,
      "eval_steps_per_second": 1.3,
      "step": 80
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.924995715705229e-05,
      "loss": 0.7619,
      "step": 81
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.922522824439872e-05,
      "loss": 0.8059,
      "step": 82
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.920010144210157e-05,
      "loss": 0.797,
      "step": 83
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.917457695326184e-05,
      "loss": 0.804,
      "step": 84
    },
    {
      "epoch": 0.38,
      "learning_rate": 9.91486549841951e-05,
      "loss": 0.8354,
      "step": 85
    },
    {
      "epoch": 0.38,
      "learning_rate": 9.912233574442971e-05,
      "loss": 0.831,
      "step": 86
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.909561944670525e-05,
      "loss": 0.8985,
      "step": 87
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.906850630697068e-05,
      "loss": 0.7296,
      "step": 88
    },
    {
      "epoch": 0.4,
      "learning_rate": 9.90409965443827e-05,
      "loss": 0.8307,
      "step": 89
    },
    {
      "epoch": 0.4,
      "learning_rate": 9.90130903813039e-05,
      "loss": 0.8038,
      "step": 90
    },
    {
      "epoch": 0.4,
      "learning_rate": 9.8984788043301e-05,
      "loss": 0.7976,
      "step": 91
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.895608975914304e-05,
      "loss": 0.7875,
      "step": 92
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.892699576079944e-05,
      "loss": 0.8176,
      "step": 93
    },
    {
      "epoch": 0.42,
      "learning_rate": 9.88975062834383e-05,
      "loss": 0.7454,
      "step": 94
    },
    {
      "epoch": 0.42,
      "learning_rate": 9.886762156542428e-05,
      "loss": 0.8036,
      "step": 95
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.88373418483169e-05,
      "loss": 0.7755,
      "step": 96
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.880666737686838e-05,
      "loss": 0.8433,
      "step": 97
    },
    {
      "epoch": 0.44,
      "learning_rate": 9.877559839902184e-05,
      "loss": 0.846,
      "step": 98
    },
    {
      "epoch": 0.44,
      "learning_rate": 9.874413516590912e-05,
      "loss": 0.7174,
      "step": 99
    },
    {
      "epoch": 0.44,
      "learning_rate": 9.871227793184892e-05,
      "loss": 0.8729,
      "step": 100
    },
    {
      "epoch": 0.44,
      "eval_loss": 0.792314887046814,
      "eval_runtime": 16.1661,
      "eval_samples_per_second": 5.01,
      "eval_steps_per_second": 1.299,
      "step": 100
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.868002695434462e-05,
      "loss": 0.7872,
      "step": 101
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.864738249408227e-05,
      "loss": 0.7842,
      "step": 102
    },
    {
      "epoch": 0.46,
      "learning_rate": 9.861434481492846e-05,
      "loss": 0.8254,
      "step": 103
    },
    {
      "epoch": 0.46,
      "learning_rate": 9.858091418392815e-05,
      "loss": 0.795,
      "step": 104
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.85470908713026e-05,
      "loss": 0.7682,
      "step": 105
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.85128751504471e-05,
      "loss": 0.8486,
      "step": 106
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.84782672979288e-05,
      "loss": 0.7597,
      "step": 107
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.844326759348443e-05,
      "loss": 0.8089,
      "step": 108
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.840787632001817e-05,
      "loss": 0.8206,
      "step": 109
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.837209376359917e-05,
      "loss": 0.8977,
      "step": 110
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.833592021345937e-05,
      "loss": 0.803,
      "step": 111
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.829935596199118e-05,
      "loss": 0.7868,
      "step": 112
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.826240130474498e-05,
      "loss": 0.8622,
      "step": 113
    },
    {
      "epoch": 0.51,
      "learning_rate": 9.822505654042686e-05,
      "loss": 0.8019,
      "step": 114
    },
    {
      "epoch": 0.51,
      "learning_rate": 9.818732197089619e-05,
      "loss": 0.8117,
      "step": 115
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.814919790116312e-05,
      "loss": 0.826,
      "step": 116
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.811068463938613e-05,
      "loss": 0.7429,
      "step": 117
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.807178249686958e-05,
      "loss": 0.8234,
      "step": 118
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.803249178806117e-05,
      "loss": 0.6938,
      "step": 119
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.79928128305494e-05,
      "loss": 0.7565,
      "step": 120
    },
    {
      "epoch": 0.53,
      "eval_loss": 0.7824007272720337,
      "eval_runtime": 16.1648,
      "eval_samples_per_second": 5.011,
      "eval_steps_per_second": 1.299,
      "step": 120
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.7952745945061e-05,
      "loss": 0.7787,
      "step": 121
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.791229145545831e-05,
      "loss": 0.7321,
      "step": 122
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.787144968873674e-05,
      "loss": 0.7939,
      "step": 123
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.783022097502204e-05,
      "loss": 0.7418,
      "step": 124
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.77886056475677e-05,
      "loss": 0.717,
      "step": 125
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.774660404275216e-05,
      "loss": 0.7836,
      "step": 126
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.770421650007626e-05,
      "loss": 0.8028,
      "step": 127
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.76614433621603e-05,
      "loss": 0.7333,
      "step": 128
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.761828497474141e-05,
      "loss": 0.7807,
      "step": 129
    },
    {
      "epoch": 0.58,
      "learning_rate": 9.757474168667071e-05,
      "loss": 0.7477,
      "step": 130
    },
    {
      "epoch": 0.58,
      "learning_rate": 9.753081384991045e-05,
      "loss": 0.7751,
      "step": 131
    },
    {
      "epoch": 0.59,
      "learning_rate": 9.748650181953126e-05,
      "loss": 0.7487,
      "step": 132
    },
    {
      "epoch": 0.59,
      "learning_rate": 9.744180595370917e-05,
      "loss": 0.7202,
      "step": 133
    },
    {
      "epoch": 0.6,
      "learning_rate": 9.739672661372279e-05,
      "loss": 0.7786,
      "step": 134
    },
    {
      "epoch": 0.6,
      "learning_rate": 9.73512641639504e-05,
      "loss": 0.7736,
      "step": 135
    },
    {
      "epoch": 0.6,
      "learning_rate": 9.730541897186689e-05,
      "loss": 0.7865,
      "step": 136
    },
    {
      "epoch": 0.61,
      "learning_rate": 9.725919140804099e-05,
      "loss": 0.8128,
      "step": 137
    },
    {
      "epoch": 0.61,
      "learning_rate": 9.721258184613203e-05,
      "loss": 0.8746,
      "step": 138
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.716559066288715e-05,
      "loss": 0.742,
      "step": 139
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.711821823813812e-05,
      "loss": 0.7634,
      "step": 140
    },
    {
      "epoch": 0.62,
      "eval_loss": 0.7769067883491516,
      "eval_runtime": 16.1735,
      "eval_samples_per_second": 5.008,
      "eval_steps_per_second": 1.298,
      "step": 140
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.707046495479826e-05,
      "loss": 0.8214,
      "step": 141
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.702233119885943e-05,
      "loss": 0.7602,
      "step": 142
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.697381735938887e-05,
      "loss": 0.8803,
      "step": 143
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.6924923828526e-05,
      "loss": 0.8203,
      "step": 144
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.687565100147939e-05,
      "loss": 0.7716,
      "step": 145
    },
    {
      "epoch": 0.65,
      "learning_rate": 9.682599927652341e-05,
      "loss": 0.7369,
      "step": 146
    },
    {
      "epoch": 0.65,
      "learning_rate": 9.677596905499507e-05,
      "loss": 0.8006,
      "step": 147
    },
    {
      "epoch": 0.66,
      "learning_rate": 9.672556074129085e-05,
      "loss": 0.7668,
      "step": 148
    },
    {
      "epoch": 0.66,
      "learning_rate": 9.667477474286328e-05,
      "loss": 0.7686,
      "step": 149
    },
    {
      "epoch": 0.67,
      "learning_rate": 9.662361147021779e-05,
      "loss": 0.7507,
      "step": 150
    },
    {
      "epoch": 0.67,
      "learning_rate": 9.657207133690936e-05,
      "loss": 0.7517,
      "step": 151
    },
    {
      "epoch": 0.68,
      "learning_rate": 9.652015475953904e-05,
      "loss": 0.818,
      "step": 152
    },
    {
      "epoch": 0.68,
      "learning_rate": 9.646786215775083e-05,
      "loss": 0.8582,
      "step": 153
    },
    {
      "epoch": 0.68,
      "learning_rate": 9.641519395422806e-05,
      "loss": 0.7712,
      "step": 154
    },
    {
      "epoch": 0.69,
      "learning_rate": 9.636215057469008e-05,
      "loss": 0.7736,
      "step": 155
    },
    {
      "epoch": 0.69,
      "learning_rate": 9.630873244788883e-05,
      "loss": 0.7466,
      "step": 156
    },
    {
      "epoch": 0.7,
      "learning_rate": 9.625494000560533e-05,
      "loss": 0.7985,
      "step": 157
    },
    {
      "epoch": 0.7,
      "learning_rate": 9.620077368264622e-05,
      "loss": 0.8343,
      "step": 158
    },
    {
      "epoch": 0.71,
      "learning_rate": 9.614623391684021e-05,
      "loss": 0.7508,
      "step": 159
    },
    {
      "epoch": 0.71,
      "learning_rate": 9.609132114903458e-05,
      "loss": 0.779,
      "step": 160
    },
    {
      "epoch": 0.71,
      "eval_loss": 0.7734922766685486,
      "eval_runtime": 16.2259,
      "eval_samples_per_second": 4.992,
      "eval_steps_per_second": 1.294,
      "step": 160
    },
    {
      "epoch": 0.72,
      "learning_rate": 9.603603582309162e-05,
      "loss": 0.8098,
      "step": 161
    },
    {
      "epoch": 0.72,
      "learning_rate": 9.598037838588499e-05,
      "loss": 0.8056,
      "step": 162
    },
    {
      "epoch": 0.72,
      "learning_rate": 9.592434928729616e-05,
      "loss": 0.7972,
      "step": 163
    },
    {
      "epoch": 0.73,
      "learning_rate": 9.586794898021075e-05,
      "loss": 0.7694,
      "step": 164
    },
    {
      "epoch": 0.73,
      "learning_rate": 9.581117792051486e-05,
      "loss": 0.7313,
      "step": 165
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.575403656709146e-05,
      "loss": 0.76,
      "step": 166
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.569652538181653e-05,
      "loss": 0.7332,
      "step": 167
    },
    {
      "epoch": 0.75,
      "learning_rate": 9.563864482955547e-05,
      "loss": 0.7197,
      "step": 168
    },
    {
      "epoch": 0.75,
      "learning_rate": 9.558039537815929e-05,
      "loss": 0.7504,
      "step": 169
    },
    {
      "epoch": 0.76,
      "learning_rate": 9.552177749846083e-05,
      "loss": 0.7278,
      "step": 170
    },
    {
      "epoch": 0.76,
      "learning_rate": 9.546279166427092e-05,
      "loss": 0.7565,
      "step": 171
    },
    {
      "epoch": 0.76,
      "learning_rate": 9.54034383523746e-05,
      "loss": 0.7512,
      "step": 172
    },
    {
      "epoch": 0.77,
      "learning_rate": 9.534371804252728e-05,
      "loss": 0.7861,
      "step": 173
    },
    {
      "epoch": 0.77,
      "learning_rate": 9.528363121745076e-05,
      "loss": 0.7935,
      "step": 174
    },
    {
      "epoch": 0.78,
      "learning_rate": 9.522317836282948e-05,
      "loss": 0.7924,
      "step": 175
    },
    {
      "epoch": 0.78,
      "learning_rate": 9.516235996730645e-05,
      "loss": 0.7461,
      "step": 176
    },
    {
      "epoch": 0.79,
      "learning_rate": 9.510117652247938e-05,
      "loss": 0.7863,
      "step": 177
    },
    {
      "epoch": 0.79,
      "learning_rate": 9.503962852289672e-05,
      "loss": 0.8284,
      "step": 178
    },
    {
      "epoch": 0.8,
      "learning_rate": 9.497771646605355e-05,
      "loss": 0.7805,
      "step": 179
    },
    {
      "epoch": 0.8,
      "learning_rate": 9.491544085238777e-05,
      "loss": 0.7776,
      "step": 180
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.7671141028404236,
      "eval_runtime": 16.2323,
      "eval_samples_per_second": 4.99,
      "eval_steps_per_second": 1.294,
      "step": 180
    },
    {
      "epoch": 0.8,
      "learning_rate": 9.485280218527581e-05,
      "loss": 0.8404,
      "step": 181
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.478980097102872e-05,
      "loss": 0.7056,
      "step": 182
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.472643771888803e-05,
      "loss": 0.8308,
      "step": 183
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.466271294102167e-05,
      "loss": 0.7528,
      "step": 184
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.459862715251973e-05,
      "loss": 0.8127,
      "step": 185
    },
    {
      "epoch": 0.83,
      "learning_rate": 9.453418087139043e-05,
      "loss": 0.7245,
      "step": 186
    },
    {
      "epoch": 0.83,
      "learning_rate": 9.446937461855583e-05,
      "loss": 0.7629,
      "step": 187
    },
    {
      "epoch": 0.84,
      "learning_rate": 9.440420891784766e-05,
      "loss": 0.7152,
      "step": 188
    },
    {
      "epoch": 0.84,
      "learning_rate": 9.43386842960031e-05,
      "loss": 0.7143,
      "step": 189
    },
    {
      "epoch": 0.84,
      "learning_rate": 9.42728012826605e-05,
      "loss": 0.7566,
      "step": 190
    },
    {
      "epoch": 0.85,
      "learning_rate": 9.42065604103551e-05,
      "loss": 0.7708,
      "step": 191
    },
    {
      "epoch": 0.85,
      "learning_rate": 9.41399622145147e-05,
      "loss": 0.7418,
      "step": 192
    },
    {
      "epoch": 0.86,
      "learning_rate": 9.407300723345543e-05,
      "loss": 0.7811,
      "step": 193
    },
    {
      "epoch": 0.86,
      "learning_rate": 9.400569600837728e-05,
      "loss": 0.7059,
      "step": 194
    },
    {
      "epoch": 0.87,
      "learning_rate": 9.393802908335977e-05,
      "loss": 0.751,
      "step": 195
    },
    {
      "epoch": 0.87,
      "learning_rate": 9.387000700535758e-05,
      "loss": 0.7053,
      "step": 196
    },
    {
      "epoch": 0.88,
      "learning_rate": 9.380163032419611e-05,
      "loss": 0.7926,
      "step": 197
    },
    {
      "epoch": 0.88,
      "learning_rate": 9.373289959256701e-05,
      "loss": 0.8703,
      "step": 198
    },
    {
      "epoch": 0.88,
      "learning_rate": 9.366381536602378e-05,
      "loss": 0.7272,
      "step": 199
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.359437820297717e-05,
      "loss": 0.7694,
      "step": 200
    },
    {
      "epoch": 0.89,
      "eval_loss": 0.7667359709739685,
      "eval_runtime": 16.1834,
      "eval_samples_per_second": 5.005,
      "eval_steps_per_second": 1.298,
      "step": 200
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.352458866469076e-05,
      "loss": 0.8422,
      "step": 201
    },
    {
      "epoch": 0.9,
      "learning_rate": 9.345444731527642e-05,
      "loss": 0.8142,
      "step": 202
    },
    {
      "epoch": 0.9,
      "learning_rate": 9.338395472168971e-05,
      "loss": 0.7815,
      "step": 203
    },
    {
      "epoch": 0.91,
      "learning_rate": 9.331311145372528e-05,
      "loss": 0.8582,
      "step": 204
    },
    {
      "epoch": 0.91,
      "learning_rate": 9.324191808401235e-05,
      "loss": 0.7516,
      "step": 205
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.317037518800998e-05,
      "loss": 0.7781,
      "step": 206
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.309848334400246e-05,
      "loss": 0.7884,
      "step": 207
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.302624313309471e-05,
      "loss": 0.7071,
      "step": 208
    },
    {
      "epoch": 0.93,
      "learning_rate": 9.295365513920748e-05,
      "loss": 0.6806,
      "step": 209
    },
    {
      "epoch": 0.93,
      "learning_rate": 9.288071994907261e-05,
      "loss": 0.7055,
      "step": 210
    },
    {
      "epoch": 0.94,
      "learning_rate": 9.280743815222841e-05,
      "loss": 0.7375,
      "step": 211
    },
    {
      "epoch": 0.94,
      "learning_rate": 9.273381034101482e-05,
      "loss": 0.73,
      "step": 212
    },
    {
      "epoch": 0.95,
      "learning_rate": 9.26598371105686e-05,
      "loss": 0.7278,
      "step": 213
    },
    {
      "epoch": 0.95,
      "learning_rate": 9.258551905881856e-05,
      "loss": 0.7403,
      "step": 214
    },
    {
      "epoch": 0.96,
      "learning_rate": 9.251085678648072e-05,
      "loss": 0.7537,
      "step": 215
    },
    {
      "epoch": 0.96,
      "learning_rate": 9.243585089705344e-05,
      "loss": 0.803,
      "step": 216
    },
    {
      "epoch": 0.96,
      "learning_rate": 9.236050199681258e-05,
      "loss": 0.7842,
      "step": 217
    },
    {
      "epoch": 0.97,
      "learning_rate": 9.228481069480655e-05,
      "loss": 0.6973,
      "step": 218
    },
    {
      "epoch": 0.97,
      "learning_rate": 9.220877760285141e-05,
      "loss": 0.7908,
      "step": 219
    },
    {
      "epoch": 0.98,
      "learning_rate": 9.213240333552589e-05,
      "loss": 0.7201,
      "step": 220
    },
    {
      "epoch": 0.98,
      "eval_loss": 0.761328935623169,
      "eval_runtime": 16.1777,
      "eval_samples_per_second": 5.007,
      "eval_steps_per_second": 1.298,
      "step": 220
    },
    {
      "epoch": 0.98,
      "learning_rate": 9.205568851016652e-05,
      "loss": 0.7408,
      "step": 221
    },
    {
      "epoch": 0.99,
      "learning_rate": 9.197863374686256e-05,
      "loss": 0.8622,
      "step": 222
    },
    {
      "epoch": 0.99,
      "learning_rate": 9.190123966845092e-05,
      "loss": 0.8189,
      "step": 223
    },
    {
      "epoch": 1.0,
      "learning_rate": 9.182350690051133e-05,
      "loss": 0.7666,
      "step": 224
    },
    {
      "epoch": 1.0,
      "learning_rate": 9.174543607136112e-05,
      "loss": 0.8323,
      "step": 225
    },
    {
      "epoch": 1.0,
      "learning_rate": 9.166702781205012e-05,
      "loss": 0.8211,
      "step": 226
    },
    {
      "epoch": 1.01,
      "learning_rate": 9.158828275635569e-05,
      "loss": 0.7923,
      "step": 227
    },
    {
      "epoch": 1.01,
      "learning_rate": 9.150920154077754e-05,
      "loss": 0.7626,
      "step": 228
    },
    {
      "epoch": 1.02,
      "learning_rate": 9.142978480453251e-05,
      "loss": 0.7593,
      "step": 229
    },
    {
      "epoch": 1.02,
      "learning_rate": 9.135003318954954e-05,
      "loss": 0.7953,
      "step": 230
    },
    {
      "epoch": 1.03,
      "learning_rate": 9.126994734046432e-05,
      "loss": 0.7352,
      "step": 231
    },
    {
      "epoch": 1.03,
      "learning_rate": 9.11895279046143e-05,
      "loss": 0.7583,
      "step": 232
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.110877553203319e-05,
      "loss": 0.7573,
      "step": 233
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.102769087544592e-05,
      "loss": 0.7988,
      "step": 234
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.094627459026325e-05,
      "loss": 0.7627,
      "step": 235
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.086452733457656e-05,
      "loss": 0.8511,
      "step": 236
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.078244976915244e-05,
      "loss": 0.7316,
      "step": 237
    },
    {
      "epoch": 1.06,
      "learning_rate": 9.070004255742737e-05,
      "loss": 0.7108,
      "step": 238
    },
    {
      "epoch": 1.06,
      "learning_rate": 9.06173063655024e-05,
      "loss": 0.797,
      "step": 239
    },
    {
      "epoch": 1.07,
      "learning_rate": 9.053424186213775e-05,
      "loss": 0.7498,
      "step": 240
    },
    {
      "epoch": 1.07,
      "eval_loss": 0.758377194404602,
      "eval_runtime": 16.1613,
      "eval_samples_per_second": 5.012,
      "eval_steps_per_second": 1.299,
      "step": 240
    },
    {
      "epoch": 1.07,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.6338,
      "step": 241
    },
    {
      "epoch": 1.08,
      "learning_rate": 9.036713060939358e-05,
      "loss": 0.8083,
      "step": 242
    },
    {
      "epoch": 1.08,
      "learning_rate": 9.028308521078152e-05,
      "loss": 0.7359,
      "step": 243
    },
    {
      "epoch": 1.08,
      "learning_rate": 9.019871420225381e-05,
      "loss": 0.6618,
      "step": 244
    },
    {
      "epoch": 1.09,
      "learning_rate": 9.011401826578492e-05,
      "loss": 0.6512,
      "step": 245
    },
    {
      "epoch": 1.09,
      "learning_rate": 9.002899808597576e-05,
      "loss": 0.6483,
      "step": 246
    },
    {
      "epoch": 1.1,
      "learning_rate": 8.994365435004815e-05,
      "loss": 0.727,
      "step": 247
    },
    {
      "epoch": 1.1,
      "learning_rate": 8.985798774783913e-05,
      "loss": 0.7561,
      "step": 248
    },
    {
      "epoch": 1.11,
      "learning_rate": 8.977199897179558e-05,
      "loss": 0.7811,
      "step": 249
    },
    {
      "epoch": 1.11,
      "learning_rate": 8.968568871696847e-05,
      "loss": 0.6946,
      "step": 250
    },
    {
      "epoch": 1.12,
      "learning_rate": 8.959905768100733e-05,
      "loss": 0.713,
      "step": 251
    },
    {
      "epoch": 1.12,
      "learning_rate": 8.951210656415456e-05,
      "loss": 0.6828,
      "step": 252
    },
    {
      "epoch": 1.12,
      "learning_rate": 8.942483606923979e-05,
      "loss": 0.6932,
      "step": 253
    },
    {
      "epoch": 1.13,
      "learning_rate": 8.933724690167417e-05,
      "loss": 0.7629,
      "step": 254
    },
    {
      "epoch": 1.13,
      "learning_rate": 8.924933976944474e-05,
      "loss": 0.6765,
      "step": 255
    },
    {
      "epoch": 1.14,
      "learning_rate": 8.916111538310863e-05,
      "loss": 0.7267,
      "step": 256
    },
    {
      "epoch": 1.14,
      "learning_rate": 8.907257445578739e-05,
      "loss": 0.729,
      "step": 257
    },
    {
      "epoch": 1.15,
      "learning_rate": 8.898371770316111e-05,
      "loss": 0.6951,
      "step": 258
    },
    {
      "epoch": 1.15,
      "learning_rate": 8.889454584346282e-05,
      "loss": 0.8412,
      "step": 259
    },
    {
      "epoch": 1.16,
      "learning_rate": 8.880505959747244e-05,
      "loss": 0.7852,
      "step": 260
    },
    {
      "epoch": 1.16,
      "eval_loss": 0.7558982372283936,
      "eval_runtime": 16.1507,
      "eval_samples_per_second": 5.015,
      "eval_steps_per_second": 1.3,
      "step": 260
    },
    {
      "epoch": 1.16,
      "learning_rate": 8.87152596885112e-05,
      "loss": 0.6959,
      "step": 261
    },
    {
      "epoch": 1.16,
      "learning_rate": 8.862514684243562e-05,
      "loss": 0.7405,
      "step": 262
    },
    {
      "epoch": 1.17,
      "learning_rate": 8.853472178763171e-05,
      "loss": 0.6729,
      "step": 263
    },
    {
      "epoch": 1.17,
      "learning_rate": 8.844398525500914e-05,
      "loss": 0.752,
      "step": 264
    },
    {
      "epoch": 1.18,
      "learning_rate": 8.835293797799517e-05,
      "loss": 0.7236,
      "step": 265
    },
    {
      "epoch": 1.18,
      "learning_rate": 8.826158069252888e-05,
      "loss": 0.6427,
      "step": 266
    },
    {
      "epoch": 1.19,
      "learning_rate": 8.816991413705516e-05,
      "loss": 0.7169,
      "step": 267
    },
    {
      "epoch": 1.19,
      "learning_rate": 8.80779390525187e-05,
      "loss": 0.7441,
      "step": 268
    },
    {
      "epoch": 1.2,
      "learning_rate": 8.798565618235813e-05,
      "loss": 0.7225,
      "step": 269
    },
    {
      "epoch": 1.2,
      "learning_rate": 8.789306627249985e-05,
      "loss": 0.7396,
      "step": 270
    },
    {
      "epoch": 1.2,
      "learning_rate": 8.780017007135207e-05,
      "loss": 0.7642,
      "step": 271
    },
    {
      "epoch": 1.21,
      "learning_rate": 8.770696832979881e-05,
      "loss": 0.7186,
      "step": 272
    },
    {
      "epoch": 1.21,
      "learning_rate": 8.761346180119376e-05,
      "loss": 0.6484,
      "step": 273
    },
    {
      "epoch": 1.22,
      "learning_rate": 8.751965124135425e-05,
      "loss": 0.731,
      "step": 274
    },
    {
      "epoch": 1.22,
      "learning_rate": 8.742553740855506e-05,
      "loss": 0.7666,
      "step": 275
    },
    {
      "epoch": 1.23,
      "learning_rate": 8.733112106352237e-05,
      "loss": 0.8041,
      "step": 276
    },
    {
      "epoch": 1.23,
      "learning_rate": 8.723640296942755e-05,
      "loss": 0.7403,
      "step": 277
    },
    {
      "epoch": 1.24,
      "learning_rate": 8.714138389188106e-05,
      "loss": 0.689,
      "step": 278
    },
    {
      "epoch": 1.24,
      "learning_rate": 8.704606459892623e-05,
      "loss": 0.7662,
      "step": 279
    },
    {
      "epoch": 1.24,
      "learning_rate": 8.695044586103296e-05,
      "loss": 0.8143,
      "step": 280
    },
    {
      "epoch": 1.24,
      "eval_loss": 0.7563924193382263,
      "eval_runtime": 16.1537,
      "eval_samples_per_second": 5.014,
      "eval_steps_per_second": 1.3,
      "step": 280
    },
    {
      "epoch": 1.25,
      "learning_rate": 8.685452845109168e-05,
      "loss": 0.7726,
      "step": 281
    },
    {
      "epoch": 1.25,
      "learning_rate": 8.675831314440693e-05,
      "loss": 0.7412,
      "step": 282
    },
    {
      "epoch": 1.26,
      "learning_rate": 8.666180071869117e-05,
      "loss": 0.7479,
      "step": 283
    },
    {
      "epoch": 1.26,
      "learning_rate": 8.656499195405852e-05,
      "loss": 0.8669,
      "step": 284
    },
    {
      "epoch": 1.27,
      "learning_rate": 8.646788763301841e-05,
      "loss": 0.8918,
      "step": 285
    },
    {
      "epoch": 1.27,
      "learning_rate": 8.637048854046926e-05,
      "loss": 0.8616,
      "step": 286
    },
    {
      "epoch": 1.28,
      "learning_rate": 8.62727954636921e-05,
      "loss": 0.7365,
      "step": 287
    },
    {
      "epoch": 1.28,
      "learning_rate": 8.617480919234433e-05,
      "loss": 0.672,
      "step": 288
    },
    {
      "epoch": 1.28,
      "learning_rate": 8.607653051845316e-05,
      "loss": 0.7442,
      "step": 289
    },
    {
      "epoch": 1.29,
      "learning_rate": 8.597796023640941e-05,
      "loss": 0.7626,
      "step": 290
    },
    {
      "epoch": 1.29,
      "learning_rate": 8.587909914296089e-05,
      "loss": 0.731,
      "step": 291
    },
    {
      "epoch": 1.3,
      "learning_rate": 8.577994803720606e-05,
      "loss": 0.7305,
      "step": 292
    },
    {
      "epoch": 1.3,
      "learning_rate": 8.568050772058762e-05,
      "loss": 0.7216,
      "step": 293
    },
    {
      "epoch": 1.31,
      "learning_rate": 8.558077899688592e-05,
      "loss": 0.7631,
      "step": 294
    },
    {
      "epoch": 1.31,
      "learning_rate": 8.548076267221256e-05,
      "loss": 0.6488,
      "step": 295
    },
    {
      "epoch": 1.32,
      "learning_rate": 8.538045955500379e-05,
      "loss": 0.7594,
      "step": 296
    },
    {
      "epoch": 1.32,
      "learning_rate": 8.527987045601404e-05,
      "loss": 0.6487,
      "step": 297
    },
    {
      "epoch": 1.32,
      "learning_rate": 8.517899618830931e-05,
      "loss": 0.7834,
      "step": 298
    },
    {
      "epoch": 1.33,
      "learning_rate": 8.50778375672607e-05,
      "loss": 0.734,
      "step": 299
    },
    {
      "epoch": 1.33,
      "learning_rate": 8.49763954105377e-05,
      "loss": 0.8413,
      "step": 300
    },
    {
      "epoch": 1.33,
      "eval_loss": 0.7527250647544861,
      "eval_runtime": 16.1668,
      "eval_samples_per_second": 5.01,
      "eval_steps_per_second": 1.299,
      "step": 300
    },
    {
      "epoch": 1.34,
      "learning_rate": 8.487467053810161e-05,
      "loss": 0.765,
      "step": 301
    },
    {
      "epoch": 1.34,
      "learning_rate": 8.477266377219898e-05,
      "loss": 0.7672,
      "step": 302
    },
    {
      "epoch": 1.35,
      "learning_rate": 8.46703759373549e-05,
      "loss": 0.7322,
      "step": 303
    },
    {
      "epoch": 1.35,
      "learning_rate": 8.456780786036635e-05,
      "loss": 0.771,
      "step": 304
    },
    {
      "epoch": 1.36,
      "learning_rate": 8.446496037029555e-05,
      "loss": 0.7365,
      "step": 305
    },
    {
      "epoch": 1.36,
      "learning_rate": 8.436183429846313e-05,
      "loss": 0.73,
      "step": 306
    },
    {
      "epoch": 1.36,
      "learning_rate": 8.425843047844165e-05,
      "loss": 0.7694,
      "step": 307
    },
    {
      "epoch": 1.37,
      "learning_rate": 8.415474974604862e-05,
      "loss": 0.6973,
      "step": 308
    },
    {
      "epoch": 1.37,
      "learning_rate": 8.405079293933986e-05,
      "loss": 0.6781,
      "step": 309
    },
    {
      "epoch": 1.38,
      "learning_rate": 8.394656089860274e-05,
      "loss": 0.7672,
      "step": 310
    },
    {
      "epoch": 1.38,
      "learning_rate": 8.384205446634935e-05,
      "loss": 0.7333,
      "step": 311
    },
    {
      "epoch": 1.39,
      "learning_rate": 8.373727448730966e-05,
      "loss": 0.6644,
      "step": 312
    },
    {
      "epoch": 1.39,
      "learning_rate": 8.363222180842478e-05,
      "loss": 0.7538,
      "step": 313
    },
    {
      "epoch": 1.4,
      "learning_rate": 8.352689727884005e-05,
      "loss": 0.6857,
      "step": 314
    },
    {
      "epoch": 1.4,
      "learning_rate": 8.342130174989818e-05,
      "loss": 0.7679,
      "step": 315
    },
    {
      "epoch": 1.4,
      "learning_rate": 8.33154360751324e-05,
      "loss": 0.6924,
      "step": 316
    },
    {
      "epoch": 1.41,
      "learning_rate": 8.320930111025951e-05,
      "loss": 0.7447,
      "step": 317
    },
    {
      "epoch": 1.41,
      "learning_rate": 8.3102897713173e-05,
      "loss": 0.7555,
      "step": 318
    },
    {
      "epoch": 1.42,
      "learning_rate": 8.299622674393614e-05,
      "loss": 0.6906,
      "step": 319
    },
    {
      "epoch": 1.42,
      "learning_rate": 8.288928906477496e-05,
      "loss": 0.7673,
      "step": 320
    },
    {
      "epoch": 1.42,
      "eval_loss": 0.7525838017463684,
      "eval_runtime": 16.2133,
      "eval_samples_per_second": 4.996,
      "eval_steps_per_second": 1.295,
      "step": 320
    },
    {
      "epoch": 1.43,
      "learning_rate": 8.278208554007136e-05,
      "loss": 0.6646,
      "step": 321
    },
    {
      "epoch": 1.43,
      "learning_rate": 8.267461703635604e-05,
      "loss": 0.7592,
      "step": 322
    },
    {
      "epoch": 1.44,
      "learning_rate": 8.256688442230154e-05,
      "loss": 0.6533,
      "step": 323
    },
    {
      "epoch": 1.44,
      "learning_rate": 8.245888856871525e-05,
      "loss": 0.7016,
      "step": 324
    },
    {
      "epoch": 1.44,
      "learning_rate": 8.235063034853228e-05,
      "loss": 0.694,
      "step": 325
    },
    {
      "epoch": 1.45,
      "learning_rate": 8.224211063680853e-05,
      "loss": 0.7522,
      "step": 326
    },
    {
      "epoch": 1.45,
      "learning_rate": 8.213333031071351e-05,
      "loss": 0.7599,
      "step": 327
    },
    {
      "epoch": 1.46,
      "learning_rate": 8.202429024952327e-05,
      "loss": 0.7784,
      "step": 328
    },
    {
      "epoch": 1.46,
      "learning_rate": 8.191499133461332e-05,
      "loss": 0.7284,
      "step": 329
    },
    {
      "epoch": 1.47,
      "learning_rate": 8.180543444945153e-05,
      "loss": 0.7748,
      "step": 330
    },
    {
      "epoch": 1.47,
      "learning_rate": 8.169562047959093e-05,
      "loss": 0.7928,
      "step": 331
    },
    {
      "epoch": 1.48,
      "learning_rate": 8.158555031266254e-05,
      "loss": 0.7354,
      "step": 332
    },
    {
      "epoch": 1.48,
      "learning_rate": 8.147522483836833e-05,
      "loss": 0.7134,
      "step": 333
    },
    {
      "epoch": 1.48,
      "learning_rate": 8.13646449484738e-05,
      "loss": 0.6899,
      "step": 334
    },
    {
      "epoch": 1.49,
      "learning_rate": 8.125381153680103e-05,
      "loss": 0.7251,
      "step": 335
    },
    {
      "epoch": 1.49,
      "learning_rate": 8.114272549922122e-05,
      "loss": 0.75,
      "step": 336
    },
    {
      "epoch": 1.5,
      "learning_rate": 8.103138773364763e-05,
      "loss": 0.7581,
      "step": 337
    },
    {
      "epoch": 1.5,
      "learning_rate": 8.091979914002823e-05,
      "loss": 0.7181,
      "step": 338
    },
    {
      "epoch": 1.51,
      "learning_rate": 8.08079606203384e-05,
      "loss": 0.7017,
      "step": 339
    },
    {
      "epoch": 1.51,
      "learning_rate": 8.069587307857375e-05,
      "loss": 0.7212,
      "step": 340
    },
    {
      "epoch": 1.51,
      "eval_loss": 0.7467713356018066,
      "eval_runtime": 16.2203,
      "eval_samples_per_second": 4.994,
      "eval_steps_per_second": 1.295,
      "step": 340
    },
    {
      "epoch": 1.52,
      "learning_rate": 8.058353742074274e-05,
      "loss": 0.7242,
      "step": 341
    },
    {
      "epoch": 1.52,
      "learning_rate": 8.047095455485926e-05,
      "loss": 0.641,
      "step": 342
    },
    {
      "epoch": 1.52,
      "learning_rate": 8.035812539093557e-05,
      "loss": 0.6913,
      "step": 343
    },
    {
      "epoch": 1.53,
      "learning_rate": 8.024505084097461e-05,
      "loss": 0.6621,
      "step": 344
    },
    {
      "epoch": 1.53,
      "learning_rate": 8.013173181896283e-05,
      "loss": 0.6989,
      "step": 345
    },
    {
      "epoch": 1.54,
      "learning_rate": 8.001816924086281e-05,
      "loss": 0.7018,
      "step": 346
    },
    {
      "epoch": 1.54,
      "learning_rate": 7.990436402460575e-05,
      "loss": 0.6998,
      "step": 347
    },
    {
      "epoch": 1.55,
      "learning_rate": 7.979031709008415e-05,
      "loss": 0.6177,
      "step": 348
    },
    {
      "epoch": 1.55,
      "learning_rate": 7.967602935914427e-05,
      "loss": 0.7046,
      "step": 349
    },
    {
      "epoch": 1.56,
      "learning_rate": 7.95615017555788e-05,
      "loss": 0.7031,
      "step": 350
    },
    {
      "epoch": 1.56,
      "learning_rate": 7.944673520511925e-05,
      "loss": 0.8082,
      "step": 351
    },
    {
      "epoch": 1.56,
      "learning_rate": 7.933173063542865e-05,
      "loss": 0.6821,
      "step": 352
    },
    {
      "epoch": 1.57,
      "learning_rate": 7.921648897609388e-05,
      "loss": 0.7141,
      "step": 353
    },
    {
      "epoch": 1.57,
      "learning_rate": 7.910101115861826e-05,
      "loss": 0.7485,
      "step": 354
    },
    {
      "epoch": 1.58,
      "learning_rate": 7.898529811641394e-05,
      "loss": 0.6949,
      "step": 355
    },
    {
      "epoch": 1.58,
      "learning_rate": 7.886935078479445e-05,
      "loss": 0.6738,
      "step": 356
    },
    {
      "epoch": 1.59,
      "learning_rate": 7.875317010096705e-05,
      "loss": 0.7596,
      "step": 357
    },
    {
      "epoch": 1.59,
      "learning_rate": 7.863675700402526e-05,
      "loss": 0.7138,
      "step": 358
    },
    {
      "epoch": 1.6,
      "learning_rate": 7.852011243494115e-05,
      "loss": 0.7938,
      "step": 359
    },
    {
      "epoch": 1.6,
      "learning_rate": 7.840323733655778e-05,
      "loss": 0.7411,
      "step": 360
    },
    {
      "epoch": 1.6,
      "eval_loss": 0.7443473935127258,
      "eval_runtime": 16.2109,
      "eval_samples_per_second": 4.997,
      "eval_steps_per_second": 1.295,
      "step": 360
    },
    {
      "epoch": 1.6,
      "learning_rate": 7.828613265358167e-05,
      "loss": 0.7394,
      "step": 361
    },
    {
      "epoch": 1.61,
      "learning_rate": 7.816879933257494e-05,
      "loss": 0.7386,
      "step": 362
    },
    {
      "epoch": 1.61,
      "learning_rate": 7.805123832194797e-05,
      "loss": 0.6989,
      "step": 363
    },
    {
      "epoch": 1.62,
      "learning_rate": 7.793345057195143e-05,
      "loss": 0.6786,
      "step": 364
    },
    {
      "epoch": 1.62,
      "learning_rate": 7.78154370346688e-05,
      "loss": 0.6932,
      "step": 365
    },
    {
      "epoch": 1.63,
      "learning_rate": 7.76971986640086e-05,
      "loss": 0.7703,
      "step": 366
    },
    {
      "epoch": 1.63,
      "learning_rate": 7.757873641569666e-05,
      "loss": 0.7093,
      "step": 367
    },
    {
      "epoch": 1.64,
      "learning_rate": 7.746005124726847e-05,
      "loss": 0.7387,
      "step": 368
    },
    {
      "epoch": 1.64,
      "learning_rate": 7.734114411806133e-05,
      "loss": 0.7532,
      "step": 369
    },
    {
      "epoch": 1.64,
      "learning_rate": 7.722201598920673e-05,
      "loss": 0.6431,
      "step": 370
    },
    {
      "epoch": 1.65,
      "learning_rate": 7.710266782362247e-05,
      "loss": 0.6197,
      "step": 371
    },
    {
      "epoch": 1.65,
      "learning_rate": 7.698310058600492e-05,
      "loss": 0.7008,
      "step": 372
    },
    {
      "epoch": 1.66,
      "learning_rate": 7.68633152428212e-05,
      "loss": 0.7036,
      "step": 373
    },
    {
      "epoch": 1.66,
      "learning_rate": 7.674331276230143e-05,
      "loss": 0.6732,
      "step": 374
    },
    {
      "epoch": 1.67,
      "learning_rate": 7.662309411443082e-05,
      "loss": 0.7795,
      "step": 375
    },
    {
      "epoch": 1.67,
      "learning_rate": 7.65026602709419e-05,
      "loss": 0.807,
      "step": 376
    },
    {
      "epoch": 1.68,
      "learning_rate": 7.638201220530665e-05,
      "loss": 0.7432,
      "step": 377
    },
    {
      "epoch": 1.68,
      "learning_rate": 7.626115089272852e-05,
      "loss": 0.7394,
      "step": 378
    },
    {
      "epoch": 1.68,
      "learning_rate": 7.614007731013477e-05,
      "loss": 0.7725,
      "step": 379
    },
    {
      "epoch": 1.69,
      "learning_rate": 7.601879243616839e-05,
      "loss": 0.7654,
      "step": 380
    },
    {
      "epoch": 1.69,
      "eval_loss": 0.7455257177352905,
      "eval_runtime": 16.2214,
      "eval_samples_per_second": 4.993,
      "eval_steps_per_second": 1.295,
      "step": 380
    },
    {
      "epoch": 1.69,
      "learning_rate": 7.58972972511802e-05,
      "loss": 0.7174,
      "step": 381
    },
    {
      "epoch": 1.7,
      "learning_rate": 7.577559273722107e-05,
      "loss": 0.7298,
      "step": 382
    },
    {
      "epoch": 1.7,
      "learning_rate": 7.565367987803382e-05,
      "loss": 0.8,
      "step": 383
    },
    {
      "epoch": 1.71,
      "learning_rate": 7.553155965904535e-05,
      "loss": 0.7282,
      "step": 384
    },
    {
      "epoch": 1.71,
      "learning_rate": 7.540923306735868e-05,
      "loss": 0.7088,
      "step": 385
    },
    {
      "epoch": 1.72,
      "learning_rate": 7.528670109174489e-05,
      "loss": 0.7395,
      "step": 386
    },
    {
      "epoch": 1.72,
      "learning_rate": 7.516396472263524e-05,
      "loss": 0.7584,
      "step": 387
    },
    {
      "epoch": 1.72,
      "learning_rate": 7.504102495211312e-05,
      "loss": 0.7816,
      "step": 388
    },
    {
      "epoch": 1.73,
      "learning_rate": 7.491788277390595e-05,
      "loss": 0.7837,
      "step": 389
    },
    {
      "epoch": 1.73,
      "learning_rate": 7.479453918337734e-05,
      "loss": 0.7132,
      "step": 390
    },
    {
      "epoch": 1.74,
      "learning_rate": 7.467099517751879e-05,
      "loss": 0.7289,
      "step": 391
    },
    {
      "epoch": 1.74,
      "learning_rate": 7.454725175494183e-05,
      "loss": 0.7453,
      "step": 392
    },
    {
      "epoch": 1.75,
      "learning_rate": 7.442330991586995e-05,
      "loss": 0.7424,
      "step": 393
    },
    {
      "epoch": 1.75,
      "learning_rate": 7.42991706621303e-05,
      "loss": 0.7824,
      "step": 394
    },
    {
      "epoch": 1.76,
      "learning_rate": 7.417483499714589e-05,
      "loss": 0.6926,
      "step": 395
    },
    {
      "epoch": 1.76,
      "learning_rate": 7.405030392592723e-05,
      "loss": 0.72,
      "step": 396
    },
    {
      "epoch": 1.76,
      "learning_rate": 7.392557845506432e-05,
      "loss": 0.7812,
      "step": 397
    },
    {
      "epoch": 1.77,
      "learning_rate": 7.380065959271858e-05,
      "loss": 0.6893,
      "step": 398
    },
    {
      "epoch": 1.77,
      "learning_rate": 7.367554834861452e-05,
      "loss": 0.712,
      "step": 399
    },
    {
      "epoch": 1.78,
      "learning_rate": 7.355024573403174e-05,
      "loss": 0.7619,
      "step": 400
    },
    {
      "epoch": 1.78,
      "eval_loss": 0.7424753904342651,
      "eval_runtime": 16.2095,
      "eval_samples_per_second": 4.997,
      "eval_steps_per_second": 1.296,
      "step": 400
    },
    {
      "epoch": 1.78,
      "learning_rate": 7.342475276179668e-05,
      "loss": 0.7277,
      "step": 401
    },
    {
      "epoch": 1.79,
      "learning_rate": 7.329907044627444e-05,
      "loss": 0.6948,
      "step": 402
    },
    {
      "epoch": 1.79,
      "learning_rate": 7.31731998033606e-05,
      "loss": 0.7299,
      "step": 403
    },
    {
      "epoch": 1.8,
      "learning_rate": 7.3047141850473e-05,
      "loss": 0.7115,
      "step": 404
    },
    {
      "epoch": 1.8,
      "learning_rate": 7.292089760654351e-05,
      "loss": 0.7089,
      "step": 405
    },
    {
      "epoch": 1.8,
      "learning_rate": 7.279446809200981e-05,
      "loss": 0.6761,
      "step": 406
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.266785432880711e-05,
      "loss": 0.837,
      "step": 407
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.254105734035991e-05,
      "loss": 0.7154,
      "step": 408
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.241407815157376e-05,
      "loss": 0.7904,
      "step": 409
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.228691778882693e-05,
      "loss": 0.713,
      "step": 410
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.215957727996207e-05,
      "loss": 0.7723,
      "step": 411
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.203205765427806e-05,
      "loss": 0.6863,
      "step": 412
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.190435994252149e-05,
      "loss": 0.6906,
      "step": 413
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.177648517687852e-05,
      "loss": 0.6936,
      "step": 414
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.164843439096636e-05,
      "loss": 0.7206,
      "step": 415
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.152020861982505e-05,
      "loss": 0.7572,
      "step": 416
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.139180889990903e-05,
      "loss": 0.7206,
      "step": 417
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.126323626907878e-05,
      "loss": 0.7221,
      "step": 418
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.113449176659241e-05,
      "loss": 0.6963,
      "step": 419
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.100557643309732e-05,
      "loss": 0.7152,
      "step": 420
    },
    {
      "epoch": 1.87,
      "eval_loss": 0.7406202554702759,
      "eval_runtime": 16.1712,
      "eval_samples_per_second": 5.009,
      "eval_steps_per_second": 1.299,
      "step": 420
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.087649131062167e-05,
      "loss": 0.6858,
      "step": 421
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.074723744256609e-05,
      "loss": 0.886,
      "step": 422
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.061781587369519e-05,
      "loss": 0.7167,
      "step": 423
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.048822765012906e-05,
      "loss": 0.6678,
      "step": 424
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.035847381933493e-05,
      "loss": 0.7218,
      "step": 425
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.02285554301186e-05,
      "loss": 0.7909,
      "step": 426
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.009847353261601e-05,
      "loss": 0.7765,
      "step": 427
    },
    {
      "epoch": 1.9,
      "learning_rate": 6.996822917828477e-05,
      "loss": 0.7674,
      "step": 428
    },
    {
      "epoch": 1.91,
      "learning_rate": 6.983782341989562e-05,
      "loss": 0.7803,
      "step": 429
    },
    {
      "epoch": 1.91,
      "learning_rate": 6.970725731152388e-05,
      "loss": 0.7154,
      "step": 430
    },
    {
      "epoch": 1.92,
      "learning_rate": 6.95765319085411e-05,
      "loss": 0.7647,
      "step": 431
    },
    {
      "epoch": 1.92,
      "learning_rate": 6.944564826760631e-05,
      "loss": 0.7262,
      "step": 432
    },
    {
      "epoch": 1.92,
      "learning_rate": 6.931460744665763e-05,
      "loss": 0.6987,
      "step": 433
    },
    {
      "epoch": 1.93,
      "learning_rate": 6.918341050490369e-05,
      "loss": 0.6296,
      "step": 434
    },
    {
      "epoch": 1.93,
      "learning_rate": 6.905205850281502e-05,
      "loss": 0.6835,
      "step": 435
    },
    {
      "epoch": 1.94,
      "learning_rate": 6.892055250211552e-05,
      "loss": 0.7104,
      "step": 436
    },
    {
      "epoch": 1.94,
      "learning_rate": 6.878889356577386e-05,
      "loss": 0.7301,
      "step": 437
    },
    {
      "epoch": 1.95,
      "learning_rate": 6.865708275799492e-05,
      "loss": 0.7071,
      "step": 438
    },
    {
      "epoch": 1.95,
      "learning_rate": 6.852512114421112e-05,
      "loss": 0.6801,
      "step": 439
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.83930097910739e-05,
      "loss": 0.7442,
      "step": 440
    },
    {
      "epoch": 1.96,
      "eval_loss": 0.7402726411819458,
      "eval_runtime": 16.1717,
      "eval_samples_per_second": 5.009,
      "eval_steps_per_second": 1.299,
      "step": 440
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.826074976644502e-05,
      "loss": 0.6664,
      "step": 441
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.812834213938794e-05,
      "loss": 0.6879,
      "step": 442
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.799578798015926e-05,
      "loss": 0.704,
      "step": 443
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.786308836019997e-05,
      "loss": 0.6899,
      "step": 444
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.773024435212678e-05,
      "loss": 0.6909,
      "step": 445
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.759725702972358e-05,
      "loss": 0.6586,
      "step": 446
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.746412746793263e-05,
      "loss": 0.6694,
      "step": 447
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.733085674284589e-05,
      "loss": 0.7332,
      "step": 448
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.719744593169641e-05,
      "loss": 0.7097,
      "step": 449
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.706389611284954e-05,
      "loss": 0.7246,
      "step": 450
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.693020836579418e-05,
      "loss": 0.7681,
      "step": 451
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.679638377113419e-05,
      "loss": 0.6958,
      "step": 452
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.666242341057958e-05,
      "loss": 0.7621,
      "step": 453
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.652832836693764e-05,
      "loss": 0.695,
      "step": 454
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.639409972410446e-05,
      "loss": 0.8196,
      "step": 455
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.625973856705593e-05,
      "loss": 0.7137,
      "step": 456
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.612524598183907e-05,
      "loss": 0.7218,
      "step": 457
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.599062305556325e-05,
      "loss": 0.6928,
      "step": 458
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.58558708763914e-05,
      "loss": 0.7925,
      "step": 459
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.57209905335312e-05,
      "loss": 0.709,
      "step": 460
    },
    {
      "epoch": 2.04,
      "eval_loss": 0.7391338348388672,
      "eval_runtime": 16.1675,
      "eval_samples_per_second": 5.01,
      "eval_steps_per_second": 1.299,
      "step": 460
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.558598311722626e-05,
      "loss": 0.6628,
      "step": 461
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.7818,
      "step": 462
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.531559143038362e-05,
      "loss": 0.7831,
      "step": 463
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.518020934543359e-05,
      "loss": 0.688,
      "step": 464
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.504470455819652e-05,
      "loss": 0.7234,
      "step": 465
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.49090781639634e-05,
      "loss": 0.6163,
      "step": 466
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.477333125900831e-05,
      "loss": 0.6454,
      "step": 467
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.463746494057928e-05,
      "loss": 0.6303,
      "step": 468
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.450148030688962e-05,
      "loss": 0.7002,
      "step": 469
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.436537845710903e-05,
      "loss": 0.6609,
      "step": 470
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.422916049135462e-05,
      "loss": 0.6433,
      "step": 471
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.409282751068206e-05,
      "loss": 0.7334,
      "step": 472
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.395638061707674e-05,
      "loss": 0.7172,
      "step": 473
    },
    {
      "epoch": 2.11,
      "learning_rate": 6.381982091344477e-05,
      "loss": 0.6729,
      "step": 474
    },
    {
      "epoch": 2.11,
      "learning_rate": 6.368314950360415e-05,
      "loss": 0.6502,
      "step": 475
    },
    {
      "epoch": 2.12,
      "learning_rate": 6.354636749227576e-05,
      "loss": 0.7175,
      "step": 476
    },
    {
      "epoch": 2.12,
      "learning_rate": 6.340947598507449e-05,
      "loss": 0.6623,
      "step": 477
    },
    {
      "epoch": 2.12,
      "learning_rate": 6.327247608850034e-05,
      "loss": 0.6991,
      "step": 478
    },
    {
      "epoch": 2.13,
      "learning_rate": 6.313536890992935e-05,
      "loss": 0.6839,
      "step": 479
    },
    {
      "epoch": 2.13,
      "learning_rate": 6.299815555760477e-05,
      "loss": 0.6784,
      "step": 480
    },
    {
      "epoch": 2.13,
      "eval_loss": 0.7386890649795532,
      "eval_runtime": 16.1698,
      "eval_samples_per_second": 5.009,
      "eval_steps_per_second": 1.299,
      "step": 480
    },
    {
      "epoch": 2.14,
      "learning_rate": 6.286083714062804e-05,
      "loss": 0.6487,
      "step": 481
    },
    {
      "epoch": 2.14,
      "learning_rate": 6.272341476894985e-05,
      "loss": 0.7033,
      "step": 482
    },
    {
      "epoch": 2.15,
      "learning_rate": 6.258588955336117e-05,
      "loss": 0.7006,
      "step": 483
    },
    {
      "epoch": 2.15,
      "learning_rate": 6.244826260548426e-05,
      "loss": 0.7005,
      "step": 484
    },
    {
      "epoch": 2.16,
      "learning_rate": 6.231053503776362e-05,
      "loss": 0.6183,
      "step": 485
    },
    {
      "epoch": 2.16,
      "learning_rate": 6.217270796345721e-05,
      "loss": 0.7022,
      "step": 486
    },
    {
      "epoch": 2.16,
      "learning_rate": 6.203478249662714e-05,
      "loss": 0.6653,
      "step": 487
    },
    {
      "epoch": 2.17,
      "learning_rate": 6.189675975213094e-05,
      "loss": 0.6226,
      "step": 488
    },
    {
      "epoch": 2.17,
      "learning_rate": 6.175864084561242e-05,
      "loss": 0.6435,
      "step": 489
    },
    {
      "epoch": 2.18,
      "learning_rate": 6.162042689349264e-05,
      "loss": 0.6788,
      "step": 490
    },
    {
      "epoch": 2.18,
      "learning_rate": 6.148211901296095e-05,
      "loss": 0.669,
      "step": 491
    },
    {
      "epoch": 2.19,
      "learning_rate": 6.13437183219659e-05,
      "loss": 0.6735,
      "step": 492
    },
    {
      "epoch": 2.19,
      "learning_rate": 6.120522593920628e-05,
      "loss": 0.6594,
      "step": 493
    },
    {
      "epoch": 2.2,
      "learning_rate": 6.106664298412196e-05,
      "loss": 0.6089,
      "step": 494
    },
    {
      "epoch": 2.2,
      "learning_rate": 6.092797057688495e-05,
      "loss": 0.6393,
      "step": 495
    },
    {
      "epoch": 2.2,
      "learning_rate": 6.078920983839031e-05,
      "loss": 0.6794,
      "step": 496
    },
    {
      "epoch": 2.21,
      "learning_rate": 6.0650361890247075e-05,
      "loss": 0.6395,
      "step": 497
    },
    {
      "epoch": 2.21,
      "learning_rate": 6.051142785476921e-05,
      "loss": 0.6425,
      "step": 498
    },
    {
      "epoch": 2.22,
      "learning_rate": 6.037240885496649e-05,
      "loss": 0.6211,
      "step": 499
    },
    {
      "epoch": 2.22,
      "learning_rate": 6.023330601453551e-05,
      "loss": 0.7064,
      "step": 500
    },
    {
      "epoch": 2.22,
      "eval_loss": 0.7412368655204773,
      "eval_runtime": 16.197,
      "eval_samples_per_second": 5.001,
      "eval_steps_per_second": 1.297,
      "step": 500
    },
    {
      "epoch": 2.23,
      "learning_rate": 6.009412045785051e-05,
      "loss": 0.7181,
      "step": 501
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.995485330995438e-05,
      "loss": 0.6969,
      "step": 502
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.981550569654947e-05,
      "loss": 0.7208,
      "step": 503
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.967607874398854e-05,
      "loss": 0.747,
      "step": 504
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.953657357926569e-05,
      "loss": 0.774,
      "step": 505
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.939699133000714e-05,
      "loss": 0.7115,
      "step": 506
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.925733312446228e-05,
      "loss": 0.7553,
      "step": 507
    },
    {
      "epoch": 2.26,
      "learning_rate": 5.911760009149438e-05,
      "loss": 0.6754,
      "step": 508
    },
    {
      "epoch": 2.26,
      "learning_rate": 5.8977793360571596e-05,
      "loss": 0.6937,
      "step": 509
    },
    {
      "epoch": 2.27,
      "learning_rate": 5.883791406175775e-05,
      "loss": 0.7532,
      "step": 510
    },
    {
      "epoch": 2.27,
      "learning_rate": 5.8697963325703255e-05,
      "loss": 0.7376,
      "step": 511
    },
    {
      "epoch": 2.28,
      "learning_rate": 5.8557942283635934e-05,
      "loss": 0.6842,
      "step": 512
    },
    {
      "epoch": 2.28,
      "learning_rate": 5.841785206735192e-05,
      "loss": 0.6874,
      "step": 513
    },
    {
      "epoch": 2.28,
      "learning_rate": 5.82776938092065e-05,
      "loss": 0.7199,
      "step": 514
    },
    {
      "epoch": 2.29,
      "learning_rate": 5.813746864210489e-05,
      "loss": 0.7406,
      "step": 515
    },
    {
      "epoch": 2.29,
      "learning_rate": 5.7997177699493175e-05,
      "loss": 0.7413,
      "step": 516
    },
    {
      "epoch": 2.3,
      "learning_rate": 5.785682211534911e-05,
      "loss": 0.6762,
      "step": 517
    },
    {
      "epoch": 2.3,
      "learning_rate": 5.771640302417291e-05,
      "loss": 0.6858,
      "step": 518
    },
    {
      "epoch": 2.31,
      "learning_rate": 5.7575921560978154e-05,
      "loss": 0.6122,
      "step": 519
    },
    {
      "epoch": 2.31,
      "learning_rate": 5.7435378861282585e-05,
      "loss": 0.718,
      "step": 520
    },
    {
      "epoch": 2.31,
      "eval_loss": 0.7355238795280457,
      "eval_runtime": 16.2114,
      "eval_samples_per_second": 4.996,
      "eval_steps_per_second": 1.295,
      "step": 520
    },
    {
      "epoch": 2.32,
      "learning_rate": 5.72947760610989e-05,
      "loss": 0.736,
      "step": 521
    },
    {
      "epoch": 2.32,
      "learning_rate": 5.71541142969256e-05,
      "loss": 0.6424,
      "step": 522
    },
    {
      "epoch": 2.32,
      "learning_rate": 5.701339470573779e-05,
      "loss": 0.6246,
      "step": 523
    },
    {
      "epoch": 2.33,
      "learning_rate": 5.6872618424977996e-05,
      "loss": 0.7625,
      "step": 524
    },
    {
      "epoch": 2.33,
      "learning_rate": 5.673178659254697e-05,
      "loss": 0.6304,
      "step": 525
    },
    {
      "epoch": 2.34,
      "learning_rate": 5.659090034679451e-05,
      "loss": 0.6563,
      "step": 526
    },
    {
      "epoch": 2.34,
      "learning_rate": 5.644996082651017e-05,
      "loss": 0.7153,
      "step": 527
    },
    {
      "epoch": 2.35,
      "learning_rate": 5.6308969170914214e-05,
      "loss": 0.6987,
      "step": 528
    },
    {
      "epoch": 2.35,
      "learning_rate": 5.6167926519648275e-05,
      "loss": 0.7146,
      "step": 529
    },
    {
      "epoch": 2.36,
      "learning_rate": 5.602683401276615e-05,
      "loss": 0.6472,
      "step": 530
    },
    {
      "epoch": 2.36,
      "learning_rate": 5.588569279072471e-05,
      "loss": 0.6651,
      "step": 531
    },
    {
      "epoch": 2.36,
      "learning_rate": 5.574450399437452e-05,
      "loss": 0.6967,
      "step": 532
    },
    {
      "epoch": 2.37,
      "learning_rate": 5.56032687649507e-05,
      "loss": 0.6471,
      "step": 533
    },
    {
      "epoch": 2.37,
      "learning_rate": 5.546198824406372e-05,
      "loss": 0.6559,
      "step": 534
    },
    {
      "epoch": 2.38,
      "learning_rate": 5.532066357369012e-05,
      "loss": 0.7009,
      "step": 535
    },
    {
      "epoch": 2.38,
      "learning_rate": 5.5179295896163306e-05,
      "loss": 0.6793,
      "step": 536
    },
    {
      "epoch": 2.39,
      "learning_rate": 5.5037886354164315e-05,
      "loss": 0.7662,
      "step": 537
    },
    {
      "epoch": 2.39,
      "learning_rate": 5.489643609071259e-05,
      "loss": 0.6045,
      "step": 538
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.475494624915668e-05,
      "loss": 0.6578,
      "step": 539
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.4613417973165106e-05,
      "loss": 0.6901,
      "step": 540
    },
    {
      "epoch": 2.4,
      "eval_loss": 0.7356687188148499,
      "eval_runtime": 16.1873,
      "eval_samples_per_second": 5.004,
      "eval_steps_per_second": 1.297,
      "step": 540
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.447185240671703e-05,
      "loss": 0.7187,
      "step": 541
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.433025069409301e-05,
      "loss": 0.6906,
      "step": 542
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.418861397986581e-05,
      "loss": 0.6637,
      "step": 543
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.404694340889112e-05,
      "loss": 0.7194,
      "step": 544
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.390524012629824e-05,
      "loss": 0.6827,
      "step": 545
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.376350527748094e-05,
      "loss": 0.6821,
      "step": 546
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.3621740008088126e-05,
      "loss": 0.7305,
      "step": 547
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.347994546401457e-05,
      "loss": 0.5709,
      "step": 548
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.333812279139169e-05,
      "loss": 0.6363,
      "step": 549
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.3196273136578286e-05,
      "loss": 0.7283,
      "step": 550
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.305439764615121e-05,
      "loss": 0.7929,
      "step": 551
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.291249746689619e-05,
      "loss": 0.7669,
      "step": 552
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.27705737457985e-05,
      "loss": 0.722,
      "step": 553
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.262862763003369e-05,
      "loss": 0.6837,
      "step": 554
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.248666026695834e-05,
      "loss": 0.753,
      "step": 555
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.234467280410078e-05,
      "loss": 0.7146,
      "step": 556
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.220266638915178e-05,
      "loss": 0.6773,
      "step": 557
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.206064216995532e-05,
      "loss": 0.7773,
      "step": 558
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.191860129449931e-05,
      "loss": 0.7431,
      "step": 559
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.177654491090627e-05,
      "loss": 0.7091,
      "step": 560
    },
    {
      "epoch": 2.49,
      "eval_loss": 0.7365412712097168,
      "eval_runtime": 16.1735,
      "eval_samples_per_second": 5.008,
      "eval_steps_per_second": 1.298,
      "step": 560
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.163447416742405e-05,
      "loss": 0.6741,
      "step": 561
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.149239021241663e-05,
      "loss": 0.6587,
      "step": 562
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.1350294194354744e-05,
      "loss": 0.6847,
      "step": 563
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.1208187261806615e-05,
      "loss": 0.7163,
      "step": 564
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.106607056342874e-05,
      "loss": 0.6476,
      "step": 565
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.092394524795649e-05,
      "loss": 0.6942,
      "step": 566
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.0781812464194955e-05,
      "loss": 0.6458,
      "step": 567
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.0639673361009544e-05,
      "loss": 0.6589,
      "step": 568
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.049752908731675e-05,
      "loss": 0.6806,
      "step": 569
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.035538079207488e-05,
      "loss": 0.6733,
      "step": 570
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.021322962427475e-05,
      "loss": 0.7576,
      "step": 571
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.0071076732930355e-05,
      "loss": 0.7129,
      "step": 572
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.9928923267069656e-05,
      "loss": 0.7436,
      "step": 573
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.978677037572526e-05,
      "loss": 0.6551,
      "step": 574
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.964461920792512e-05,
      "loss": 0.6978,
      "step": 575
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.950247091268326e-05,
      "loss": 0.7198,
      "step": 576
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.936032663899046e-05,
      "loss": 0.6327,
      "step": 577
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.921818753580506e-05,
      "loss": 0.7077,
      "step": 578
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.907605475204352e-05,
      "loss": 0.7271,
      "step": 579
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.893392943657127e-05,
      "loss": 0.6951,
      "step": 580
    },
    {
      "epoch": 2.58,
      "eval_loss": 0.7344306111335754,
      "eval_runtime": 16.1779,
      "eval_samples_per_second": 5.007,
      "eval_steps_per_second": 1.298,
      "step": 580
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.87918127381934e-05,
      "loss": 0.6326,
      "step": 581
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.8649705805645274e-05,
      "loss": 0.733,
      "step": 582
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.850760978758338e-05,
      "loss": 0.6771,
      "step": 583
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.836552583257597e-05,
      "loss": 0.673,
      "step": 584
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.822345508909376e-05,
      "loss": 0.7064,
      "step": 585
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.8081398705500706e-05,
      "loss": 0.6944,
      "step": 586
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.79393578300447e-05,
      "loss": 0.6399,
      "step": 587
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.7797333610848246e-05,
      "loss": 0.646,
      "step": 588
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.7655327195899244e-05,
      "loss": 0.6829,
      "step": 589
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.7513339733041663e-05,
      "loss": 0.6971,
      "step": 590
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.737137236996631e-05,
      "loss": 0.6354,
      "step": 591
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.72294262542015e-05,
      "loss": 0.6524,
      "step": 592
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.7087502533103806e-05,
      "loss": 0.6613,
      "step": 593
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.694560235384879e-05,
      "loss": 0.6854,
      "step": 594
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.6803726863421725e-05,
      "loss": 0.6929,
      "step": 595
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.666187720860831e-05,
      "loss": 0.6584,
      "step": 596
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.652005453598544e-05,
      "loss": 0.7853,
      "step": 597
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.6378259991911886e-05,
      "loss": 0.6494,
      "step": 598
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.623649472251907e-05,
      "loss": 0.617,
      "step": 599
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.609475987370177e-05,
      "loss": 0.5959,
      "step": 600
    },
    {
      "epoch": 2.67,
      "eval_loss": 0.7361752390861511,
      "eval_runtime": 16.2017,
      "eval_samples_per_second": 4.999,
      "eval_steps_per_second": 1.296,
      "step": 600
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.595305659110889e-05,
      "loss": 0.7532,
      "step": 601
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.5811386020134205e-05,
      "loss": 0.7152,
      "step": 602
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.566974930590701e-05,
      "loss": 0.6756,
      "step": 603
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.552814759328299e-05,
      "loss": 0.7294,
      "step": 604
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.5386582026834906e-05,
      "loss": 0.7813,
      "step": 605
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.524505375084333e-05,
      "loss": 0.721,
      "step": 606
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.510356390928742e-05,
      "loss": 0.7941,
      "step": 607
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.49621136458357e-05,
      "loss": 0.7977,
      "step": 608
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.4820704103836705e-05,
      "loss": 0.6903,
      "step": 609
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.467933642630989e-05,
      "loss": 0.7076,
      "step": 610
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.453801175593629e-05,
      "loss": 0.6853,
      "step": 611
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.439673123504931e-05,
      "loss": 0.6855,
      "step": 612
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.425549600562549e-05,
      "loss": 0.7278,
      "step": 613
    },
    {
      "epoch": 2.73,
      "learning_rate": 4.411430720927531e-05,
      "loss": 0.7256,
      "step": 614
    },
    {
      "epoch": 2.73,
      "learning_rate": 4.397316598723385e-05,
      "loss": 0.7019,
      "step": 615
    },
    {
      "epoch": 2.74,
      "learning_rate": 4.383207348035175e-05,
      "loss": 0.6817,
      "step": 616
    },
    {
      "epoch": 2.74,
      "learning_rate": 4.369103082908581e-05,
      "loss": 0.7147,
      "step": 617
    },
    {
      "epoch": 2.75,
      "learning_rate": 4.3550039173489845e-05,
      "loss": 0.6466,
      "step": 618
    },
    {
      "epoch": 2.75,
      "learning_rate": 4.340909965320552e-05,
      "loss": 0.7691,
      "step": 619
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.326821340745304e-05,
      "loss": 0.619,
      "step": 620
    },
    {
      "epoch": 2.76,
      "eval_loss": 0.7325919270515442,
      "eval_runtime": 16.2144,
      "eval_samples_per_second": 4.996,
      "eval_steps_per_second": 1.295,
      "step": 620
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.312738157502202e-05,
      "loss": 0.5971,
      "step": 621
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.298660529426223e-05,
      "loss": 0.6656,
      "step": 622
    },
    {
      "epoch": 2.77,
      "learning_rate": 4.284588570307442e-05,
      "loss": 0.7092,
      "step": 623
    },
    {
      "epoch": 2.77,
      "learning_rate": 4.270522393890112e-05,
      "loss": 0.6887,
      "step": 624
    },
    {
      "epoch": 2.78,
      "learning_rate": 4.2564621138717407e-05,
      "loss": 0.739,
      "step": 625
    },
    {
      "epoch": 2.78,
      "learning_rate": 4.2424078439021844e-05,
      "loss": 0.6589,
      "step": 626
    },
    {
      "epoch": 2.79,
      "learning_rate": 4.22835969758271e-05,
      "loss": 0.6241,
      "step": 627
    },
    {
      "epoch": 2.79,
      "learning_rate": 4.21431778846509e-05,
      "loss": 0.6595,
      "step": 628
    },
    {
      "epoch": 2.8,
      "learning_rate": 4.200282230050683e-05,
      "loss": 0.6708,
      "step": 629
    },
    {
      "epoch": 2.8,
      "learning_rate": 4.186253135789511e-05,
      "loss": 0.661,
      "step": 630
    },
    {
      "epoch": 2.8,
      "learning_rate": 4.1722306190793495e-05,
      "loss": 0.7229,
      "step": 631
    },
    {
      "epoch": 2.81,
      "learning_rate": 4.1582147932648074e-05,
      "loss": 0.5988,
      "step": 632
    },
    {
      "epoch": 2.81,
      "learning_rate": 4.144205771636407e-05,
      "loss": 0.7131,
      "step": 633
    },
    {
      "epoch": 2.82,
      "learning_rate": 4.1302036674296756e-05,
      "loss": 0.6935,
      "step": 634
    },
    {
      "epoch": 2.82,
      "learning_rate": 4.116208593824227e-05,
      "loss": 0.7132,
      "step": 635
    },
    {
      "epoch": 2.83,
      "learning_rate": 4.102220663942841e-05,
      "loss": 0.7209,
      "step": 636
    },
    {
      "epoch": 2.83,
      "learning_rate": 4.088239990850562e-05,
      "loss": 0.712,
      "step": 637
    },
    {
      "epoch": 2.84,
      "learning_rate": 4.074266687553773e-05,
      "loss": 0.7514,
      "step": 638
    },
    {
      "epoch": 2.84,
      "learning_rate": 4.060300866999286e-05,
      "loss": 0.6708,
      "step": 639
    },
    {
      "epoch": 2.84,
      "learning_rate": 4.0463426420734323e-05,
      "loss": 0.6953,
      "step": 640
    },
    {
      "epoch": 2.84,
      "eval_loss": 0.7320604920387268,
      "eval_runtime": 16.1931,
      "eval_samples_per_second": 5.002,
      "eval_steps_per_second": 1.297,
      "step": 640
    },
    {
      "epoch": 2.85,
      "learning_rate": 4.032392125601147e-05,
      "loss": 0.6768,
      "step": 641
    },
    {
      "epoch": 2.85,
      "learning_rate": 4.018449430345054e-05,
      "loss": 0.6666,
      "step": 642
    },
    {
      "epoch": 2.86,
      "learning_rate": 4.004514669004562e-05,
      "loss": 0.6697,
      "step": 643
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.99058795421495e-05,
      "loss": 0.6809,
      "step": 644
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.976669398546451e-05,
      "loss": 0.6717,
      "step": 645
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.962759114503353e-05,
      "loss": 0.6091,
      "step": 646
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.948857214523081e-05,
      "loss": 0.6579,
      "step": 647
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.934963810975294e-05,
      "loss": 0.7374,
      "step": 648
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.92107901616097e-05,
      "loss": 0.6407,
      "step": 649
    },
    {
      "epoch": 2.89,
      "learning_rate": 3.9072029423115055e-05,
      "loss": 0.6414,
      "step": 650
    },
    {
      "epoch": 2.89,
      "learning_rate": 3.8933357015878064e-05,
      "loss": 0.7717,
      "step": 651
    },
    {
      "epoch": 2.9,
      "learning_rate": 3.8794774060793735e-05,
      "loss": 0.6792,
      "step": 652
    },
    {
      "epoch": 2.9,
      "learning_rate": 3.86562816780341e-05,
      "loss": 0.7022,
      "step": 653
    },
    {
      "epoch": 2.91,
      "learning_rate": 3.851788098703907e-05,
      "loss": 0.729,
      "step": 654
    },
    {
      "epoch": 2.91,
      "learning_rate": 3.837957310650738e-05,
      "loss": 0.7356,
      "step": 655
    },
    {
      "epoch": 2.92,
      "learning_rate": 3.8241359154387595e-05,
      "loss": 0.6932,
      "step": 656
    },
    {
      "epoch": 2.92,
      "learning_rate": 3.8103240247869075e-05,
      "loss": 0.6775,
      "step": 657
    },
    {
      "epoch": 2.92,
      "learning_rate": 3.7965217503372877e-05,
      "loss": 0.6587,
      "step": 658
    },
    {
      "epoch": 2.93,
      "learning_rate": 3.7827292036542805e-05,
      "loss": 0.7077,
      "step": 659
    },
    {
      "epoch": 2.93,
      "learning_rate": 3.768946496223637e-05,
      "loss": 0.6363,
      "step": 660
    },
    {
      "epoch": 2.93,
      "eval_loss": 0.7314029932022095,
      "eval_runtime": 16.2089,
      "eval_samples_per_second": 4.997,
      "eval_steps_per_second": 1.296,
      "step": 660
    },
    {
      "epoch": 2.94,
      "learning_rate": 3.755173739451575e-05,
      "loss": 0.6865,
      "step": 661
    },
    {
      "epoch": 2.94,
      "learning_rate": 3.7414110446638825e-05,
      "loss": 0.7104,
      "step": 662
    },
    {
      "epoch": 2.95,
      "learning_rate": 3.7276585231050145e-05,
      "loss": 0.6059,
      "step": 663
    },
    {
      "epoch": 2.95,
      "learning_rate": 3.7139162859371955e-05,
      "loss": 0.6314,
      "step": 664
    },
    {
      "epoch": 2.96,
      "learning_rate": 3.700184444239524e-05,
      "loss": 0.7693,
      "step": 665
    },
    {
      "epoch": 2.96,
      "learning_rate": 3.6864631090070655e-05,
      "loss": 0.6617,
      "step": 666
    },
    {
      "epoch": 2.96,
      "learning_rate": 3.672752391149966e-05,
      "loss": 0.6895,
      "step": 667
    },
    {
      "epoch": 2.97,
      "learning_rate": 3.659052401492551e-05,
      "loss": 0.7244,
      "step": 668
    },
    {
      "epoch": 2.97,
      "learning_rate": 3.645363250772425e-05,
      "loss": 0.6243,
      "step": 669
    },
    {
      "epoch": 2.98,
      "learning_rate": 3.631685049639586e-05,
      "loss": 0.5942,
      "step": 670
    },
    {
      "epoch": 2.98,
      "learning_rate": 3.618017908655523e-05,
      "loss": 0.6391,
      "step": 671
    },
    {
      "epoch": 2.99,
      "learning_rate": 3.6043619382923274e-05,
      "loss": 0.6535,
      "step": 672
    },
    {
      "epoch": 2.99,
      "learning_rate": 3.590717248931794e-05,
      "loss": 0.602,
      "step": 673
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.5770839508645385e-05,
      "loss": 0.6396,
      "step": 674
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.563462154289098e-05,
      "loss": 0.725,
      "step": 675
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.5498519693110375e-05,
      "loss": 0.8034,
      "step": 676
    },
    {
      "epoch": 3.01,
      "learning_rate": 3.536253505942073e-05,
      "loss": 0.6909,
      "step": 677
    },
    {
      "epoch": 3.01,
      "learning_rate": 3.5226668740991705e-05,
      "loss": 0.7072,
      "step": 678
    },
    {
      "epoch": 3.02,
      "learning_rate": 3.509092183603659e-05,
      "loss": 0.7086,
      "step": 679
    },
    {
      "epoch": 3.02,
      "learning_rate": 3.4955295441803495e-05,
      "loss": 0.7257,
      "step": 680
    },
    {
      "epoch": 3.02,
      "eval_loss": 0.7309479117393494,
      "eval_runtime": 16.2583,
      "eval_samples_per_second": 4.982,
      "eval_steps_per_second": 1.292,
      "step": 680
    },
    {
      "epoch": 3.03,
      "learning_rate": 3.481979065456642e-05,
      "loss": 0.6641,
      "step": 681
    },
    {
      "epoch": 3.03,
      "learning_rate": 3.468440856961639e-05,
      "loss": 0.675,
      "step": 682
    },
    {
      "epoch": 3.04,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.6748,
      "step": 683
    },
    {
      "epoch": 3.04,
      "learning_rate": 3.4414016882773757e-05,
      "loss": 0.6625,
      "step": 684
    },
    {
      "epoch": 3.04,
      "learning_rate": 3.427900946646882e-05,
      "loss": 0.6625,
      "step": 685
    },
    {
      "epoch": 3.05,
      "learning_rate": 3.414412912360861e-05,
      "loss": 0.6641,
      "step": 686
    },
    {
      "epoch": 3.05,
      "learning_rate": 3.400937694443678e-05,
      "loss": 0.6629,
      "step": 687
    },
    {
      "epoch": 3.06,
      "learning_rate": 3.387475401816096e-05,
      "loss": 0.6882,
      "step": 688
    },
    {
      "epoch": 3.06,
      "learning_rate": 3.37402614329441e-05,
      "loss": 0.6896,
      "step": 689
    },
    {
      "epoch": 3.07,
      "learning_rate": 3.360590027589556e-05,
      "loss": 0.6455,
      "step": 690
    },
    {
      "epoch": 3.07,
      "learning_rate": 3.3471671633062376e-05,
      "loss": 0.6925,
      "step": 691
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.333757658942045e-05,
      "loss": 0.5948,
      "step": 692
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.320361622886581e-05,
      "loss": 0.6794,
      "step": 693
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.306979163420582e-05,
      "loss": 0.733,
      "step": 694
    },
    {
      "epoch": 3.09,
      "learning_rate": 3.293610388715048e-05,
      "loss": 0.6766,
      "step": 695
    },
    {
      "epoch": 3.09,
      "learning_rate": 3.2802554068303596e-05,
      "loss": 0.6086,
      "step": 696
    },
    {
      "epoch": 3.1,
      "learning_rate": 3.2669143257154113e-05,
      "loss": 0.6252,
      "step": 697
    },
    {
      "epoch": 3.1,
      "learning_rate": 3.253587253206738e-05,
      "loss": 0.6326,
      "step": 698
    },
    {
      "epoch": 3.11,
      "learning_rate": 3.2402742970276426e-05,
      "loss": 0.7659,
      "step": 699
    },
    {
      "epoch": 3.11,
      "learning_rate": 3.226975564787322e-05,
      "loss": 0.6001,
      "step": 700
    },
    {
      "epoch": 3.11,
      "eval_loss": 0.7316482067108154,
      "eval_runtime": 16.2988,
      "eval_samples_per_second": 4.97,
      "eval_steps_per_second": 1.288,
      "step": 700
    },
    {
      "epoch": 3.12,
      "learning_rate": 3.213691163980004e-05,
      "loss": 0.6963,
      "step": 701
    },
    {
      "epoch": 3.12,
      "learning_rate": 3.200421201984074e-05,
      "loss": 0.6698,
      "step": 702
    },
    {
      "epoch": 3.12,
      "learning_rate": 3.187165786061206e-05,
      "loss": 0.6681,
      "step": 703
    },
    {
      "epoch": 3.13,
      "learning_rate": 3.1739250233554996e-05,
      "loss": 0.6811,
      "step": 704
    },
    {
      "epoch": 3.13,
      "learning_rate": 3.160699020892612e-05,
      "loss": 0.6669,
      "step": 705
    },
    {
      "epoch": 3.14,
      "learning_rate": 3.1474878855788895e-05,
      "loss": 0.6001,
      "step": 706
    },
    {
      "epoch": 3.14,
      "learning_rate": 3.13429172420051e-05,
      "loss": 0.6296,
      "step": 707
    },
    {
      "epoch": 3.15,
      "learning_rate": 3.121110643422615e-05,
      "loss": 0.6481,
      "step": 708
    },
    {
      "epoch": 3.15,
      "learning_rate": 3.107944749788449e-05,
      "loss": 0.6581,
      "step": 709
    },
    {
      "epoch": 3.16,
      "learning_rate": 3.094794149718498e-05,
      "loss": 0.7019,
      "step": 710
    },
    {
      "epoch": 3.16,
      "learning_rate": 3.0816589495096317e-05,
      "loss": 0.6645,
      "step": 711
    },
    {
      "epoch": 3.16,
      "learning_rate": 3.068539255334237e-05,
      "loss": 0.6239,
      "step": 712
    },
    {
      "epoch": 3.17,
      "learning_rate": 3.05543517323937e-05,
      "loss": 0.6124,
      "step": 713
    },
    {
      "epoch": 3.17,
      "learning_rate": 3.0423468091458918e-05,
      "loss": 0.6291,
      "step": 714
    },
    {
      "epoch": 3.18,
      "learning_rate": 3.0292742688476122e-05,
      "loss": 0.6312,
      "step": 715
    },
    {
      "epoch": 3.18,
      "learning_rate": 3.0162176580104397e-05,
      "loss": 0.6532,
      "step": 716
    },
    {
      "epoch": 3.19,
      "learning_rate": 3.003177082171523e-05,
      "loss": 0.6687,
      "step": 717
    },
    {
      "epoch": 3.19,
      "learning_rate": 2.9901526467383987e-05,
      "loss": 0.5563,
      "step": 718
    },
    {
      "epoch": 3.2,
      "learning_rate": 2.9771444569881413e-05,
      "loss": 0.666,
      "step": 719
    },
    {
      "epoch": 3.2,
      "learning_rate": 2.964152618066508e-05,
      "loss": 0.6246,
      "step": 720
    },
    {
      "epoch": 3.2,
      "eval_loss": 0.7331957221031189,
      "eval_runtime": 16.3691,
      "eval_samples_per_second": 4.948,
      "eval_steps_per_second": 1.283,
      "step": 720
    },
    {
      "epoch": 3.2,
      "learning_rate": 2.9511772349870958e-05,
      "loss": 0.6019,
      "step": 721
    },
    {
      "epoch": 3.21,
      "learning_rate": 2.9382184126304834e-05,
      "loss": 0.7261,
      "step": 722
    },
    {
      "epoch": 3.21,
      "learning_rate": 2.9252762557433922e-05,
      "loss": 0.6073,
      "step": 723
    },
    {
      "epoch": 3.22,
      "learning_rate": 2.9123508689378352e-05,
      "loss": 0.6041,
      "step": 724
    },
    {
      "epoch": 3.22,
      "learning_rate": 2.8994423566902707e-05,
      "loss": 0.6246,
      "step": 725
    },
    {
      "epoch": 3.23,
      "learning_rate": 2.8865508233407595e-05,
      "loss": 0.7411,
      "step": 726
    },
    {
      "epoch": 3.23,
      "learning_rate": 2.8736763730921228e-05,
      "loss": 0.7378,
      "step": 727
    },
    {
      "epoch": 3.24,
      "learning_rate": 2.8608191100090977e-05,
      "loss": 0.694,
      "step": 728
    },
    {
      "epoch": 3.24,
      "learning_rate": 2.847979138017496e-05,
      "loss": 0.7049,
      "step": 729
    },
    {
      "epoch": 3.24,
      "learning_rate": 2.835156560903365e-05,
      "loss": 0.6776,
      "step": 730
    },
    {
      "epoch": 3.25,
      "learning_rate": 2.822351482312149e-05,
      "loss": 0.7103,
      "step": 731
    },
    {
      "epoch": 3.25,
      "learning_rate": 2.80956400574785e-05,
      "loss": 0.6989,
      "step": 732
    },
    {
      "epoch": 3.26,
      "learning_rate": 2.7967942345721964e-05,
      "loss": 0.7524,
      "step": 733
    },
    {
      "epoch": 3.26,
      "learning_rate": 2.784042272003794e-05,
      "loss": 0.7126,
      "step": 734
    },
    {
      "epoch": 3.27,
      "learning_rate": 2.771308221117309e-05,
      "loss": 0.6255,
      "step": 735
    },
    {
      "epoch": 3.27,
      "learning_rate": 2.7585921848426242e-05,
      "loss": 0.663,
      "step": 736
    },
    {
      "epoch": 3.28,
      "learning_rate": 2.7458942659640085e-05,
      "loss": 0.6511,
      "step": 737
    },
    {
      "epoch": 3.28,
      "learning_rate": 2.73321456711929e-05,
      "loss": 0.6723,
      "step": 738
    },
    {
      "epoch": 3.28,
      "learning_rate": 2.720553190799019e-05,
      "loss": 0.6504,
      "step": 739
    },
    {
      "epoch": 3.29,
      "learning_rate": 2.7079102393456503e-05,
      "loss": 0.6465,
      "step": 740
    },
    {
      "epoch": 3.29,
      "eval_loss": 0.7326642274856567,
      "eval_runtime": 16.2926,
      "eval_samples_per_second": 4.972,
      "eval_steps_per_second": 1.289,
      "step": 740
    },
    {
      "epoch": 3.29,
      "learning_rate": 2.6952858149527017e-05,
      "loss": 0.6594,
      "step": 741
    },
    {
      "epoch": 3.3,
      "learning_rate": 2.6826800196639412e-05,
      "loss": 0.6149,
      "step": 742
    },
    {
      "epoch": 3.3,
      "learning_rate": 2.6700929553725573e-05,
      "loss": 0.6817,
      "step": 743
    },
    {
      "epoch": 3.31,
      "learning_rate": 2.6575247238203328e-05,
      "loss": 0.6066,
      "step": 744
    },
    {
      "epoch": 3.31,
      "learning_rate": 2.6449754265968264e-05,
      "loss": 0.6141,
      "step": 745
    },
    {
      "epoch": 3.32,
      "learning_rate": 2.6324451651385495e-05,
      "loss": 0.6716,
      "step": 746
    },
    {
      "epoch": 3.32,
      "learning_rate": 2.6199340407281435e-05,
      "loss": 0.6897,
      "step": 747
    },
    {
      "epoch": 3.32,
      "learning_rate": 2.607442154493568e-05,
      "loss": 0.6212,
      "step": 748
    },
    {
      "epoch": 3.33,
      "learning_rate": 2.5949696074072786e-05,
      "loss": 0.6493,
      "step": 749
    },
    {
      "epoch": 3.33,
      "learning_rate": 2.5825165002854124e-05,
      "loss": 0.6363,
      "step": 750
    },
    {
      "epoch": 3.34,
      "learning_rate": 2.57008293378697e-05,
      "loss": 0.6591,
      "step": 751
    },
    {
      "epoch": 3.34,
      "learning_rate": 2.5576690084130085e-05,
      "loss": 0.6951,
      "step": 752
    },
    {
      "epoch": 3.35,
      "learning_rate": 2.5452748245058177e-05,
      "loss": 0.6369,
      "step": 753
    },
    {
      "epoch": 3.35,
      "learning_rate": 2.532900482248124e-05,
      "loss": 0.6685,
      "step": 754
    },
    {
      "epoch": 3.36,
      "learning_rate": 2.5205460816622684e-05,
      "loss": 0.6467,
      "step": 755
    },
    {
      "epoch": 3.36,
      "learning_rate": 2.5082117226094047e-05,
      "loss": 0.6581,
      "step": 756
    },
    {
      "epoch": 3.36,
      "learning_rate": 2.49589750478869e-05,
      "loss": 0.7478,
      "step": 757
    },
    {
      "epoch": 3.37,
      "learning_rate": 2.4836035277364767e-05,
      "loss": 0.7247,
      "step": 758
    },
    {
      "epoch": 3.37,
      "learning_rate": 2.471329890825514e-05,
      "loss": 0.6335,
      "step": 759
    },
    {
      "epoch": 3.38,
      "learning_rate": 2.4590766932641353e-05,
      "loss": 0.6239,
      "step": 760
    },
    {
      "epoch": 3.38,
      "eval_loss": 0.7320056557655334,
      "eval_runtime": 16.2722,
      "eval_samples_per_second": 4.978,
      "eval_steps_per_second": 1.291,
      "step": 760
    },
    {
      "epoch": 3.38,
      "learning_rate": 2.446844034095466e-05,
      "loss": 0.6315,
      "step": 761
    },
    {
      "epoch": 3.39,
      "learning_rate": 2.4346320121966194e-05,
      "loss": 0.6445,
      "step": 762
    },
    {
      "epoch": 3.39,
      "learning_rate": 2.4224407262778925e-05,
      "loss": 0.5742,
      "step": 763
    },
    {
      "epoch": 3.4,
      "learning_rate": 2.410270274881981e-05,
      "loss": 0.581,
      "step": 764
    },
    {
      "epoch": 3.4,
      "learning_rate": 2.398120756383163e-05,
      "loss": 0.6181,
      "step": 765
    },
    {
      "epoch": 3.4,
      "learning_rate": 2.3859922689865234e-05,
      "loss": 0.624,
      "step": 766
    },
    {
      "epoch": 3.41,
      "learning_rate": 2.3738849107271476e-05,
      "loss": 0.6646,
      "step": 767
    },
    {
      "epoch": 3.41,
      "learning_rate": 2.361798779469336e-05,
      "loss": 0.6389,
      "step": 768
    },
    {
      "epoch": 3.42,
      "learning_rate": 2.3497339729058083e-05,
      "loss": 0.6622,
      "step": 769
    },
    {
      "epoch": 3.42,
      "learning_rate": 2.3376905885569182e-05,
      "loss": 0.6558,
      "step": 770
    },
    {
      "epoch": 3.43,
      "learning_rate": 2.325668723769858e-05,
      "loss": 0.6956,
      "step": 771
    },
    {
      "epoch": 3.43,
      "learning_rate": 2.3136684757178806e-05,
      "loss": 0.6325,
      "step": 772
    },
    {
      "epoch": 3.44,
      "learning_rate": 2.3016899413995092e-05,
      "loss": 0.5749,
      "step": 773
    },
    {
      "epoch": 3.44,
      "learning_rate": 2.2897332176377528e-05,
      "loss": 0.5596,
      "step": 774
    },
    {
      "epoch": 3.44,
      "learning_rate": 2.2777984010793264e-05,
      "loss": 0.6344,
      "step": 775
    },
    {
      "epoch": 3.45,
      "learning_rate": 2.2658855881938685e-05,
      "loss": 0.7134,
      "step": 776
    },
    {
      "epoch": 3.45,
      "learning_rate": 2.2539948752731555e-05,
      "loss": 0.6477,
      "step": 777
    },
    {
      "epoch": 3.46,
      "learning_rate": 2.2421263584303355e-05,
      "loss": 0.7192,
      "step": 778
    },
    {
      "epoch": 3.46,
      "learning_rate": 2.2302801335991413e-05,
      "loss": 0.7168,
      "step": 779
    },
    {
      "epoch": 3.47,
      "learning_rate": 2.2184562965331202e-05,
      "loss": 0.754,
      "step": 780
    },
    {
      "epoch": 3.47,
      "eval_loss": 0.7345401048660278,
      "eval_runtime": 16.2731,
      "eval_samples_per_second": 4.978,
      "eval_steps_per_second": 1.29,
      "step": 780
    },
    {
      "epoch": 3.47,
      "learning_rate": 2.206654942804857e-05,
      "loss": 0.7341,
      "step": 781
    },
    {
      "epoch": 3.48,
      "learning_rate": 2.1948761678052028e-05,
      "loss": 0.7202,
      "step": 782
    },
    {
      "epoch": 3.48,
      "learning_rate": 2.183120066742506e-05,
      "loss": 0.7192,
      "step": 783
    },
    {
      "epoch": 3.48,
      "learning_rate": 2.1713867346418354e-05,
      "loss": 0.8172,
      "step": 784
    },
    {
      "epoch": 3.49,
      "learning_rate": 2.1596762663442218e-05,
      "loss": 0.6996,
      "step": 785
    },
    {
      "epoch": 3.49,
      "learning_rate": 2.147988756505886e-05,
      "loss": 0.7763,
      "step": 786
    },
    {
      "epoch": 3.5,
      "learning_rate": 2.136324299597474e-05,
      "loss": 0.6673,
      "step": 787
    },
    {
      "epoch": 3.5,
      "learning_rate": 2.124682989903295e-05,
      "loss": 0.6465,
      "step": 788
    },
    {
      "epoch": 3.51,
      "learning_rate": 2.1130649215205584e-05,
      "loss": 0.7541,
      "step": 789
    },
    {
      "epoch": 3.51,
      "learning_rate": 2.1014701883586085e-05,
      "loss": 0.669,
      "step": 790
    },
    {
      "epoch": 3.52,
      "learning_rate": 2.089898884138177e-05,
      "loss": 0.7063,
      "step": 791
    },
    {
      "epoch": 3.52,
      "learning_rate": 2.0783511023906128e-05,
      "loss": 0.6963,
      "step": 792
    },
    {
      "epoch": 3.52,
      "learning_rate": 2.0668269364571358e-05,
      "loss": 0.6246,
      "step": 793
    },
    {
      "epoch": 3.53,
      "learning_rate": 2.0553264794880756e-05,
      "loss": 0.658,
      "step": 794
    },
    {
      "epoch": 3.53,
      "learning_rate": 2.043849824442124e-05,
      "loss": 0.6113,
      "step": 795
    },
    {
      "epoch": 3.54,
      "learning_rate": 2.0323970640855748e-05,
      "loss": 0.6036,
      "step": 796
    },
    {
      "epoch": 3.54,
      "learning_rate": 2.0209682909915855e-05,
      "loss": 0.6178,
      "step": 797
    },
    {
      "epoch": 3.55,
      "learning_rate": 2.0095635975394238e-05,
      "loss": 0.7395,
      "step": 798
    },
    {
      "epoch": 3.55,
      "learning_rate": 1.9981830759137186e-05,
      "loss": 0.6803,
      "step": 799
    },
    {
      "epoch": 3.56,
      "learning_rate": 1.9868268181037185e-05,
      "loss": 0.6661,
      "step": 800
    },
    {
      "epoch": 3.56,
      "eval_loss": 0.7309223413467407,
      "eval_runtime": 16.2587,
      "eval_samples_per_second": 4.982,
      "eval_steps_per_second": 1.292,
      "step": 800
    },
    {
      "epoch": 3.56,
      "learning_rate": 1.9754949159025415e-05,
      "loss": 0.665,
      "step": 801
    },
    {
      "epoch": 3.56,
      "learning_rate": 1.9641874609064443e-05,
      "loss": 0.6056,
      "step": 802
    },
    {
      "epoch": 3.57,
      "learning_rate": 1.9529045445140725e-05,
      "loss": 0.6633,
      "step": 803
    },
    {
      "epoch": 3.57,
      "learning_rate": 1.9416462579257273e-05,
      "loss": 0.6446,
      "step": 804
    },
    {
      "epoch": 3.58,
      "learning_rate": 1.9304126921426236e-05,
      "loss": 0.6842,
      "step": 805
    },
    {
      "epoch": 3.58,
      "learning_rate": 1.919203937966161e-05,
      "loss": 0.6465,
      "step": 806
    },
    {
      "epoch": 3.59,
      "learning_rate": 1.9080200859971793e-05,
      "loss": 0.6094,
      "step": 807
    },
    {
      "epoch": 3.59,
      "learning_rate": 1.8968612266352376e-05,
      "loss": 0.684,
      "step": 808
    },
    {
      "epoch": 3.6,
      "learning_rate": 1.885727450077879e-05,
      "loss": 0.6624,
      "step": 809
    },
    {
      "epoch": 3.6,
      "learning_rate": 1.8746188463198982e-05,
      "loss": 0.6649,
      "step": 810
    },
    {
      "epoch": 3.6,
      "learning_rate": 1.8635355051526195e-05,
      "loss": 0.7099,
      "step": 811
    },
    {
      "epoch": 3.61,
      "learning_rate": 1.8524775161631675e-05,
      "loss": 0.658,
      "step": 812
    },
    {
      "epoch": 3.61,
      "learning_rate": 1.8414449687337464e-05,
      "loss": 0.6599,
      "step": 813
    },
    {
      "epoch": 3.62,
      "learning_rate": 1.8304379520409087e-05,
      "loss": 0.6879,
      "step": 814
    },
    {
      "epoch": 3.62,
      "learning_rate": 1.8194565550548476e-05,
      "loss": 0.6552,
      "step": 815
    },
    {
      "epoch": 3.63,
      "learning_rate": 1.8085008665386687e-05,
      "loss": 0.6411,
      "step": 816
    },
    {
      "epoch": 3.63,
      "learning_rate": 1.7975709750476744e-05,
      "loss": 0.6946,
      "step": 817
    },
    {
      "epoch": 3.64,
      "learning_rate": 1.7866669689286496e-05,
      "loss": 0.693,
      "step": 818
    },
    {
      "epoch": 3.64,
      "learning_rate": 1.7757889363191483e-05,
      "loss": 0.6364,
      "step": 819
    },
    {
      "epoch": 3.64,
      "learning_rate": 1.764936965146773e-05,
      "loss": 0.6127,
      "step": 820
    },
    {
      "epoch": 3.64,
      "eval_loss": 0.730964720249176,
      "eval_runtime": 16.277,
      "eval_samples_per_second": 4.976,
      "eval_steps_per_second": 1.29,
      "step": 820
    },
    {
      "epoch": 3.65,
      "learning_rate": 1.7541111431284773e-05,
      "loss": 0.5951,
      "step": 821
    },
    {
      "epoch": 3.65,
      "learning_rate": 1.743311557769847e-05,
      "loss": 0.6106,
      "step": 822
    },
    {
      "epoch": 3.66,
      "learning_rate": 1.7325382963643976e-05,
      "loss": 0.5896,
      "step": 823
    },
    {
      "epoch": 3.66,
      "learning_rate": 1.7217914459928646e-05,
      "loss": 0.7827,
      "step": 824
    },
    {
      "epoch": 3.67,
      "learning_rate": 1.7110710935225055e-05,
      "loss": 0.5832,
      "step": 825
    },
    {
      "epoch": 3.67,
      "learning_rate": 1.700377325606388e-05,
      "loss": 0.7961,
      "step": 826
    },
    {
      "epoch": 3.68,
      "learning_rate": 1.689710228682702e-05,
      "loss": 0.6715,
      "step": 827
    },
    {
      "epoch": 3.68,
      "learning_rate": 1.679069888974052e-05,
      "loss": 0.6424,
      "step": 828
    },
    {
      "epoch": 3.68,
      "learning_rate": 1.668456392486762e-05,
      "loss": 0.7368,
      "step": 829
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.6578698250101825e-05,
      "loss": 0.7096,
      "step": 830
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.6473102721159957e-05,
      "loss": 0.718,
      "step": 831
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.6367778191575223e-05,
      "loss": 0.7177,
      "step": 832
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.6262725512690345e-05,
      "loss": 0.7212,
      "step": 833
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.615794553365066e-05,
      "loss": 0.6808,
      "step": 834
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.6053439101397255e-05,
      "loss": 0.7125,
      "step": 835
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.5949207060660138e-05,
      "loss": 0.6934,
      "step": 836
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.5845250253951393e-05,
      "loss": 0.6788,
      "step": 837
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.5741569521558352e-05,
      "loss": 0.6613,
      "step": 838
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.5638165701536868e-05,
      "loss": 0.6954,
      "step": 839
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.553503962970447e-05,
      "loss": 0.6538,
      "step": 840
    },
    {
      "epoch": 3.73,
      "eval_loss": 0.731265664100647,
      "eval_runtime": 16.2788,
      "eval_samples_per_second": 4.976,
      "eval_steps_per_second": 1.29,
      "step": 840
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.5432192139633645e-05,
      "loss": 0.6583,
      "step": 841
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.53296240626451e-05,
      "loss": 0.6118,
      "step": 842
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.5227336227801037e-05,
      "loss": 0.5973,
      "step": 843
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.5125329461898408e-05,
      "loss": 0.6266,
      "step": 844
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.502360458946232e-05,
      "loss": 0.6101,
      "step": 845
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.49221624327393e-05,
      "loss": 0.7363,
      "step": 846
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.4821003811690686e-05,
      "loss": 0.6602,
      "step": 847
    },
    {
      "epoch": 3.77,
      "learning_rate": 1.4720129543985973e-05,
      "loss": 0.6837,
      "step": 848
    },
    {
      "epoch": 3.77,
      "learning_rate": 1.4619540444996227e-05,
      "loss": 0.6254,
      "step": 849
    },
    {
      "epoch": 3.78,
      "learning_rate": 1.4519237327787449e-05,
      "loss": 0.6692,
      "step": 850
    },
    {
      "epoch": 3.78,
      "learning_rate": 1.4419221003114081e-05,
      "loss": 0.6508,
      "step": 851
    },
    {
      "epoch": 3.79,
      "learning_rate": 1.4319492279412388e-05,
      "loss": 0.6578,
      "step": 852
    },
    {
      "epoch": 3.79,
      "learning_rate": 1.422005196279395e-05,
      "loss": 0.6257,
      "step": 853
    },
    {
      "epoch": 3.8,
      "learning_rate": 1.4120900857039126e-05,
      "loss": 0.6523,
      "step": 854
    },
    {
      "epoch": 3.8,
      "learning_rate": 1.4022039763590594e-05,
      "loss": 0.6703,
      "step": 855
    },
    {
      "epoch": 3.8,
      "learning_rate": 1.3923469481546841e-05,
      "loss": 0.6597,
      "step": 856
    },
    {
      "epoch": 3.81,
      "learning_rate": 1.3825190807655692e-05,
      "loss": 0.6728,
      "step": 857
    },
    {
      "epoch": 3.81,
      "learning_rate": 1.3727204536307908e-05,
      "loss": 0.6369,
      "step": 858
    },
    {
      "epoch": 3.82,
      "learning_rate": 1.3629511459530758e-05,
      "loss": 0.6164,
      "step": 859
    },
    {
      "epoch": 3.82,
      "learning_rate": 1.3532112366981598e-05,
      "loss": 0.6646,
      "step": 860
    },
    {
      "epoch": 3.82,
      "eval_loss": 0.7303179502487183,
      "eval_runtime": 16.2683,
      "eval_samples_per_second": 4.979,
      "eval_steps_per_second": 1.291,
      "step": 860
    },
    {
      "epoch": 3.83,
      "learning_rate": 1.3435008045941483e-05,
      "loss": 0.6108,
      "step": 861
    },
    {
      "epoch": 3.83,
      "learning_rate": 1.3338199281308855e-05,
      "loss": 0.7907,
      "step": 862
    },
    {
      "epoch": 3.84,
      "learning_rate": 1.3241686855593105e-05,
      "loss": 0.6953,
      "step": 863
    },
    {
      "epoch": 3.84,
      "learning_rate": 1.3145471548908345e-05,
      "loss": 0.5835,
      "step": 864
    },
    {
      "epoch": 3.84,
      "learning_rate": 1.3049554138967051e-05,
      "loss": 0.6939,
      "step": 865
    },
    {
      "epoch": 3.85,
      "learning_rate": 1.2953935401073786e-05,
      "loss": 0.6388,
      "step": 866
    },
    {
      "epoch": 3.85,
      "learning_rate": 1.2858616108118932e-05,
      "loss": 0.5938,
      "step": 867
    },
    {
      "epoch": 3.86,
      "learning_rate": 1.2763597030572449e-05,
      "loss": 0.6354,
      "step": 868
    },
    {
      "epoch": 3.86,
      "learning_rate": 1.266887893647764e-05,
      "loss": 0.6286,
      "step": 869
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.257446259144494e-05,
      "loss": 0.6327,
      "step": 870
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.248034875864575e-05,
      "loss": 0.5474,
      "step": 871
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.238653819880623e-05,
      "loss": 0.6045,
      "step": 872
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.2293031670201205e-05,
      "loss": 0.6502,
      "step": 873
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.2199829928647949e-05,
      "loss": 0.6524,
      "step": 874
    },
    {
      "epoch": 3.89,
      "learning_rate": 1.2106933727500169e-05,
      "loss": 0.6121,
      "step": 875
    },
    {
      "epoch": 3.89,
      "learning_rate": 1.2014343817641872e-05,
      "loss": 0.6873,
      "step": 876
    },
    {
      "epoch": 3.9,
      "learning_rate": 1.192206094748129e-05,
      "loss": 0.7149,
      "step": 877
    },
    {
      "epoch": 3.9,
      "learning_rate": 1.183008586294485e-05,
      "loss": 0.708,
      "step": 878
    },
    {
      "epoch": 3.91,
      "learning_rate": 1.1738419307471138e-05,
      "loss": 0.7516,
      "step": 879
    },
    {
      "epoch": 3.91,
      "learning_rate": 1.1647062022004845e-05,
      "loss": 0.7091,
      "step": 880
    },
    {
      "epoch": 3.91,
      "eval_loss": 0.7308124303817749,
      "eval_runtime": 16.2664,
      "eval_samples_per_second": 4.98,
      "eval_steps_per_second": 1.291,
      "step": 880
    },
    {
      "epoch": 3.92,
      "learning_rate": 1.1556014744990874e-05,
      "loss": 0.7198,
      "step": 881
    },
    {
      "epoch": 3.92,
      "learning_rate": 1.1465278212368285e-05,
      "loss": 0.6556,
      "step": 882
    },
    {
      "epoch": 3.92,
      "learning_rate": 1.1374853157564391e-05,
      "loss": 0.6451,
      "step": 883
    },
    {
      "epoch": 3.93,
      "learning_rate": 1.1284740311488812e-05,
      "loss": 0.6918,
      "step": 884
    },
    {
      "epoch": 3.93,
      "learning_rate": 1.1194940402527564e-05,
      "loss": 0.6494,
      "step": 885
    },
    {
      "epoch": 3.94,
      "learning_rate": 1.1105454156537204e-05,
      "loss": 0.6872,
      "step": 886
    },
    {
      "epoch": 3.94,
      "learning_rate": 1.1016282296838887e-05,
      "loss": 0.6875,
      "step": 887
    },
    {
      "epoch": 3.95,
      "learning_rate": 1.0927425544212622e-05,
      "loss": 0.6745,
      "step": 888
    },
    {
      "epoch": 3.95,
      "learning_rate": 1.083888461689137e-05,
      "loss": 0.6712,
      "step": 889
    },
    {
      "epoch": 3.96,
      "learning_rate": 1.0750660230555271e-05,
      "loss": 0.6447,
      "step": 890
    },
    {
      "epoch": 3.96,
      "learning_rate": 1.066275309832584e-05,
      "loss": 0.6174,
      "step": 891
    },
    {
      "epoch": 3.96,
      "learning_rate": 1.0575163930760235e-05,
      "loss": 0.6796,
      "step": 892
    },
    {
      "epoch": 3.97,
      "learning_rate": 1.0487893435845447e-05,
      "loss": 0.5846,
      "step": 893
    },
    {
      "epoch": 3.97,
      "learning_rate": 1.0400942318992668e-05,
      "loss": 0.6586,
      "step": 894
    },
    {
      "epoch": 3.98,
      "learning_rate": 1.031431128303153e-05,
      "loss": 0.6335,
      "step": 895
    },
    {
      "epoch": 3.98,
      "learning_rate": 1.0228001028204426e-05,
      "loss": 0.6112,
      "step": 896
    },
    {
      "epoch": 3.99,
      "learning_rate": 1.0142012252160877e-05,
      "loss": 0.6517,
      "step": 897
    },
    {
      "epoch": 3.99,
      "learning_rate": 1.0056345649951881e-05,
      "loss": 0.5948,
      "step": 898
    },
    {
      "epoch": 4.0,
      "learning_rate": 9.971001914024247e-06,
      "loss": 0.6068,
      "step": 899
    },
    {
      "epoch": 4.0,
      "learning_rate": 9.885981734215094e-06,
      "loss": 0.6261,
      "step": 900
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.7309070229530334,
      "eval_runtime": 16.3022,
      "eval_samples_per_second": 4.969,
      "eval_steps_per_second": 1.288,
      "step": 900
    },
    {
      "epoch": 4.0,
      "learning_rate": 9.80128579774619e-06,
      "loss": 0.7467,
      "step": 901
    },
    {
      "epoch": 4.01,
      "learning_rate": 9.716914789218468e-06,
      "loss": 0.6475,
      "step": 902
    },
    {
      "epoch": 4.01,
      "learning_rate": 9.63286939060643e-06,
      "loss": 0.6467,
      "step": 903
    },
    {
      "epoch": 4.02,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.7299,
      "step": 904
    },
    {
      "epoch": 4.02,
      "learning_rate": 9.465758137862263e-06,
      "loss": 0.6303,
      "step": 905
    },
    {
      "epoch": 4.03,
      "learning_rate": 9.382693634497609e-06,
      "loss": 0.6799,
      "step": 906
    },
    {
      "epoch": 4.03,
      "learning_rate": 9.299957442572643e-06,
      "loss": 0.6997,
      "step": 907
    },
    {
      "epoch": 4.04,
      "learning_rate": 9.217550230847565e-06,
      "loss": 0.7221,
      "step": 908
    },
    {
      "epoch": 4.04,
      "learning_rate": 9.135472665423433e-06,
      "loss": 0.6812,
      "step": 909
    },
    {
      "epoch": 4.04,
      "learning_rate": 9.053725409736751e-06,
      "loss": 0.8037,
      "step": 910
    },
    {
      "epoch": 4.05,
      "learning_rate": 8.972309124554101e-06,
      "loss": 0.6067,
      "step": 911
    },
    {
      "epoch": 4.05,
      "learning_rate": 8.891224467966824e-06,
      "loss": 0.6613,
      "step": 912
    },
    {
      "epoch": 4.06,
      "learning_rate": 8.810472095385713e-06,
      "loss": 0.6712,
      "step": 913
    },
    {
      "epoch": 4.06,
      "learning_rate": 8.730052659535675e-06,
      "loss": 0.6991,
      "step": 914
    },
    {
      "epoch": 4.07,
      "learning_rate": 8.649966810450472e-06,
      "loss": 0.6237,
      "step": 915
    },
    {
      "epoch": 4.07,
      "learning_rate": 8.570215195467502e-06,
      "loss": 0.6039,
      "step": 916
    },
    {
      "epoch": 4.08,
      "learning_rate": 8.490798459222476e-06,
      "loss": 0.6738,
      "step": 917
    },
    {
      "epoch": 4.08,
      "learning_rate": 8.411717243644312e-06,
      "loss": 0.6461,
      "step": 918
    },
    {
      "epoch": 4.08,
      "learning_rate": 8.332972187949889e-06,
      "loss": 0.7234,
      "step": 919
    },
    {
      "epoch": 4.09,
      "learning_rate": 8.254563928638893e-06,
      "loss": 0.6206,
      "step": 920
    },
    {
      "epoch": 4.09,
      "eval_loss": 0.7307308316230774,
      "eval_runtime": 16.2336,
      "eval_samples_per_second": 4.99,
      "eval_steps_per_second": 1.294,
      "step": 920
    },
    {
      "epoch": 4.09,
      "learning_rate": 8.176493099488663e-06,
      "loss": 0.5918,
      "step": 921
    },
    {
      "epoch": 4.1,
      "learning_rate": 8.098760331549087e-06,
      "loss": 0.7177,
      "step": 922
    },
    {
      "epoch": 4.1,
      "learning_rate": 8.02136625313747e-06,
      "loss": 0.6874,
      "step": 923
    },
    {
      "epoch": 4.11,
      "learning_rate": 7.94431148983349e-06,
      "loss": 0.631,
      "step": 924
    },
    {
      "epoch": 4.11,
      "learning_rate": 7.86759666447412e-06,
      "loss": 0.6572,
      "step": 925
    },
    {
      "epoch": 4.12,
      "learning_rate": 7.791222397148612e-06,
      "loss": 0.6798,
      "step": 926
    },
    {
      "epoch": 4.12,
      "learning_rate": 7.715189305193454e-06,
      "loss": 0.6469,
      "step": 927
    },
    {
      "epoch": 4.12,
      "learning_rate": 7.639498003187417e-06,
      "loss": 0.6807,
      "step": 928
    },
    {
      "epoch": 4.13,
      "learning_rate": 7.564149102946572e-06,
      "loss": 0.6737,
      "step": 929
    },
    {
      "epoch": 4.13,
      "learning_rate": 7.489143213519301e-06,
      "loss": 0.608,
      "step": 930
    },
    {
      "epoch": 4.14,
      "learning_rate": 7.414480941181462e-06,
      "loss": 0.6284,
      "step": 931
    },
    {
      "epoch": 4.14,
      "learning_rate": 7.340162889431418e-06,
      "loss": 0.5924,
      "step": 932
    },
    {
      "epoch": 4.15,
      "learning_rate": 7.2661896589851895e-06,
      "loss": 0.6561,
      "step": 933
    },
    {
      "epoch": 4.15,
      "learning_rate": 7.192561847771589e-06,
      "loss": 0.5905,
      "step": 934
    },
    {
      "epoch": 4.16,
      "learning_rate": 7.119280050927407e-06,
      "loss": 0.6716,
      "step": 935
    },
    {
      "epoch": 4.16,
      "learning_rate": 7.046344860792525e-06,
      "loss": 0.6473,
      "step": 936
    },
    {
      "epoch": 4.16,
      "learning_rate": 6.973756866905279e-06,
      "loss": 0.6545,
      "step": 937
    },
    {
      "epoch": 4.17,
      "learning_rate": 6.901516655997536e-06,
      "loss": 0.6174,
      "step": 938
    },
    {
      "epoch": 4.17,
      "learning_rate": 6.829624811990038e-06,
      "loss": 0.636,
      "step": 939
    },
    {
      "epoch": 4.18,
      "learning_rate": 6.758081915987669e-06,
      "loss": 0.6257,
      "step": 940
    },
    {
      "epoch": 4.18,
      "eval_loss": 0.730650007724762,
      "eval_runtime": 16.243,
      "eval_samples_per_second": 4.987,
      "eval_steps_per_second": 1.293,
      "step": 940
    },
    {
      "epoch": 4.18,
      "learning_rate": 6.6868885462747276e-06,
      "loss": 0.6443,
      "step": 941
    },
    {
      "epoch": 4.19,
      "learning_rate": 6.616045278310301e-06,
      "loss": 0.672,
      "step": 942
    },
    {
      "epoch": 4.19,
      "learning_rate": 6.5455526847235825e-06,
      "loss": 0.6328,
      "step": 943
    },
    {
      "epoch": 4.2,
      "learning_rate": 6.475411335309245e-06,
      "loss": 0.5984,
      "step": 944
    },
    {
      "epoch": 4.2,
      "learning_rate": 6.405621797022848e-06,
      "loss": 0.5437,
      "step": 945
    },
    {
      "epoch": 4.2,
      "learning_rate": 6.3361846339762386e-06,
      "loss": 0.6539,
      "step": 946
    },
    {
      "epoch": 4.21,
      "learning_rate": 6.2671004074329965e-06,
      "loss": 0.6656,
      "step": 947
    },
    {
      "epoch": 4.21,
      "learning_rate": 6.1983696758039e-06,
      "loss": 0.5519,
      "step": 948
    },
    {
      "epoch": 4.22,
      "learning_rate": 6.129992994642425e-06,
      "loss": 0.6081,
      "step": 949
    },
    {
      "epoch": 4.22,
      "learning_rate": 6.061970916640236e-06,
      "loss": 0.6414,
      "step": 950
    },
    {
      "epoch": 4.23,
      "learning_rate": 5.9943039916227294e-06,
      "loss": 0.7595,
      "step": 951
    },
    {
      "epoch": 4.23,
      "learning_rate": 5.926992766544576e-06,
      "loss": 0.6684,
      "step": 952
    },
    {
      "epoch": 4.24,
      "learning_rate": 5.860037785485301e-06,
      "loss": 0.6643,
      "step": 953
    },
    {
      "epoch": 4.24,
      "learning_rate": 5.793439589644917e-06,
      "loss": 0.6311,
      "step": 954
    },
    {
      "epoch": 4.24,
      "learning_rate": 5.727198717339511e-06,
      "loss": 0.7277,
      "step": 955
    },
    {
      "epoch": 4.25,
      "learning_rate": 5.6613157039969055e-06,
      "loss": 0.6828,
      "step": 956
    },
    {
      "epoch": 4.25,
      "learning_rate": 5.595791082152352e-06,
      "loss": 0.6596,
      "step": 957
    },
    {
      "epoch": 4.26,
      "learning_rate": 5.530625381444182e-06,
      "loss": 0.6875,
      "step": 958
    },
    {
      "epoch": 4.26,
      "learning_rate": 5.4658191286095895e-06,
      "loss": 0.7416,
      "step": 959
    },
    {
      "epoch": 4.27,
      "learning_rate": 5.4013728474802846e-06,
      "loss": 0.7023,
      "step": 960
    },
    {
      "epoch": 4.27,
      "eval_loss": 0.7306858897209167,
      "eval_runtime": 16.2653,
      "eval_samples_per_second": 4.98,
      "eval_steps_per_second": 1.291,
      "step": 960
    },
    {
      "epoch": 4.27,
      "learning_rate": 5.337287058978347e-06,
      "loss": 0.6474,
      "step": 961
    },
    {
      "epoch": 4.28,
      "learning_rate": 5.2735622811119715e-06,
      "loss": 0.5983,
      "step": 962
    },
    {
      "epoch": 4.28,
      "learning_rate": 5.210199028971291e-06,
      "loss": 0.6468,
      "step": 963
    },
    {
      "epoch": 4.28,
      "learning_rate": 5.1471978147241974e-06,
      "loss": 0.7072,
      "step": 964
    },
    {
      "epoch": 4.29,
      "learning_rate": 5.0845591476122444e-06,
      "loss": 0.6677,
      "step": 965
    },
    {
      "epoch": 4.29,
      "learning_rate": 5.022283533946448e-06,
      "loss": 0.655,
      "step": 966
    },
    {
      "epoch": 4.3,
      "learning_rate": 4.960371477103304e-06,
      "loss": 0.6534,
      "step": 967
    },
    {
      "epoch": 4.3,
      "learning_rate": 4.898823477520625e-06,
      "loss": 0.5957,
      "step": 968
    },
    {
      "epoch": 4.31,
      "learning_rate": 4.837640032693558e-06,
      "loss": 0.6261,
      "step": 969
    },
    {
      "epoch": 4.31,
      "learning_rate": 4.776821637170526e-06,
      "loss": 0.6309,
      "step": 970
    },
    {
      "epoch": 4.32,
      "learning_rate": 4.7163687825492355e-06,
      "loss": 0.5786,
      "step": 971
    },
    {
      "epoch": 4.32,
      "learning_rate": 4.65628195747273e-06,
      "loss": 0.6396,
      "step": 972
    },
    {
      "epoch": 4.32,
      "learning_rate": 4.596561647625397e-06,
      "loss": 0.6884,
      "step": 973
    },
    {
      "epoch": 4.33,
      "learning_rate": 4.537208335729088e-06,
      "loss": 0.5858,
      "step": 974
    },
    {
      "epoch": 4.33,
      "learning_rate": 4.478222501539175e-06,
      "loss": 0.7217,
      "step": 975
    },
    {
      "epoch": 4.34,
      "learning_rate": 4.419604621840717e-06,
      "loss": 0.6265,
      "step": 976
    },
    {
      "epoch": 4.34,
      "learning_rate": 4.3613551704445365e-06,
      "loss": 0.5914,
      "step": 977
    },
    {
      "epoch": 4.35,
      "learning_rate": 4.303474618183484e-06,
      "loss": 0.662,
      "step": 978
    },
    {
      "epoch": 4.35,
      "learning_rate": 4.245963432908556e-06,
      "loss": 0.6491,
      "step": 979
    },
    {
      "epoch": 4.36,
      "learning_rate": 4.188822079485139e-06,
      "loss": 0.6101,
      "step": 980
    },
    {
      "epoch": 4.36,
      "eval_loss": 0.7306349277496338,
      "eval_runtime": 16.2893,
      "eval_samples_per_second": 4.973,
      "eval_steps_per_second": 1.289,
      "step": 980
    },
    {
      "epoch": 4.36,
      "learning_rate": 4.132051019789263e-06,
      "loss": 0.6338,
      "step": 981
    },
    {
      "epoch": 4.36,
      "learning_rate": 4.075650712703849e-06,
      "loss": 0.5961,
      "step": 982
    },
    {
      "epoch": 4.37,
      "learning_rate": 4.0196216141150214e-06,
      "loss": 0.622,
      "step": 983
    },
    {
      "epoch": 4.37,
      "learning_rate": 3.963964176908391e-06,
      "loss": 0.6119,
      "step": 984
    },
    {
      "epoch": 4.38,
      "learning_rate": 3.908678850965425e-06,
      "loss": 0.6194,
      "step": 985
    },
    {
      "epoch": 4.38,
      "learning_rate": 3.853766083159799e-06,
      "loss": 0.6721,
      "step": 986
    },
    {
      "epoch": 4.39,
      "learning_rate": 3.799226317353788e-06,
      "loss": 0.6138,
      "step": 987
    },
    {
      "epoch": 4.39,
      "learning_rate": 3.7450599943946728e-06,
      "loss": 0.6869,
      "step": 988
    },
    {
      "epoch": 4.4,
      "learning_rate": 3.691267552111183e-06,
      "loss": 0.6511,
      "step": 989
    },
    {
      "epoch": 4.4,
      "learning_rate": 3.6378494253099304e-06,
      "loss": 0.6247,
      "step": 990
    },
    {
      "epoch": 4.4,
      "learning_rate": 3.5848060457719547e-06,
      "loss": 0.6228,
      "step": 991
    },
    {
      "epoch": 4.41,
      "learning_rate": 3.532137842249178e-06,
      "loss": 0.5655,
      "step": 992
    },
    {
      "epoch": 4.41,
      "learning_rate": 3.4798452404609603e-06,
      "loss": 0.5822,
      "step": 993
    },
    {
      "epoch": 4.42,
      "learning_rate": 3.4279286630906572e-06,
      "loss": 0.6357,
      "step": 994
    },
    {
      "epoch": 4.42,
      "learning_rate": 3.376388529782215e-06,
      "loss": 0.6089,
      "step": 995
    },
    {
      "epoch": 4.43,
      "learning_rate": 3.325225257136738e-06,
      "loss": 0.6363,
      "step": 996
    },
    {
      "epoch": 4.43,
      "learning_rate": 3.2744392587091722e-06,
      "loss": 0.5946,
      "step": 997
    },
    {
      "epoch": 4.44,
      "learning_rate": 3.2240309450049355e-06,
      "loss": 0.5804,
      "step": 998
    },
    {
      "epoch": 4.44,
      "learning_rate": 3.1740007234766002e-06,
      "loss": 0.6326,
      "step": 999
    },
    {
      "epoch": 4.44,
      "learning_rate": 3.1243489985206097e-06,
      "loss": 0.6465,
      "step": 1000
    },
    {
      "epoch": 4.44,
      "eval_loss": 0.7306743860244751,
      "eval_runtime": 16.3111,
      "eval_samples_per_second": 4.966,
      "eval_steps_per_second": 1.287,
      "step": 1000
    },
    {
      "epoch": 4.45,
      "learning_rate": 3.075076171473995e-06,
      "loss": 0.7062,
      "step": 1001
    },
    {
      "epoch": 4.45,
      "learning_rate": 3.0261826406111525e-06,
      "loss": 0.7212,
      "step": 1002
    },
    {
      "epoch": 4.46,
      "learning_rate": 2.9776688011405817e-06,
      "loss": 0.7193,
      "step": 1003
    },
    {
      "epoch": 4.46,
      "learning_rate": 2.9295350452017535e-06,
      "loss": 0.7064,
      "step": 1004
    },
    {
      "epoch": 4.47,
      "learning_rate": 2.8817817618618847e-06,
      "loss": 0.7188,
      "step": 1005
    },
    {
      "epoch": 4.47,
      "learning_rate": 2.8344093371128424e-06,
      "loss": 0.7245,
      "step": 1006
    },
    {
      "epoch": 4.48,
      "learning_rate": 2.787418153867971e-06,
      "loss": 0.7176,
      "step": 1007
    },
    {
      "epoch": 4.48,
      "learning_rate": 2.7408085919590264e-06,
      "loss": 0.7244,
      "step": 1008
    },
    {
      "epoch": 4.48,
      "learning_rate": 2.6945810281331085e-06,
      "loss": 0.6673,
      "step": 1009
    },
    {
      "epoch": 4.49,
      "learning_rate": 2.648735836049615e-06,
      "loss": 0.6861,
      "step": 1010
    },
    {
      "epoch": 4.49,
      "learning_rate": 2.6032733862772105e-06,
      "loss": 0.6591,
      "step": 1011
    },
    {
      "epoch": 4.5,
      "learning_rate": 2.558194046290835e-06,
      "loss": 0.6897,
      "step": 1012
    },
    {
      "epoch": 4.5,
      "learning_rate": 2.5134981804687484e-06,
      "loss": 0.6069,
      "step": 1013
    },
    {
      "epoch": 4.51,
      "learning_rate": 2.4691861500895474e-06,
      "loss": 0.6301,
      "step": 1014
    },
    {
      "epoch": 4.51,
      "learning_rate": 2.4252583133292927e-06,
      "loss": 0.6238,
      "step": 1015
    },
    {
      "epoch": 4.52,
      "learning_rate": 2.381715025258585e-06,
      "loss": 0.6159,
      "step": 1016
    },
    {
      "epoch": 4.52,
      "learning_rate": 2.3385566378397007e-06,
      "loss": 0.6171,
      "step": 1017
    },
    {
      "epoch": 4.52,
      "learning_rate": 2.2957834999237424e-06,
      "loss": 0.5524,
      "step": 1018
    },
    {
      "epoch": 4.53,
      "learning_rate": 2.2533959572478392e-06,
      "loss": 0.6197,
      "step": 1019
    },
    {
      "epoch": 4.53,
      "learning_rate": 2.2113943524323167e-06,
      "loss": 0.7135,
      "step": 1020
    },
    {
      "epoch": 4.53,
      "eval_loss": 0.730741024017334,
      "eval_runtime": 16.3211,
      "eval_samples_per_second": 4.963,
      "eval_steps_per_second": 1.287,
      "step": 1020
    },
    {
      "epoch": 4.54,
      "learning_rate": 2.1697790249779636e-06,
      "loss": 0.6207,
      "step": 1021
    },
    {
      "epoch": 4.54,
      "learning_rate": 2.128550311263261e-06,
      "loss": 0.6479,
      "step": 1022
    },
    {
      "epoch": 4.55,
      "learning_rate": 2.087708544541689e-06,
      "loss": 0.6496,
      "step": 1023
    },
    {
      "epoch": 4.55,
      "learning_rate": 2.0472540549390074e-06,
      "loss": 0.6942,
      "step": 1024
    },
    {
      "epoch": 4.56,
      "learning_rate": 2.007187169450603e-06,
      "loss": 0.7154,
      "step": 1025
    },
    {
      "epoch": 4.56,
      "learning_rate": 1.9675082119388344e-06,
      "loss": 0.6424,
      "step": 1026
    },
    {
      "epoch": 4.56,
      "learning_rate": 1.9282175031304305e-06,
      "loss": 0.6996,
      "step": 1027
    },
    {
      "epoch": 4.57,
      "learning_rate": 1.8893153606138802e-06,
      "loss": 0.633,
      "step": 1028
    },
    {
      "epoch": 4.57,
      "learning_rate": 1.85080209883689e-06,
      "loss": 0.6168,
      "step": 1029
    },
    {
      "epoch": 4.58,
      "learning_rate": 1.8126780291038037e-06,
      "loss": 0.5781,
      "step": 1030
    },
    {
      "epoch": 4.58,
      "learning_rate": 1.7749434595731363e-06,
      "loss": 0.663,
      "step": 1031
    },
    {
      "epoch": 4.59,
      "learning_rate": 1.7375986952550326e-06,
      "loss": 0.655,
      "step": 1032
    },
    {
      "epoch": 4.59,
      "learning_rate": 1.7006440380088362e-06,
      "loss": 0.6346,
      "step": 1033
    },
    {
      "epoch": 4.6,
      "learning_rate": 1.6640797865406288e-06,
      "loss": 0.6138,
      "step": 1034
    },
    {
      "epoch": 4.6,
      "learning_rate": 1.6279062364008447e-06,
      "loss": 0.5901,
      "step": 1035
    },
    {
      "epoch": 4.6,
      "learning_rate": 1.5921236799818384e-06,
      "loss": 0.5759,
      "step": 1036
    },
    {
      "epoch": 4.61,
      "learning_rate": 1.5567324065155653e-06,
      "loss": 0.6452,
      "step": 1037
    },
    {
      "epoch": 4.61,
      "learning_rate": 1.5217327020712158e-06,
      "loss": 0.5874,
      "step": 1038
    },
    {
      "epoch": 4.62,
      "learning_rate": 1.4871248495529011e-06,
      "loss": 0.6349,
      "step": 1039
    },
    {
      "epoch": 4.62,
      "learning_rate": 1.4529091286973995e-06,
      "loss": 0.6658,
      "step": 1040
    },
    {
      "epoch": 4.62,
      "eval_loss": 0.7307910919189453,
      "eval_runtime": 16.3086,
      "eval_samples_per_second": 4.967,
      "eval_steps_per_second": 1.288,
      "step": 1040
    },
    {
      "epoch": 4.63,
      "learning_rate": 1.4190858160718467e-06,
      "loss": 0.6129,
      "step": 1041
    },
    {
      "epoch": 4.63,
      "learning_rate": 1.3856551850715494e-06,
      "loss": 0.6186,
      "step": 1042
    },
    {
      "epoch": 4.64,
      "learning_rate": 1.352617505917736e-06,
      "loss": 0.6641,
      "step": 1043
    },
    {
      "epoch": 4.64,
      "learning_rate": 1.3199730456553926e-06,
      "loss": 0.5646,
      "step": 1044
    },
    {
      "epoch": 4.64,
      "learning_rate": 1.2877220681510927e-06,
      "loss": 0.5679,
      "step": 1045
    },
    {
      "epoch": 4.65,
      "learning_rate": 1.2558648340908862e-06,
      "loss": 0.6184,
      "step": 1046
    },
    {
      "epoch": 4.65,
      "learning_rate": 1.2244016009781701e-06,
      "loss": 0.6944,
      "step": 1047
    },
    {
      "epoch": 4.66,
      "learning_rate": 1.19333262313161e-06,
      "loss": 0.5864,
      "step": 1048
    },
    {
      "epoch": 4.66,
      "learning_rate": 1.1626581516831048e-06,
      "loss": 0.5279,
      "step": 1049
    },
    {
      "epoch": 4.67,
      "learning_rate": 1.1323784345757205e-06,
      "loss": 0.6324,
      "step": 1050
    },
    {
      "epoch": 4.67,
      "learning_rate": 1.1024937165617145e-06,
      "loss": 0.7198,
      "step": 1051
    },
    {
      "epoch": 4.68,
      "learning_rate": 1.0730042392005647e-06,
      "loss": 0.7036,
      "step": 1052
    },
    {
      "epoch": 4.68,
      "learning_rate": 1.0439102408569769e-06,
      "loss": 0.73,
      "step": 1053
    },
    {
      "epoch": 4.68,
      "learning_rate": 1.0152119566990025e-06,
      "loss": 0.7607,
      "step": 1054
    },
    {
      "epoch": 4.69,
      "learning_rate": 9.869096186961025e-07,
      "loss": 0.7188,
      "step": 1055
    },
    {
      "epoch": 4.69,
      "learning_rate": 9.590034556173079e-07,
      "loss": 0.7229,
      "step": 1056
    },
    {
      "epoch": 4.7,
      "learning_rate": 9.314936930293283e-07,
      "loss": 0.7239,
      "step": 1057
    },
    {
      "epoch": 4.7,
      "learning_rate": 9.043805532947647e-07,
      "loss": 0.664,
      "step": 1058
    },
    {
      "epoch": 4.71,
      "learning_rate": 8.776642555702985e-07,
      "loss": 0.7025,
      "step": 1059
    },
    {
      "epoch": 4.71,
      "learning_rate": 8.513450158049108e-07,
      "loss": 0.6855,
      "step": 1060
    },
    {
      "epoch": 4.71,
      "eval_loss": 0.7306917309761047,
      "eval_runtime": 16.3221,
      "eval_samples_per_second": 4.963,
      "eval_steps_per_second": 1.287,
      "step": 1060
    },
    {
      "epoch": 4.72,
      "learning_rate": 8.254230467381551e-07,
      "loss": 0.6641,
      "step": 1061
    },
    {
      "epoch": 4.72,
      "learning_rate": 7.99898557898432e-07,
      "loss": 0.7383,
      "step": 1062
    },
    {
      "epoch": 4.72,
      "learning_rate": 7.747717556012779e-07,
      "loss": 0.6531,
      "step": 1063
    },
    {
      "epoch": 4.73,
      "learning_rate": 7.50042842947718e-07,
      "loss": 0.6781,
      "step": 1064
    },
    {
      "epoch": 4.73,
      "learning_rate": 7.25712019822622e-07,
      "loss": 0.6386,
      "step": 1065
    },
    {
      "epoch": 4.74,
      "learning_rate": 7.01779482893089e-07,
      "loss": 0.6938,
      "step": 1066
    },
    {
      "epoch": 4.74,
      "learning_rate": 6.782454256068438e-07,
      "loss": 0.7193,
      "step": 1067
    },
    {
      "epoch": 4.75,
      "learning_rate": 6.551100381906872e-07,
      "loss": 0.5737,
      "step": 1068
    },
    {
      "epoch": 4.75,
      "learning_rate": 6.323735076489535e-07,
      "loss": 0.6546,
      "step": 1069
    },
    {
      "epoch": 4.76,
      "learning_rate": 6.100360177619946e-07,
      "loss": 0.6558,
      "step": 1070
    },
    {
      "epoch": 4.76,
      "learning_rate": 5.880977490847151e-07,
      "loss": 0.5834,
      "step": 1071
    },
    {
      "epoch": 4.76,
      "learning_rate": 5.665588789450782e-07,
      "loss": 0.6137,
      "step": 1072
    },
    {
      "epoch": 4.77,
      "learning_rate": 5.454195814427021e-07,
      "loss": 0.7225,
      "step": 1073
    },
    {
      "epoch": 4.77,
      "learning_rate": 5.246800274474439e-07,
      "loss": 0.6166,
      "step": 1074
    },
    {
      "epoch": 4.78,
      "learning_rate": 5.043403845980122e-07,
      "loss": 0.6703,
      "step": 1075
    },
    {
      "epoch": 4.78,
      "learning_rate": 4.844008173006232e-07,
      "loss": 0.5972,
      "step": 1076
    },
    {
      "epoch": 4.79,
      "learning_rate": 4.6486148672764687e-07,
      "loss": 0.6436,
      "step": 1077
    },
    {
      "epoch": 4.79,
      "learning_rate": 4.457225508163465e-07,
      "loss": 0.6859,
      "step": 1078
    },
    {
      "epoch": 4.8,
      "learning_rate": 4.269841642675576e-07,
      "loss": 0.6171,
      "step": 1079
    },
    {
      "epoch": 4.8,
      "learning_rate": 4.086464785444777e-07,
      "loss": 0.5769,
      "step": 1080
    },
    {
      "epoch": 4.8,
      "eval_loss": 0.7307865023612976,
      "eval_runtime": 16.328,
      "eval_samples_per_second": 4.961,
      "eval_steps_per_second": 1.286,
      "step": 1080
    },
    {
      "epoch": 4.8,
      "learning_rate": 3.907096418714062e-07,
      "loss": 0.6674,
      "step": 1081
    },
    {
      "epoch": 4.81,
      "learning_rate": 3.7317379923256233e-07,
      "loss": 0.7353,
      "step": 1082
    },
    {
      "epoch": 4.81,
      "learning_rate": 3.5603909237092445e-07,
      "loss": 0.7192,
      "step": 1083
    },
    {
      "epoch": 4.82,
      "learning_rate": 3.393056597870703e-07,
      "loss": 0.7237,
      "step": 1084
    },
    {
      "epoch": 4.82,
      "learning_rate": 3.229736367380498e-07,
      "loss": 0.7342,
      "step": 1085
    },
    {
      "epoch": 4.83,
      "learning_rate": 3.0704315523631953e-07,
      "loss": 0.6708,
      "step": 1086
    },
    {
      "epoch": 4.83,
      "learning_rate": 2.915143440486379e-07,
      "loss": 0.5742,
      "step": 1087
    },
    {
      "epoch": 4.84,
      "learning_rate": 2.7638732869506025e-07,
      "loss": 0.5832,
      "step": 1088
    },
    {
      "epoch": 4.84,
      "learning_rate": 2.616622314479067e-07,
      "loss": 0.5766,
      "step": 1089
    },
    {
      "epoch": 4.84,
      "learning_rate": 2.473391713307738e-07,
      "loss": 0.6198,
      "step": 1090
    },
    {
      "epoch": 4.85,
      "learning_rate": 2.334182641175686e-07,
      "loss": 0.6962,
      "step": 1091
    },
    {
      "epoch": 4.85,
      "learning_rate": 2.19899622331593e-07,
      "loss": 0.596,
      "step": 1092
    },
    {
      "epoch": 4.86,
      "learning_rate": 2.0678335524460524e-07,
      "loss": 0.6843,
      "step": 1093
    },
    {
      "epoch": 4.86,
      "learning_rate": 1.9406956887595418e-07,
      "loss": 0.6217,
      "step": 1094
    },
    {
      "epoch": 4.87,
      "learning_rate": 1.8175836599173546e-07,
      "loss": 0.6389,
      "step": 1095
    },
    {
      "epoch": 4.87,
      "learning_rate": 1.6984984610392552e-07,
      "loss": 0.6922,
      "step": 1096
    },
    {
      "epoch": 4.88,
      "learning_rate": 1.5834410546960443e-07,
      "loss": 0.649,
      "step": 1097
    },
    {
      "epoch": 4.88,
      "learning_rate": 1.4724123709017878e-07,
      "loss": 0.6382,
      "step": 1098
    },
    {
      "epoch": 4.88,
      "learning_rate": 1.3654133071059893e-07,
      "loss": 0.6124,
      "step": 1099
    },
    {
      "epoch": 4.89,
      "learning_rate": 1.2624447281867624e-07,
      "loss": 0.5613,
      "step": 1100
    },
    {
      "epoch": 4.89,
      "eval_loss": 0.7309412956237793,
      "eval_runtime": 16.3302,
      "eval_samples_per_second": 4.96,
      "eval_steps_per_second": 1.286,
      "step": 1100
    },
    {
      "epoch": 4.89,
      "learning_rate": 1.1635074664436141e-07,
      "loss": 0.7437,
      "step": 1101
    },
    {
      "epoch": 4.9,
      "learning_rate": 1.0686023215906171e-07,
      "loss": 0.6838,
      "step": 1102
    },
    {
      "epoch": 4.9,
      "learning_rate": 9.777300607501927e-08,
      "loss": 0.7432,
      "step": 1103
    },
    {
      "epoch": 4.91,
      "learning_rate": 8.90891418446782e-08,
      "loss": 0.6784,
      "step": 1104
    },
    {
      "epoch": 4.91,
      "learning_rate": 8.080870966008514e-08,
      "loss": 0.7191,
      "step": 1105
    },
    {
      "epoch": 4.92,
      "learning_rate": 7.293177645232297e-08,
      "loss": 0.7513,
      "step": 1106
    },
    {
      "epoch": 4.92,
      "learning_rate": 6.545840589099461e-08,
      "loss": 0.673,
      "step": 1107
    },
    {
      "epoch": 4.92,
      "learning_rate": 5.838865838366792e-08,
      "loss": 0.6339,
      "step": 1108
    },
    {
      "epoch": 4.93,
      "learning_rate": 5.1722591075425986e-08,
      "loss": 0.7272,
      "step": 1109
    },
    {
      "epoch": 4.93,
      "learning_rate": 4.5460257848373156e-08,
      "loss": 0.7292,
      "step": 1110
    },
    {
      "epoch": 4.94,
      "learning_rate": 3.96017093212131e-08,
      "loss": 0.6564,
      "step": 1111
    },
    {
      "epoch": 4.94,
      "learning_rate": 3.4146992848854695e-08,
      "loss": 0.6613,
      "step": 1112
    },
    {
      "epoch": 4.95,
      "learning_rate": 2.9096152522006815e-08,
      "loss": 0.6325,
      "step": 1113
    },
    {
      "epoch": 4.95,
      "learning_rate": 2.4449229166823016e-08,
      "loss": 0.6954,
      "step": 1114
    },
    {
      "epoch": 4.96,
      "learning_rate": 2.020626034459072e-08,
      "loss": 0.7139,
      "step": 1115
    },
    {
      "epoch": 4.96,
      "learning_rate": 1.636728035140922e-08,
      "loss": 0.6613,
      "step": 1116
    },
    {
      "epoch": 4.96,
      "learning_rate": 1.2932320217917682e-08,
      "loss": 0.6576,
      "step": 1117
    },
    {
      "epoch": 4.97,
      "learning_rate": 9.901407709050902e-09,
      "loss": 0.6583,
      "step": 1118
    },
    {
      "epoch": 4.97,
      "learning_rate": 7.2745673238006076e-09,
      "loss": 0.6937,
      "step": 1119
    },
    {
      "epoch": 4.98,
      "learning_rate": 5.051820295032261e-09,
      "loss": 0.6591,
      "step": 1120
    },
    {
      "epoch": 4.98,
      "eval_loss": 0.7308295965194702,
      "eval_runtime": 16.298,
      "eval_samples_per_second": 4.97,
      "eval_steps_per_second": 1.288,
      "step": 1120
    },
    {
      "epoch": 4.98,
      "learning_rate": 3.2331845893074366e-09,
      "loss": 0.7024,
      "step": 1121
    },
    {
      "epoch": 4.99,
      "learning_rate": 1.8186749067339303e-09,
      "loss": 0.6562,
      "step": 1122
    },
    {
      "epoch": 4.99,
      "learning_rate": 8.083026808602956e-10,
      "loss": 0.5939,
      "step": 1123
    },
    {
      "epoch": 5.0,
      "learning_rate": 2.020760785648168e-10,
      "loss": 0.6105,
      "step": 1124
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0,
      "loss": 0.6582,
      "step": 1125
    }
  ],
  "logging_steps": 1,
  "max_steps": 1125,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 20,
  "total_flos": 2.6726410071834624e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}