{
  "best_metric": 0.7217394709587097,
  "best_model_checkpoint": "./open-long-orca-13b/checkpoint-4096",
  "epoch": 2.8845070422535213,
  "eval_steps": 64,
  "global_step": 4096,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.66015625e-08,
      "loss": 1.216,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 6.640625e-08,
      "loss": 1.0668,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.494140625e-07,
      "loss": 1.0028,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 2.65625e-07,
      "loss": 0.9744,
      "step": 4
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.150390625e-07,
      "loss": 0.9335,
      "step": 5
    },
    {
      "epoch": 0.0,
      "learning_rate": 5.9765625e-07,
      "loss": 0.9319,
      "step": 6
    },
    {
      "epoch": 0.0,
      "learning_rate": 8.134765625e-07,
      "loss": 0.9267,
      "step": 7
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.0625e-06,
      "loss": 0.9021,
      "step": 8
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.3447265625e-06,
      "loss": 0.9244,
      "step": 9
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.66015625e-06,
      "loss": 0.8741,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 2.0087890625e-06,
      "loss": 0.8839,
      "step": 11
    },
    {
      "epoch": 0.01,
      "learning_rate": 2.390625e-06,
      "loss": 0.8737,
      "step": 12
    },
    {
      "epoch": 0.01,
      "learning_rate": 2.8056640625e-06,
      "loss": 0.8887,
      "step": 13
    },
    {
      "epoch": 0.01,
      "learning_rate": 3.25390625e-06,
      "loss": 0.9123,
      "step": 14
    },
    {
      "epoch": 0.01,
      "learning_rate": 3.7353515625e-06,
      "loss": 0.8705,
      "step": 15
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.25e-06,
      "loss": 0.8738,
      "step": 16
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.7978515625e-06,
      "loss": 0.9001,
      "step": 17
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.37890625e-06,
      "loss": 0.8693,
      "step": 18
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.9931640625e-06,
      "loss": 0.8689,
      "step": 19
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.640625e-06,
      "loss": 0.8399,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 7.3212890625e-06,
      "loss": 0.8806,
      "step": 21
    },
    {
      "epoch": 0.02,
      "learning_rate": 8.03515625e-06,
      "loss": 0.8683,
      "step": 22
    },
    {
      "epoch": 0.02,
      "learning_rate": 8.7822265625e-06,
      "loss": 0.8444,
      "step": 23
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.5625e-06,
      "loss": 0.8619,
      "step": 24
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.03759765625e-05,
      "loss": 0.8647,
      "step": 25
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.122265625e-05,
      "loss": 0.8665,
      "step": 26
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.21025390625e-05,
      "loss": 0.8415,
      "step": 27
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.3015625e-05,
      "loss": 0.8239,
      "step": 28
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.39619140625e-05,
      "loss": 0.8758,
      "step": 29
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.494140625e-05,
      "loss": 0.8213,
      "step": 30
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.59541015625e-05,
      "loss": 0.8685,
      "step": 31
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.7e-05,
      "loss": 0.8758,
      "step": 32
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.699999868508013e-05,
      "loss": 0.8605,
      "step": 33
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.6999994740320928e-05,
      "loss": 0.8425,
      "step": 34
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.699998816572361e-05,
      "loss": 0.838,
      "step": 35
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6999978961290217e-05,
      "loss": 0.8427,
      "step": 36
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6999967127023588e-05,
      "loss": 0.8792,
      "step": 37
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6999952662927394e-05,
      "loss": 0.8509,
      "step": 38
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6999935569006104e-05,
      "loss": 0.8594,
      "step": 39
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.699991584526501e-05,
      "loss": 0.8562,
      "step": 40
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.699989349171021e-05,
      "loss": 0.8739,
      "step": 41
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6999868508348625e-05,
      "loss": 0.8483,
      "step": 42
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.699984089518798e-05,
      "loss": 0.8666,
      "step": 43
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6999810652236823e-05,
      "loss": 0.877,
      "step": 44
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.699977777950451e-05,
      "loss": 0.8618,
      "step": 45
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6999742277001208e-05,
      "loss": 0.8322,
      "step": 46
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.69997041447379e-05,
      "loss": 0.855,
      "step": 47
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6999663382726394e-05,
      "loss": 0.838,
      "step": 48
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.699961999097929e-05,
      "loss": 0.8335,
      "step": 49
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.699957396951002e-05,
      "loss": 0.829,
      "step": 50
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.699952531833282e-05,
      "loss": 0.8028,
      "step": 51
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.6999474037462744e-05,
      "loss": 0.8335,
      "step": 52
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.6999420126915656e-05,
      "loss": 0.8372,
      "step": 53
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.6999363586708237e-05,
      "loss": 0.8146,
      "step": 54
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.699930441685798e-05,
      "loss": 0.8233,
      "step": 55
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.699924261738319e-05,
      "loss": 0.8361,
      "step": 56
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.699917818830299e-05,
      "loss": 0.811,
      "step": 57
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.6999111129637313e-05,
      "loss": 0.8238,
      "step": 58
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.6999041441406904e-05,
      "loss": 0.8621,
      "step": 59
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.699896912363333e-05,
      "loss": 0.8619,
      "step": 60
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.6998894176338956e-05,
      "loss": 0.8156,
      "step": 61
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.699881659954698e-05,
      "loss": 0.8674,
      "step": 62
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.6998736393281398e-05,
      "loss": 0.8227,
      "step": 63
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.699865355756703e-05,
      "loss": 0.8377,
      "step": 64
    },
    {
      "epoch": 0.05,
      "eval_loss": 0.8097721934318542,
      "eval_runtime": 30.8354,
      "eval_samples_per_second": 265.636,
      "eval_steps_per_second": 16.604,
      "step": 64
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.6998568092429495e-05,
      "loss": 0.8323,
      "step": 65
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.6998479997895247e-05,
      "loss": 0.8272,
      "step": 66
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.6998389273991536e-05,
      "loss": 0.843,
      "step": 67
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.699829592074643e-05,
      "loss": 0.8196,
      "step": 68
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.6998199938188817e-05,
      "loss": 0.8393,
      "step": 69
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.6998101326348386e-05,
      "loss": 0.8088,
      "step": 70
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.6998000085255654e-05,
      "loss": 0.8245,
      "step": 71
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.699789621494194e-05,
      "loss": 0.8397,
      "step": 72
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.6997789715439384e-05,
      "loss": 0.8008,
      "step": 73
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.6997680586780933e-05,
      "loss": 0.8405,
      "step": 74
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.699756882900035e-05,
      "loss": 0.8773,
      "step": 75
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.6997454442132217e-05,
      "loss": 0.8315,
      "step": 76
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.699733742621192e-05,
      "loss": 0.8245,
      "step": 77
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.6997217781275666e-05,
      "loss": 0.8439,
      "step": 78
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.6997095507360467e-05,
      "loss": 0.8516,
      "step": 79
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.699697060450416e-05,
      "loss": 0.8367,
      "step": 80
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.6996843072745386e-05,
      "loss": 0.8089,
      "step": 81
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.6996712912123603e-05,
      "loss": 0.8437,
      "step": 82
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.699658012267908e-05,
      "loss": 0.799,
      "step": 83
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.6996444704452903e-05,
      "loss": 0.843,
      "step": 84
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.6996306657486972e-05,
      "loss": 0.8081,
      "step": 85
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.699616598182399e-05,
      "loss": 0.8041,
      "step": 86
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.699602267750749e-05,
      "loss": 0.8223,
      "step": 87
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.6995876744581804e-05,
      "loss": 0.8581,
      "step": 88
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.699572818309208e-05,
      "loss": 0.8252,
      "step": 89
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.6995576993084292e-05,
      "loss": 0.8241,
      "step": 90
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.699542317460521e-05,
      "loss": 0.8242,
      "step": 91
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.699526672770242e-05,
      "loss": 0.8291,
      "step": 92
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.6995107652424334e-05,
      "loss": 0.813,
      "step": 93
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.6994945948820167e-05,
      "loss": 0.8593,
      "step": 94
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.6994781616939947e-05,
      "loss": 0.89,
      "step": 95
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.6994614656834514e-05,
      "loss": 0.786,
      "step": 96
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.699444506855553e-05,
      "loss": 0.8135,
      "step": 97
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.6994272852155465e-05,
      "loss": 0.8374,
      "step": 98
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.6994098007687595e-05,
      "loss": 0.8466,
      "step": 99
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.699392053520602e-05,
      "loss": 0.8225,
      "step": 100
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.699374043476565e-05,
      "loss": 0.8558,
      "step": 101
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.6993557706422203e-05,
      "loss": 0.8618,
      "step": 102
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.6993372350232215e-05,
      "loss": 0.8073,
      "step": 103
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.699318436625304e-05,
      "loss": 0.8552,
      "step": 104
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.6992993754542827e-05,
      "loss": 0.8,
      "step": 105
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.699280051516056e-05,
      "loss": 0.8037,
      "step": 106
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.699260464816602e-05,
      "loss": 0.8855,
      "step": 107
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.6992406153619813e-05,
      "loss": 0.8277,
      "step": 108
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.6992205031583348e-05,
      "loss": 0.8386,
      "step": 109
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.699200128211885e-05,
      "loss": 0.8042,
      "step": 110
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.6991794905289355e-05,
      "loss": 0.8367,
      "step": 111
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.6991585901158723e-05,
      "loss": 0.8261,
      "step": 112
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.6991374269791612e-05,
      "loss": 0.8202,
      "step": 113
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.6991160011253498e-05,
      "loss": 0.8063,
      "step": 114
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.6990943125610674e-05,
      "loss": 0.8268,
      "step": 115
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.6990723612930245e-05,
      "loss": 0.8307,
      "step": 116
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.6990501473280122e-05,
      "loss": 0.8151,
      "step": 117
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.6990276706729038e-05,
      "loss": 0.8078,
      "step": 118
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.699004931334653e-05,
      "loss": 0.8151,
      "step": 119
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.6989819293202955e-05,
      "loss": 0.8197,
      "step": 120
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.698958664636948e-05,
      "loss": 0.8272,
      "step": 121
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.6989351372918076e-05,
      "loss": 0.8452,
      "step": 122
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.6989113472921546e-05,
      "loss": 0.8143,
      "step": 123
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.698887294645349e-05,
      "loss": 0.7984,
      "step": 124
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.6988629793588324e-05,
      "loss": 0.881,
      "step": 125
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.6988384014401282e-05,
      "loss": 0.7801,
      "step": 126
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.69881356089684e-05,
      "loss": 0.8244,
      "step": 127
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.698788457736654e-05,
      "loss": 0.7886,
      "step": 128
    },
    {
      "epoch": 0.09,
      "eval_loss": 0.7922429442405701,
      "eval_runtime": 30.8621,
      "eval_samples_per_second": 265.406,
      "eval_steps_per_second": 16.59,
      "step": 128
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.6987630919673366e-05,
      "loss": 0.791,
      "step": 129
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.6987374635967355e-05,
      "loss": 0.782,
      "step": 130
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.6987115726327804e-05,
      "loss": 0.8234,
      "step": 131
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.698685419083481e-05,
      "loss": 0.8186,
      "step": 132
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.6986590029569304e-05,
      "loss": 0.7954,
      "step": 133
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.6986323242613e-05,
      "loss": 0.8032,
      "step": 134
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.6986053830048454e-05,
      "loss": 0.8198,
      "step": 135
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.698578179195901e-05,
      "loss": 0.7948,
      "step": 136
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.698550712842884e-05,
      "loss": 0.8733,
      "step": 137
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.6985229839542924e-05,
      "loss": 0.8047,
      "step": 138
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.6984949925387047e-05,
      "loss": 0.8023,
      "step": 139
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.698466738604782e-05,
      "loss": 0.8132,
      "step": 140
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.6984382221612652e-05,
      "loss": 0.7971,
      "step": 141
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.698409443216977e-05,
      "loss": 0.7974,
      "step": 142
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.6983804017808227e-05,
      "loss": 0.8699,
      "step": 143
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.698351097861786e-05,
      "loss": 0.8462,
      "step": 144
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.698321531468934e-05,
      "loss": 0.8091,
      "step": 145
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.6982917026114144e-05,
      "loss": 0.8224,
      "step": 146
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.6982616112984555e-05,
      "loss": 0.8063,
      "step": 147
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.698231257539368e-05,
      "loss": 0.8253,
      "step": 148
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.698200641343543e-05,
      "loss": 0.8057,
      "step": 149
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.6981697627204525e-05,
      "loss": 0.8186,
      "step": 150
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.6981386216796505e-05,
      "loss": 0.8029,
      "step": 151
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.698107218230772e-05,
      "loss": 0.811,
      "step": 152
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.6980755523835324e-05,
      "loss": 0.8182,
      "step": 153
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.6980436241477295e-05,
      "loss": 0.8047,
      "step": 154
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.6980114335332414e-05,
      "loss": 0.818,
      "step": 155
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.6979789805500272e-05,
      "loss": 0.7974,
      "step": 156
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.6979462652081286e-05,
      "loss": 0.7906,
      "step": 157
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.6979132875176666e-05,
      "loss": 0.8333,
      "step": 158
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.697880047488845e-05,
      "loss": 0.7958,
      "step": 159
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.6978465451319474e-05,
      "loss": 0.791,
      "step": 160
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.6978127804573394e-05,
      "loss": 0.8087,
      "step": 161
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.6977787534754678e-05,
      "loss": 0.7994,
      "step": 162
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.69774446419686e-05,
      "loss": 0.7964,
      "step": 163
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.6977099126321253e-05,
      "loss": 0.8083,
      "step": 164
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.697675098791953e-05,
      "loss": 0.8313,
      "step": 165
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.6976400226871147e-05,
      "loss": 0.8234,
      "step": 166
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.6976046843284627e-05,
      "loss": 0.8608,
      "step": 167
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.6975690837269304e-05,
      "loss": 0.7633,
      "step": 168
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.6975332208935324e-05,
      "loss": 0.7978,
      "step": 169
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.697497095839364e-05,
      "loss": 0.8473,
      "step": 170
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.6974607085756025e-05,
      "loss": 0.8098,
      "step": 171
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.697424059113506e-05,
      "loss": 0.8225,
      "step": 172
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.6973871474644132e-05,
      "loss": 0.8077,
      "step": 173
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.6973499736397444e-05,
      "loss": 0.8143,
      "step": 174
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.697312537651001e-05,
      "loss": 0.8154,
      "step": 175
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.6972748395097657e-05,
      "loss": 0.8136,
      "step": 176
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.6972368792277013e-05,
      "loss": 0.8283,
      "step": 177
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.6971986568165528e-05,
      "loss": 0.8048,
      "step": 178
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.6971601722881463e-05,
      "loss": 0.836,
      "step": 179
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.697121425654388e-05,
      "loss": 0.7826,
      "step": 180
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.6970824169272666e-05,
      "loss": 0.8149,
      "step": 181
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.6970431461188505e-05,
      "loss": 0.8262,
      "step": 182
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.69700361324129e-05,
      "loss": 0.7936,
      "step": 183
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.6969638183068165e-05,
      "loss": 0.8045,
      "step": 184
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.696923761327742e-05,
      "loss": 0.8283,
      "step": 185
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.6968834423164595e-05,
      "loss": 0.8239,
      "step": 186
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.696842861285444e-05,
      "loss": 0.8348,
      "step": 187
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.6968020182472512e-05,
      "loss": 0.8058,
      "step": 188
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.696760913214517e-05,
      "loss": 0.815,
      "step": 189
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.6967195461999595e-05,
      "loss": 0.7741,
      "step": 190
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.6966779172163767e-05,
      "loss": 0.8009,
      "step": 191
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.696636026276649e-05,
      "loss": 0.8107,
      "step": 192
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.7826950550079346,
      "eval_runtime": 30.8432,
      "eval_samples_per_second": 265.569,
      "eval_steps_per_second": 16.6,
      "step": 192
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.696593873393737e-05,
      "loss": 0.8324,
      "step": 193
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.6965514585806825e-05,
      "loss": 0.8162,
      "step": 194
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.696508781850608e-05,
      "loss": 0.8235,
      "step": 195
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.6964658432167176e-05,
      "loss": 0.8163,
      "step": 196
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.6964226426922963e-05,
      "loss": 0.8419,
      "step": 197
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.69637918029071e-05,
      "loss": 0.8075,
      "step": 198
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.6963354560254054e-05,
      "loss": 0.823,
      "step": 199
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.6962914699099112e-05,
      "loss": 0.7444,
      "step": 200
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.6962472219578356e-05,
      "loss": 0.7704,
      "step": 201
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.696202712182869e-05,
      "loss": 0.8065,
      "step": 202
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.6961579405987824e-05,
      "loss": 0.7784,
      "step": 203
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.6961129072194273e-05,
      "loss": 0.772,
      "step": 204
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.6960676120587377e-05,
      "loss": 0.7833,
      "step": 205
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.6960220551307266e-05,
      "loss": 0.8009,
      "step": 206
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.6959762364494895e-05,
      "loss": 0.8256,
      "step": 207
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.6959301560292025e-05,
      "loss": 0.7892,
      "step": 208
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.6958838138841224e-05,
      "loss": 0.8052,
      "step": 209
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.695837210028587e-05,
      "loss": 0.803,
      "step": 210
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.695790344477015e-05,
      "loss": 0.7894,
      "step": 211
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.6957432172439068e-05,
      "loss": 0.8169,
      "step": 212
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.695695828343843e-05,
      "loss": 0.815,
      "step": 213
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.6956481777914853e-05,
      "loss": 0.7869,
      "step": 214
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.6956002656015766e-05,
      "loss": 0.8465,
      "step": 215
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.6955520917889403e-05,
      "loss": 0.8193,
      "step": 216
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.6955036563684815e-05,
      "loss": 0.7775,
      "step": 217
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.695454959355186e-05,
      "loss": 0.8082,
      "step": 218
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.695406000764119e-05,
      "loss": 0.7888,
      "step": 219
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.695356780610429e-05,
      "loss": 0.794,
      "step": 220
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.6953072989093443e-05,
      "loss": 0.8027,
      "step": 221
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.6952575556761742e-05,
      "loss": 0.8116,
      "step": 222
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.6952075509263084e-05,
      "loss": 0.7781,
      "step": 223
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.6951572846752185e-05,
      "loss": 0.7995,
      "step": 224
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.6951067569384564e-05,
      "loss": 0.799,
      "step": 225
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.6950559677316548e-05,
      "loss": 0.8077,
      "step": 226
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.695004917070528e-05,
      "loss": 0.8142,
      "step": 227
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.69495360497087e-05,
      "loss": 0.8121,
      "step": 228
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.6949020314485574e-05,
      "loss": 0.8044,
      "step": 229
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.6948501965195458e-05,
      "loss": 0.7923,
      "step": 230
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.694798100199873e-05,
      "loss": 0.8066,
      "step": 231
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.6947457425056568e-05,
      "loss": 0.7512,
      "step": 232
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.694693123453097e-05,
      "loss": 0.7617,
      "step": 233
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.694640243058473e-05,
      "loss": 0.7869,
      "step": 234
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.6945871013381458e-05,
      "loss": 0.817,
      "step": 235
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.694533698308557e-05,
      "loss": 0.7898,
      "step": 236
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.694480033986229e-05,
      "loss": 0.8127,
      "step": 237
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.6944261083877654e-05,
      "loss": 0.8105,
      "step": 238
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.6943719215298506e-05,
      "loss": 0.8103,
      "step": 239
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.6943174734292492e-05,
      "loss": 0.7975,
      "step": 240
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.6942627641028073e-05,
      "loss": 0.7615,
      "step": 241
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.6942077935674514e-05,
      "loss": 0.8061,
      "step": 242
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.6941525618401894e-05,
      "loss": 0.7807,
      "step": 243
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.694097068938109e-05,
      "loss": 0.8169,
      "step": 244
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.6940413148783794e-05,
      "loss": 0.7953,
      "step": 245
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.693985299678251e-05,
      "loss": 0.8267,
      "step": 246
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.6939290233550543e-05,
      "loss": 0.7623,
      "step": 247
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.6938724859262002e-05,
      "loss": 0.8176,
      "step": 248
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.693815687409182e-05,
      "loss": 0.8046,
      "step": 249
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.6937586278215724e-05,
      "loss": 0.8294,
      "step": 250
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.6937013071810247e-05,
      "loss": 0.7812,
      "step": 251
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.693643725505274e-05,
      "loss": 0.8049,
      "step": 252
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.6935858828121354e-05,
      "loss": 0.7456,
      "step": 253
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.6935277791195055e-05,
      "loss": 0.7983,
      "step": 254
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.6934694144453608e-05,
      "loss": 0.7983,
      "step": 255
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.693410788807759e-05,
      "loss": 0.8016,
      "step": 256
    },
    {
      "epoch": 0.18,
      "eval_loss": 0.7754120230674744,
      "eval_runtime": 30.8738,
      "eval_samples_per_second": 265.306,
      "eval_steps_per_second": 16.584,
      "step": 256
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.693351902224838e-05,
      "loss": 0.7946,
      "step": 257
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.693292754714818e-05,
      "loss": 0.8031,
      "step": 258
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.6932333462959973e-05,
      "loss": 0.8106,
      "step": 259
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.693173676986758e-05,
      "loss": 0.8148,
      "step": 260
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.69311374680556e-05,
      "loss": 0.8,
      "step": 261
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.6930535557709463e-05,
      "loss": 0.7816,
      "step": 262
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.692993103901539e-05,
      "loss": 0.7945,
      "step": 263
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.692932391216042e-05,
      "loss": 0.8225,
      "step": 264
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.6928714177332386e-05,
      "loss": 0.8285,
      "step": 265
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.6928101834719942e-05,
      "loss": 0.8155,
      "step": 266
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.692748688451254e-05,
      "loss": 0.8228,
      "step": 267
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.6926869326900444e-05,
      "loss": 0.8353,
      "step": 268
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.6926249162074718e-05,
      "loss": 0.8233,
      "step": 269
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.6925626390227236e-05,
      "loss": 0.8081,
      "step": 270
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.6925001011550684e-05,
      "loss": 0.7618,
      "step": 271
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.6924373026238546e-05,
      "loss": 0.7921,
      "step": 272
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.6923742434485117e-05,
      "loss": 0.7971,
      "step": 273
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.69231092364855e-05,
      "loss": 0.7896,
      "step": 274
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.69224734324356e-05,
      "loss": 0.809,
      "step": 275
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.6921835022532127e-05,
      "loss": 0.7643,
      "step": 276
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.692119400697261e-05,
      "loss": 0.7724,
      "step": 277
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.6920550385955362e-05,
      "loss": 0.7917,
      "step": 278
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.6919904159679522e-05,
      "loss": 0.795,
      "step": 279
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.6919255328345027e-05,
      "loss": 0.7697,
      "step": 280
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.6918603892152623e-05,
      "loss": 0.8005,
      "step": 281
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.6917949851303857e-05,
      "loss": 0.8377,
      "step": 282
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.691729320600109e-05,
      "loss": 0.7902,
      "step": 283
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.691663395644747e-05,
      "loss": 0.7926,
      "step": 284
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.6915972102846976e-05,
      "loss": 0.8037,
      "step": 285
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.6915307645404377e-05,
      "loss": 0.7978,
      "step": 286
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.6914640584325254e-05,
      "loss": 0.774,
      "step": 287
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.6913970919815987e-05,
      "loss": 0.7894,
      "step": 288
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.6913298652083768e-05,
      "loss": 0.7951,
      "step": 289
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.691262378133659e-05,
      "loss": 0.8233,
      "step": 290
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.6911946307783257e-05,
      "loss": 0.8108,
      "step": 291
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.691126623163337e-05,
      "loss": 0.8113,
      "step": 292
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.6910583553097343e-05,
      "loss": 0.8154,
      "step": 293
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.6909898272386392e-05,
      "loss": 0.8217,
      "step": 294
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.6909210389712532e-05,
      "loss": 0.7712,
      "step": 295
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.6908519905288598e-05,
      "loss": 0.8128,
      "step": 296
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.6907826819328216e-05,
      "loss": 0.8046,
      "step": 297
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.690713113204582e-05,
      "loss": 0.7937,
      "step": 298
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.6906432843656652e-05,
      "loss": 0.7848,
      "step": 299
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.6905731954376763e-05,
      "loss": 0.8199,
      "step": 300
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.6905028464422996e-05,
      "loss": 0.788,
      "step": 301
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.690432237401301e-05,
      "loss": 0.8293,
      "step": 302
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.6903613683365264e-05,
      "loss": 0.7895,
      "step": 303
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.690290239269902e-05,
      "loss": 0.7983,
      "step": 304
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.6902188502234345e-05,
      "loss": 0.8147,
      "step": 305
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.690147201219212e-05,
      "loss": 0.7921,
      "step": 306
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.6900752922794008e-05,
      "loss": 0.8301,
      "step": 307
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.6900031234262502e-05,
      "loss": 0.8058,
      "step": 308
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.6899306946820885e-05,
      "loss": 0.8151,
      "step": 309
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.689858006069324e-05,
      "loss": 0.8258,
      "step": 310
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.6897850576104464e-05,
      "loss": 0.7922,
      "step": 311
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.6897118493280262e-05,
      "loss": 0.8135,
      "step": 312
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.6896383812447125e-05,
      "loss": 0.8037,
      "step": 313
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.689564653383236e-05,
      "loss": 0.7698,
      "step": 314
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.689490665766408e-05,
      "loss": 0.7941,
      "step": 315
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.6894164184171197e-05,
      "loss": 0.8212,
      "step": 316
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.689341911358342e-05,
      "loss": 0.8062,
      "step": 317
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.6892671446131277e-05,
      "loss": 0.7917,
      "step": 318
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.6891921182046087e-05,
      "loss": 0.7739,
      "step": 319
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.6891168321559976e-05,
      "loss": 0.7763,
      "step": 320
    },
    {
      "epoch": 0.23,
      "eval_loss": 0.7702016234397888,
      "eval_runtime": 30.8579,
      "eval_samples_per_second": 265.442,
      "eval_steps_per_second": 16.592,
      "step": 320
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.6890412864905874e-05,
      "loss": 0.7889,
      "step": 321
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.6889654812317518e-05,
      "loss": 0.7881,
      "step": 322
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.6888894164029435e-05,
      "loss": 0.7885,
      "step": 323
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.6888130920276974e-05,
      "loss": 0.7864,
      "step": 324
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.6887365081296272e-05,
      "loss": 0.8041,
      "step": 325
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.6886596647324277e-05,
      "loss": 0.8064,
      "step": 326
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.688582561859873e-05,
      "loss": 0.792,
      "step": 327
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.6885051995358192e-05,
      "loss": 0.7877,
      "step": 328
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.6884275777842013e-05,
      "loss": 0.8236,
      "step": 329
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.6883496966290342e-05,
      "loss": 0.8306,
      "step": 330
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.688271556094415e-05,
      "loss": 0.788,
      "step": 331
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.6881931562045185e-05,
      "loss": 0.8034,
      "step": 332
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.688114496983602e-05,
      "loss": 0.7898,
      "step": 333
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.688035578456002e-05,
      "loss": 0.803,
      "step": 334
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6879564006461352e-05,
      "loss": 0.8026,
      "step": 335
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6878769635784984e-05,
      "loss": 0.8316,
      "step": 336
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.687797267277669e-05,
      "loss": 0.8129,
      "step": 337
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6877173117683052e-05,
      "loss": 0.775,
      "step": 338
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6876370970751435e-05,
      "loss": 0.7659,
      "step": 339
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6875566232230026e-05,
      "loss": 0.824,
      "step": 340
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6874758902367802e-05,
      "loss": 0.8013,
      "step": 341
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6873948981414544e-05,
      "loss": 0.8139,
      "step": 342
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6873136469620837e-05,
      "loss": 0.801,
      "step": 343
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6872321367238068e-05,
      "loss": 0.7706,
      "step": 344
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6871503674518422e-05,
      "loss": 0.8227,
      "step": 345
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.687068339171489e-05,
      "loss": 0.7697,
      "step": 346
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.686986051908126e-05,
      "loss": 0.7895,
      "step": 347
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.686903505687212e-05,
      "loss": 0.7857,
      "step": 348
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.6868207005342865e-05,
      "loss": 0.7814,
      "step": 349
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.6867376364749692e-05,
      "loss": 0.8044,
      "step": 350
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.6866543135349586e-05,
      "loss": 0.7951,
      "step": 351
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.686570731740035e-05,
      "loss": 0.7987,
      "step": 352
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.686486891116058e-05,
      "loss": 0.8218,
      "step": 353
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.6864027916889666e-05,
      "loss": 0.7704,
      "step": 354
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.6863184334847812e-05,
      "loss": 0.7551,
      "step": 355
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.6862338165296013e-05,
      "loss": 0.7808,
      "step": 356
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.6861489408496077e-05,
      "loss": 0.8005,
      "step": 357
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.686063806471059e-05,
      "loss": 0.797,
      "step": 358
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.685978413420296e-05,
      "loss": 0.7495,
      "step": 359
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.685892761723738e-05,
      "loss": 0.8075,
      "step": 360
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.685806851407886e-05,
      "loss": 0.774,
      "step": 361
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.6857206824993192e-05,
      "loss": 0.8075,
      "step": 362
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.6856342550246983e-05,
      "loss": 0.7783,
      "step": 363
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.685547569010763e-05,
      "loss": 0.8634,
      "step": 364
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.6854606244843333e-05,
      "loss": 0.7866,
      "step": 365
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.685373421472309e-05,
      "loss": 0.7688,
      "step": 366
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.6852859600016707e-05,
      "loss": 0.7752,
      "step": 367
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.685198240099478e-05,
      "loss": 0.8219,
      "step": 368
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.6851102617928706e-05,
      "loss": 0.7731,
      "step": 369
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.685022025109069e-05,
      "loss": 0.7933,
      "step": 370
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.6849335300753726e-05,
      "loss": 0.8216,
      "step": 371
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.684844776719161e-05,
      "loss": 0.7697,
      "step": 372
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.684755765067894e-05,
      "loss": 0.7979,
      "step": 373
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.6846664951491114e-05,
      "loss": 0.8294,
      "step": 374
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.6845769669904323e-05,
      "loss": 0.7956,
      "step": 375
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.6844871806195564e-05,
      "loss": 0.7685,
      "step": 376
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.6843971360642628e-05,
      "loss": 0.8083,
      "step": 377
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.6843068333524106e-05,
      "loss": 0.7766,
      "step": 378
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.684216272511939e-05,
      "loss": 0.7904,
      "step": 379
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.6841254535708666e-05,
      "loss": 0.8092,
      "step": 380
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.6840343765572924e-05,
      "loss": 0.7931,
      "step": 381
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.683943041499395e-05,
      "loss": 0.7797,
      "step": 382
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.6838514484254326e-05,
      "loss": 0.785,
      "step": 383
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.6837595973637433e-05,
      "loss": 0.7809,
      "step": 384
    },
    {
      "epoch": 0.27,
      "eval_loss": 0.7656841278076172,
      "eval_runtime": 30.883,
      "eval_samples_per_second": 265.227,
      "eval_steps_per_second": 16.579,
      "step": 384
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.683667488342746e-05,
      "loss": 0.7759,
      "step": 385
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.6835751213909377e-05,
      "loss": 0.7962,
      "step": 386
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.6834824965368965e-05,
      "loss": 0.7989,
      "step": 387
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.683389613809279e-05,
      "loss": 0.8343,
      "step": 388
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.683296473236824e-05,
      "loss": 0.8368,
      "step": 389
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.683203074848348e-05,
      "loss": 0.7931,
      "step": 390
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.683109418672747e-05,
      "loss": 0.8007,
      "step": 391
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.6830155047389983e-05,
      "loss": 0.7701,
      "step": 392
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.682921333076158e-05,
      "loss": 0.7993,
      "step": 393
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.682826903713362e-05,
      "loss": 0.7663,
      "step": 394
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.682732216679826e-05,
      "loss": 0.8011,
      "step": 395
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.6826372720048464e-05,
      "loss": 0.8051,
      "step": 396
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.6825420697177973e-05,
      "loss": 0.772,
      "step": 397
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.682446609848134e-05,
      "loss": 0.7901,
      "step": 398
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.6823508924253916e-05,
      "loss": 0.8279,
      "step": 399
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.6822549174791836e-05,
      "loss": 0.7657,
      "step": 400
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.6821586850392043e-05,
      "loss": 0.8119,
      "step": 401
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.6820621951352272e-05,
      "loss": 0.7933,
      "step": 402
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.681965447797106e-05,
      "loss": 0.7838,
      "step": 403
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.6818684430547733e-05,
      "loss": 0.7944,
      "step": 404
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.681771180938242e-05,
      "loss": 0.8052,
      "step": 405
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.681673661477604e-05,
      "loss": 0.785,
      "step": 406
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.681575884703031e-05,
      "loss": 0.7816,
      "step": 407
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.681477850644775e-05,
      "loss": 0.7619,
      "step": 408
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.681379559333166e-05,
      "loss": 0.7699,
      "step": 409
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.6812810107986157e-05,
      "loss": 0.7532,
      "step": 410
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.681182205071614e-05,
      "loss": 0.7902,
      "step": 411
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.6810831421827303e-05,
      "loss": 0.761,
      "step": 412
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.6809838221626147e-05,
      "loss": 0.7818,
      "step": 413
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.6808842450419957e-05,
      "loss": 0.7538,
      "step": 414
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.6807844108516815e-05,
      "loss": 0.7797,
      "step": 415
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.6806843196225605e-05,
      "loss": 0.7718,
      "step": 416
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.6805839713855998e-05,
      "loss": 0.8179,
      "step": 417
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.680483366171847e-05,
      "loss": 0.7914,
      "step": 418
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.6803825040124284e-05,
      "loss": 0.7762,
      "step": 419
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.6802813849385497e-05,
      "loss": 0.7898,
      "step": 420
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.6801800089814968e-05,
      "loss": 0.7869,
      "step": 421
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.6800783761726345e-05,
      "loss": 0.814,
      "step": 422
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.6799764865434077e-05,
      "loss": 0.8048,
      "step": 423
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.6798743401253395e-05,
      "loss": 0.82,
      "step": 424
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.679771936950034e-05,
      "loss": 0.7757,
      "step": 425
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.679669277049174e-05,
      "loss": 0.7881,
      "step": 426
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.6795663604545214e-05,
      "loss": 0.8054,
      "step": 427
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.6794631871979177e-05,
      "loss": 0.7939,
      "step": 428
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.6793597573112846e-05,
      "loss": 0.8007,
      "step": 429
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.679256070826622e-05,
      "loss": 0.7876,
      "step": 430
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.6791521277760102e-05,
      "loss": 0.7787,
      "step": 431
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.6790479281916082e-05,
      "loss": 0.7895,
      "step": 432
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.6789434721056544e-05,
      "loss": 0.8033,
      "step": 433
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.6788387595504675e-05,
      "loss": 0.8041,
      "step": 434
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.6787337905584444e-05,
      "loss": 0.8019,
      "step": 435
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.678628565162061e-05,
      "loss": 0.8234,
      "step": 436
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.6785230833938748e-05,
      "loss": 0.7683,
      "step": 437
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.67841734528652e-05,
      "loss": 0.7853,
      "step": 438
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.6783113508727115e-05,
      "loss": 0.8335,
      "step": 439
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.6782051001852433e-05,
      "loss": 0.8013,
      "step": 440
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.6780985932569888e-05,
      "loss": 0.7891,
      "step": 441
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.6779918301208996e-05,
      "loss": 0.7884,
      "step": 442
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.6778848108100087e-05,
      "loss": 0.7991,
      "step": 443
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.677777535357427e-05,
      "loss": 0.7787,
      "step": 444
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.6776700037963432e-05,
      "loss": 0.7895,
      "step": 445
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.6775622161600286e-05,
      "loss": 0.7851,
      "step": 446
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.677454172481831e-05,
      "loss": 0.8025,
      "step": 447
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.6773458727951787e-05,
      "loss": 0.7687,
      "step": 448
    },
    {
      "epoch": 0.32,
      "eval_loss": 0.7622755765914917,
      "eval_runtime": 30.8779,
      "eval_samples_per_second": 265.271,
      "eval_steps_per_second": 16.581,
      "step": 448
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.6772373171335787e-05,
      "loss": 0.7883,
      "step": 449
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.6771285055306175e-05,
      "loss": 0.8017,
      "step": 450
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.6770194380199603e-05,
      "loss": 0.7921,
      "step": 451
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.6769101146353522e-05,
      "loss": 0.794,
      "step": 452
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.676800535410617e-05,
      "loss": 0.7928,
      "step": 453
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.6766907003796574e-05,
      "loss": 0.7817,
      "step": 454
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.6765806095764557e-05,
      "loss": 0.7619,
      "step": 455
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.676470263035074e-05,
      "loss": 0.7909,
      "step": 456
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.6763596607896517e-05,
      "loss": 0.7796,
      "step": 457
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.6762488028744085e-05,
      "loss": 0.7942,
      "step": 458
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.6761376893236436e-05,
      "loss": 0.789,
      "step": 459
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.6760263201717344e-05,
      "loss": 0.7755,
      "step": 460
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.675914695453138e-05,
      "loss": 0.7967,
      "step": 461
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.6758028152023896e-05,
      "loss": 0.8041,
      "step": 462
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.6756906794541047e-05,
      "loss": 0.7954,
      "step": 463
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.6755782882429774e-05,
      "loss": 0.8182,
      "step": 464
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.6754656416037803e-05,
      "loss": 0.7715,
      "step": 465
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.6753527395713663e-05,
      "loss": 0.8045,
      "step": 466
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.6752395821806655e-05,
      "loss": 0.7951,
      "step": 467
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.6751261694666885e-05,
      "loss": 0.7655,
      "step": 468
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.6750125014645243e-05,
      "loss": 0.7813,
      "step": 469
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.674898578209341e-05,
      "loss": 0.7936,
      "step": 470
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.6747843997363857e-05,
      "loss": 0.7683,
      "step": 471
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.674669966080984e-05,
      "loss": 0.795,
      "step": 472
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.6745552772785417e-05,
      "loss": 0.756,
      "step": 473
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.674440333364542e-05,
      "loss": 0.7557,
      "step": 474
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.6743251343745478e-05,
      "loss": 0.785,
      "step": 475
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.6742096803442012e-05,
      "loss": 0.8145,
      "step": 476
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.6740939713092225e-05,
      "loss": 0.7759,
      "step": 477
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.6739780073054112e-05,
      "loss": 0.759,
      "step": 478
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.6738617883686464e-05,
      "loss": 0.7754,
      "step": 479
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.673745314534885e-05,
      "loss": 0.7706,
      "step": 480
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.6736285858401627e-05,
      "loss": 0.7806,
      "step": 481
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.673511602320595e-05,
      "loss": 0.7916,
      "step": 482
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.673394364012376e-05,
      "loss": 0.7854,
      "step": 483
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.6732768709517784e-05,
      "loss": 0.7895,
      "step": 484
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.673159123175153e-05,
      "loss": 0.8096,
      "step": 485
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.6730411207189306e-05,
      "loss": 0.8298,
      "step": 486
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.6729228636196203e-05,
      "loss": 0.7829,
      "step": 487
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.6728043519138102e-05,
      "loss": 0.7667,
      "step": 488
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.6726855856381668e-05,
      "loss": 0.787,
      "step": 489
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.6725665648294355e-05,
      "loss": 0.7929,
      "step": 490
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.6724472895244405e-05,
      "loss": 0.7538,
      "step": 491
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.6723277597600847e-05,
      "loss": 0.7755,
      "step": 492
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.67220797557335e-05,
      "loss": 0.8047,
      "step": 493
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.6720879370012967e-05,
      "loss": 0.7884,
      "step": 494
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.6719676440810636e-05,
      "loss": 0.8054,
      "step": 495
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.6718470968498685e-05,
      "loss": 0.7881,
      "step": 496
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.6717262953450083e-05,
      "loss": 0.8083,
      "step": 497
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.671605239603858e-05,
      "loss": 0.8027,
      "step": 498
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.671483929663871e-05,
      "loss": 0.8077,
      "step": 499
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.67136236556258e-05,
      "loss": 0.7509,
      "step": 500
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.671240547337596e-05,
      "loss": 0.764,
      "step": 501
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.671118475026609e-05,
      "loss": 0.7879,
      "step": 502
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.670996148667387e-05,
      "loss": 0.7931,
      "step": 503
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.6708735682977767e-05,
      "loss": 0.8074,
      "step": 504
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.6707507339557045e-05,
      "loss": 0.7592,
      "step": 505
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.670627645679173e-05,
      "loss": 0.7944,
      "step": 506
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.6705043035062662e-05,
      "loss": 0.7969,
      "step": 507
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.6703807074751447e-05,
      "loss": 0.8282,
      "step": 508
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.6702568576240482e-05,
      "loss": 0.8053,
      "step": 509
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.6701327539912952e-05,
      "loss": 0.7942,
      "step": 510
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.670008396615282e-05,
      "loss": 0.7756,
      "step": 511
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.6698837855344846e-05,
      "loss": 0.793,
      "step": 512
    },
    {
      "epoch": 0.36,
      "eval_loss": 0.7590444087982178,
      "eval_runtime": 30.9135,
      "eval_samples_per_second": 264.965,
      "eval_steps_per_second": 16.562,
      "step": 512
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.6697589207874564e-05,
      "loss": 0.792,
      "step": 513
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.6696338024128295e-05,
      "loss": 0.7735,
      "step": 514
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.669508430449315e-05,
      "loss": 0.7991,
      "step": 515
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.6693828049357022e-05,
      "loss": 0.8135,
      "step": 516
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.669256925910858e-05,
      "loss": 0.7769,
      "step": 517
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.6691307934137292e-05,
      "loss": 0.7803,
      "step": 518
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.66900440748334e-05,
      "loss": 0.7557,
      "step": 519
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.6688777681587937e-05,
      "loss": 0.7734,
      "step": 520
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.668750875479271e-05,
      "loss": 0.7795,
      "step": 521
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.6686237294840315e-05,
      "loss": 0.7796,
      "step": 522
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.6684963302124142e-05,
      "loss": 0.7821,
      "step": 523
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.668368677703835e-05,
      "loss": 0.8237,
      "step": 524
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.668240771997788e-05,
      "loss": 0.7892,
      "step": 525
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.6681126131338476e-05,
      "loss": 0.7648,
      "step": 526
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.6679842011516643e-05,
      "loss": 0.7819,
      "step": 527
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.6678555360909683e-05,
      "loss": 0.8292,
      "step": 528
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.6677266179915677e-05,
      "loss": 0.794,
      "step": 529
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.6675974468933486e-05,
      "loss": 0.7861,
      "step": 530
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.6674680228362756e-05,
      "loss": 0.7603,
      "step": 531
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.6673383458603918e-05,
      "loss": 0.765,
      "step": 532
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.667208416005818e-05,
      "loss": 0.8044,
      "step": 533
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.667078233312754e-05,
      "loss": 0.7763,
      "step": 534
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.666947797821477e-05,
      "loss": 0.7959,
      "step": 535
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.6668171095723432e-05,
      "loss": 0.7692,
      "step": 536
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.6666861686057864e-05,
      "loss": 0.7898,
      "step": 537
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.6665549749623188e-05,
      "loss": 0.7892,
      "step": 538
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.6664235286825306e-05,
      "loss": 0.7741,
      "step": 539
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.6662918298070913e-05,
      "loss": 0.785,
      "step": 540
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.666159878376746e-05,
      "loss": 0.7591,
      "step": 541
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.666027674432321e-05,
      "loss": 0.7472,
      "step": 542
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.6658952180147184e-05,
      "loss": 0.7996,
      "step": 543
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.6657625091649198e-05,
      "loss": 0.7807,
      "step": 544
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.665629547923984e-05,
      "loss": 0.7836,
      "step": 545
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.6654963343330485e-05,
      "loss": 0.7762,
      "step": 546
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.665362868433328e-05,
      "loss": 0.7841,
      "step": 547
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.6652291502661167e-05,
      "loss": 0.7646,
      "step": 548
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.665095179872786e-05,
      "loss": 0.7986,
      "step": 549
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.6649609572947853e-05,
      "loss": 0.8159,
      "step": 550
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.6648264825736417e-05,
      "loss": 0.7722,
      "step": 551
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.664691755750961e-05,
      "loss": 0.7801,
      "step": 552
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.6645567768684268e-05,
      "loss": 0.7287,
      "step": 553
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.6644215459678008e-05,
      "loss": 0.7672,
      "step": 554
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.6642860630909218e-05,
      "loss": 0.8058,
      "step": 555
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.664150328279708e-05,
      "loss": 0.7678,
      "step": 556
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.6640143415761543e-05,
      "loss": 0.7725,
      "step": 557
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.6638781030223344e-05,
      "loss": 0.7584,
      "step": 558
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.663741612660399e-05,
      "loss": 0.7555,
      "step": 559
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.6636048705325778e-05,
      "loss": 0.7906,
      "step": 560
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.6634678766811776e-05,
      "loss": 0.8196,
      "step": 561
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.6633306311485835e-05,
      "loss": 0.8064,
      "step": 562
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.6631931339772576e-05,
      "loss": 0.8127,
      "step": 563
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.6630553852097416e-05,
      "loss": 0.7607,
      "step": 564
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.6629173848886533e-05,
      "loss": 0.7629,
      "step": 565
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.6627791330566892e-05,
      "loss": 0.7744,
      "step": 566
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.6626406297566236e-05,
      "loss": 0.785,
      "step": 567
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.662501875031308e-05,
      "loss": 0.7996,
      "step": 568
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.6623628689236728e-05,
      "loss": 0.7841,
      "step": 569
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.662223611476725e-05,
      "loss": 0.7845,
      "step": 570
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.6620841027335495e-05,
      "loss": 0.7827,
      "step": 571
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.6619443427373103e-05,
      "loss": 0.7512,
      "step": 572
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.6618043315312473e-05,
      "loss": 0.7909,
      "step": 573
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.6616640691586798e-05,
      "loss": 0.7786,
      "step": 574
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.661523555663003e-05,
      "loss": 0.7608,
      "step": 575
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.6613827910876918e-05,
      "loss": 0.7449,
      "step": 576
    },
    {
      "epoch": 0.41,
      "eval_loss": 0.7557086944580078,
      "eval_runtime": 30.9158,
      "eval_samples_per_second": 264.945,
      "eval_steps_per_second": 16.561,
      "step": 576
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.661241775476297e-05,
      "loss": 0.805,
      "step": 577
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.661100508872448e-05,
      "loss": 0.7903,
      "step": 578
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.6609589913198517e-05,
      "loss": 0.7779,
      "step": 579
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.6608172228622932e-05,
      "loss": 0.7601,
      "step": 580
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.6606752035436335e-05,
      "loss": 0.7902,
      "step": 581
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.6605329334078134e-05,
      "loss": 0.7902,
      "step": 582
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.6603904124988496e-05,
      "loss": 0.7611,
      "step": 583
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.660247640860838e-05,
      "loss": 0.7786,
      "step": 584
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.6601046185379498e-05,
      "loss": 0.7677,
      "step": 585
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.6599613455744358e-05,
      "loss": 0.7635,
      "step": 586
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.6598178220146238e-05,
      "loss": 0.7456,
      "step": 587
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.659674047902919e-05,
      "loss": 0.8087,
      "step": 588
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.6595300232838038e-05,
      "loss": 0.7694,
      "step": 589
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.6593857482018383e-05,
      "loss": 0.768,
      "step": 590
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.6592412227016604e-05,
      "loss": 0.8055,
      "step": 591
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.659096446827985e-05,
      "loss": 0.7865,
      "step": 592
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.6589514206256052e-05,
      "loss": 0.8172,
      "step": 593
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.6588061441393908e-05,
      "loss": 0.7468,
      "step": 594
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.6586606174142894e-05,
      "loss": 0.8009,
      "step": 595
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.6585148404953258e-05,
      "loss": 0.7906,
      "step": 596
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.6583688134276026e-05,
      "loss": 0.7863,
      "step": 597
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.6582225362562994e-05,
      "loss": 0.7776,
      "step": 598
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.658076009026673e-05,
      "loss": 0.7848,
      "step": 599
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.6579292317840583e-05,
      "loss": 0.7828,
      "step": 600
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.657782204573867e-05,
      "loss": 0.7654,
      "step": 601
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.6576349274415883e-05,
      "loss": 0.7833,
      "step": 602
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.657487400432789e-05,
      "loss": 0.7917,
      "step": 603
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.657339623593112e-05,
      "loss": 0.7811,
      "step": 604
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.6571915969682794e-05,
      "loss": 0.76,
      "step": 605
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.6570433206040893e-05,
      "loss": 0.7989,
      "step": 606
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.6568947945464168e-05,
      "loss": 0.8041,
      "step": 607
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.6567460188412156e-05,
      "loss": 0.7782,
      "step": 608
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.6565969935345153e-05,
      "loss": 0.8132,
      "step": 609
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.6564477186724236e-05,
      "loss": 0.7627,
      "step": 610
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.6562981943011252e-05,
      "loss": 0.7814,
      "step": 611
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.6561484204668815e-05,
      "loss": 0.7635,
      "step": 612
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.6559983972160318e-05,
      "loss": 0.7532,
      "step": 613
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.6558481245949917e-05,
      "loss": 0.7719,
      "step": 614
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.6556976026502548e-05,
      "loss": 0.8005,
      "step": 615
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.6555468314283918e-05,
      "loss": 0.7867,
      "step": 616
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.6553958109760497e-05,
      "loss": 0.7965,
      "step": 617
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.6552445413399537e-05,
      "loss": 0.796,
      "step": 618
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.6550930225669052e-05,
      "loss": 0.7645,
      "step": 619
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.6549412547037832e-05,
      "loss": 0.7613,
      "step": 620
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.6547892377975435e-05,
      "loss": 0.7674,
      "step": 621
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.654636971895219e-05,
      "loss": 0.7814,
      "step": 622
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.6544844570439202e-05,
      "loss": 0.7879,
      "step": 623
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.654331693290833e-05,
      "loss": 0.776,
      "step": 624
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.6541786806832228e-05,
      "loss": 0.7747,
      "step": 625
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.6540254192684294e-05,
      "loss": 0.8548,
      "step": 626
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.6538719090938718e-05,
      "loss": 0.7913,
      "step": 627
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.653718150207044e-05,
      "loss": 0.7653,
      "step": 628
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.6535641426555188e-05,
      "loss": 0.7859,
      "step": 629
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.6534098864869448e-05,
      "loss": 0.7828,
      "step": 630
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.653255381749047e-05,
      "loss": 0.799,
      "step": 631
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.653100628489629e-05,
      "loss": 0.7708,
      "step": 632
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.6529456267565702e-05,
      "loss": 0.7913,
      "step": 633
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.6527903765978267e-05,
      "loss": 0.7811,
      "step": 634
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.6526348780614318e-05,
      "loss": 0.7775,
      "step": 635
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.652479131195496e-05,
      "loss": 0.7663,
      "step": 636
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.6523231360482063e-05,
      "loss": 0.7824,
      "step": 637
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.652166892667826e-05,
      "loss": 0.7487,
      "step": 638
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.6520104011026962e-05,
      "loss": 0.7872,
      "step": 639
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.651853661401234e-05,
      "loss": 0.8014,
      "step": 640
    },
    {
      "epoch": 0.45,
      "eval_loss": 0.7533817887306213,
      "eval_runtime": 30.9,
      "eval_samples_per_second": 265.081,
      "eval_steps_per_second": 16.57,
      "step": 640
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.6516966736119334e-05,
      "loss": 0.7999,
      "step": 641
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.6515394377833657e-05,
      "loss": 0.7854,
      "step": 642
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.6513819539641783e-05,
      "loss": 0.7899,
      "step": 643
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.651224222203096e-05,
      "loss": 0.7785,
      "step": 644
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.6510662425489193e-05,
      "loss": 0.7772,
      "step": 645
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.6509080150505262e-05,
      "loss": 0.7572,
      "step": 646
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.650749539756871e-05,
      "loss": 0.7648,
      "step": 647
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.6505908167169852e-05,
      "loss": 0.82,
      "step": 648
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.6504318459799767e-05,
      "loss": 0.7513,
      "step": 649
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.6502726275950295e-05,
      "loss": 0.7941,
      "step": 650
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.6501131616114044e-05,
      "loss": 0.7326,
      "step": 651
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.6499534480784395e-05,
      "loss": 0.7693,
      "step": 652
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.649793487045549e-05,
      "loss": 0.7657,
      "step": 653
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.649633278562224e-05,
      "loss": 0.7786,
      "step": 654
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.6494728226780315e-05,
      "loss": 0.7793,
      "step": 655
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.649312119442615e-05,
      "loss": 0.7791,
      "step": 656
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.649151168905696e-05,
      "loss": 0.7831,
      "step": 657
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.64898997111707e-05,
      "loss": 0.8001,
      "step": 658
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.6488285261266118e-05,
      "loss": 0.7976,
      "step": 659
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.648666833984271e-05,
      "loss": 0.7723,
      "step": 660
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6485048947400734e-05,
      "loss": 0.7682,
      "step": 661
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6483427084441225e-05,
      "loss": 0.7946,
      "step": 662
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6481802751465975e-05,
      "loss": 0.7905,
      "step": 663
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.648017594897754e-05,
      "loss": 0.8145,
      "step": 664
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6478546677479236e-05,
      "loss": 0.776,
      "step": 665
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6476914937475158e-05,
      "loss": 0.7828,
      "step": 666
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6475280729470148e-05,
      "loss": 0.7992,
      "step": 667
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.647364405396982e-05,
      "loss": 0.7863,
      "step": 668
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6472004911480552e-05,
      "loss": 0.7896,
      "step": 669
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.647036330250948e-05,
      "loss": 0.8092,
      "step": 670
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6468719227564504e-05,
      "loss": 0.7609,
      "step": 671
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6467072687154296e-05,
      "loss": 0.7865,
      "step": 672
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.646542368178828e-05,
      "loss": 0.8037,
      "step": 673
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.646377221197665e-05,
      "loss": 0.7817,
      "step": 674
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.646211827823035e-05,
      "loss": 0.7647,
      "step": 675
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6460461881061103e-05,
      "loss": 0.7596,
      "step": 676
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6458803020981386e-05,
      "loss": 0.7722,
      "step": 677
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6457141698504437e-05,
      "loss": 0.8003,
      "step": 678
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6455477914144258e-05,
      "loss": 0.7948,
      "step": 679
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.645381166841561e-05,
      "loss": 0.7822,
      "step": 680
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6452142961834022e-05,
      "loss": 0.7955,
      "step": 681
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6450471794915778e-05,
      "loss": 0.7758,
      "step": 682
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.644879816817792e-05,
      "loss": 0.7708,
      "step": 683
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6447122082138263e-05,
      "loss": 0.7679,
      "step": 684
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6445443537315378e-05,
      "loss": 0.7871,
      "step": 685
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6443762534228585e-05,
      "loss": 0.7795,
      "step": 686
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6442079073397986e-05,
      "loss": 0.7704,
      "step": 687
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6440393155344423e-05,
      "loss": 0.7305,
      "step": 688
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.643870478058951e-05,
      "loss": 0.7427,
      "step": 689
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.643701394965562e-05,
      "loss": 0.758,
      "step": 690
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6435320663065885e-05,
      "loss": 0.7601,
      "step": 691
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6433624921344188e-05,
      "loss": 0.774,
      "step": 692
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6431926725015187e-05,
      "loss": 0.7939,
      "step": 693
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6430226074604292e-05,
      "loss": 0.7806,
      "step": 694
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6428522970637667e-05,
      "loss": 0.7476,
      "step": 695
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6426817413642244e-05,
      "loss": 0.7933,
      "step": 696
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6425109404145712e-05,
      "loss": 0.7713,
      "step": 697
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6423398942676513e-05,
      "loss": 0.8022,
      "step": 698
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.642168602976385e-05,
      "loss": 0.7871,
      "step": 699
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6419970665937696e-05,
      "loss": 0.8019,
      "step": 700
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.641825285172876e-05,
      "loss": 0.7336,
      "step": 701
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.641653258766853e-05,
      "loss": 0.81,
      "step": 702
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6414809874289243e-05,
      "loss": 0.7829,
      "step": 703
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.641308471212389e-05,
      "loss": 0.782,
      "step": 704
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.7518882751464844,
      "eval_runtime": 30.8954,
      "eval_samples_per_second": 265.12,
      "eval_steps_per_second": 16.572,
      "step": 704
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6411357101706223e-05,
      "loss": 0.7866,
      "step": 705
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6409627043570755e-05,
      "loss": 0.7656,
      "step": 706
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.640789453825276e-05,
      "loss": 0.7707,
      "step": 707
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6406159586288254e-05,
      "loss": 0.8005,
      "step": 708
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.640442218821402e-05,
      "loss": 0.7751,
      "step": 709
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.64026823445676e-05,
      "loss": 0.7733,
      "step": 710
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6400940055887288e-05,
      "loss": 0.7986,
      "step": 711
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6399195322712136e-05,
      "loss": 0.8281,
      "step": 712
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6397448145581953e-05,
      "loss": 0.761,
      "step": 713
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.63956985250373e-05,
      "loss": 0.7431,
      "step": 714
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.63939464616195e-05,
      "loss": 0.7732,
      "step": 715
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6392191955870627e-05,
      "loss": 0.7682,
      "step": 716
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6390435008333514e-05,
      "loss": 0.7752,
      "step": 717
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.638867561955175e-05,
      "loss": 0.7982,
      "step": 718
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.638691379006967e-05,
      "loss": 0.7752,
      "step": 719
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.638514952043238e-05,
      "loss": 0.7775,
      "step": 720
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.638338281118573e-05,
      "loss": 0.7656,
      "step": 721
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6381613662876327e-05,
      "loss": 0.7557,
      "step": 722
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.637984207605153e-05,
      "loss": 0.7934,
      "step": 723
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.637806805125946e-05,
      "loss": 0.7843,
      "step": 724
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6376291589048985e-05,
      "loss": 0.7595,
      "step": 725
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6374512689969732e-05,
      "loss": 0.7543,
      "step": 726
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6372731354572076e-05,
      "loss": 0.7708,
      "step": 727
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6370947583407154e-05,
      "loss": 0.7687,
      "step": 728
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6369161377026846e-05,
      "loss": 0.7863,
      "step": 729
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.63673727359838e-05,
      "loss": 0.7536,
      "step": 730
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6365581660831405e-05,
      "loss": 0.8022,
      "step": 731
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.63637881521238e-05,
      "loss": 0.7868,
      "step": 732
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6361992210415895e-05,
      "loss": 0.7825,
      "step": 733
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6360193836263337e-05,
      "loss": 0.7532,
      "step": 734
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.635839303022253e-05,
      "loss": 0.7716,
      "step": 735
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6356589792850627e-05,
      "loss": 0.7334,
      "step": 736
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6354784124705545e-05,
      "loss": 0.7997,
      "step": 737
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.635297602634594e-05,
      "loss": 0.733,
      "step": 738
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6351165498331222e-05,
      "loss": 0.7805,
      "step": 739
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.634935254122156e-05,
      "loss": 0.7724,
      "step": 740
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6347537155577872e-05,
      "loss": 0.7577,
      "step": 741
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6345719341961818e-05,
      "loss": 0.7888,
      "step": 742
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6343899100935822e-05,
      "loss": 0.8009,
      "step": 743
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6342076433063055e-05,
      "loss": 0.7517,
      "step": 744
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6340251338907433e-05,
      "loss": 0.7781,
      "step": 745
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.633842381903363e-05,
      "loss": 0.7482,
      "step": 746
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.633659387400707e-05,
      "loss": 0.7898,
      "step": 747
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6334761504393915e-05,
      "loss": 0.8087,
      "step": 748
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.63329267107611e-05,
      "loss": 0.8032,
      "step": 749
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.633108949367629e-05,
      "loss": 0.7988,
      "step": 750
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.632924985370791e-05,
      "loss": 0.7656,
      "step": 751
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6327407791425127e-05,
      "loss": 0.7677,
      "step": 752
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.632556330739787e-05,
      "loss": 0.7738,
      "step": 753
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6323716402196798e-05,
      "loss": 0.7818,
      "step": 754
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6321867076393336e-05,
      "loss": 0.7761,
      "step": 755
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.632001533055966e-05,
      "loss": 0.7984,
      "step": 756
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6318161165268673e-05,
      "loss": 0.7924,
      "step": 757
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.631630458109405e-05,
      "loss": 0.7722,
      "step": 758
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6314445578610206e-05,
      "loss": 0.7843,
      "step": 759
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6312584158392295e-05,
      "loss": 0.7962,
      "step": 760
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.631072032101623e-05,
      "loss": 0.8121,
      "step": 761
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6308854067058677e-05,
      "loss": 0.7868,
      "step": 762
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6306985397097032e-05,
      "loss": 0.746,
      "step": 763
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6305114311709452e-05,
      "loss": 0.7596,
      "step": 764
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.630324081147484e-05,
      "loss": 0.7831,
      "step": 765
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.630136489697284e-05,
      "loss": 0.7597,
      "step": 766
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6299486568783848e-05,
      "loss": 0.7601,
      "step": 767
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6297605827489002e-05,
      "loss": 0.746,
      "step": 768
    },
    {
      "epoch": 0.54,
      "eval_loss": 0.7493306398391724,
      "eval_runtime": 30.9193,
      "eval_samples_per_second": 264.916,
      "eval_steps_per_second": 16.559,
      "step": 768
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6295722673670196e-05,
      "loss": 0.7807,
      "step": 769
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.629383710791006e-05,
      "loss": 0.7803,
      "step": 770
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6291949130791977e-05,
      "loss": 0.7413,
      "step": 771
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6290058742900076e-05,
      "loss": 0.7829,
      "step": 772
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6288165944819227e-05,
      "loss": 0.7523,
      "step": 773
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6286270737135042e-05,
      "loss": 0.8266,
      "step": 774
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6284373120433895e-05,
      "loss": 0.7653,
      "step": 775
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.628247309530289e-05,
      "loss": 0.7592,
      "step": 776
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6280570662329884e-05,
      "loss": 0.7569,
      "step": 777
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6278665822103475e-05,
      "loss": 0.7637,
      "step": 778
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6276758575213005e-05,
      "loss": 0.7847,
      "step": 779
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6274848922248565e-05,
      "loss": 0.8165,
      "step": 780
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6272936863800987e-05,
      "loss": 0.7819,
      "step": 781
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6271022400461847e-05,
      "loss": 0.7588,
      "step": 782
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.626910553282347e-05,
      "loss": 0.7652,
      "step": 783
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6267186261478922e-05,
      "loss": 0.7668,
      "step": 784
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6265264587022005e-05,
      "loss": 0.8075,
      "step": 785
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.626334051004728e-05,
      "loss": 0.7678,
      "step": 786
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6261414031150035e-05,
      "loss": 0.7553,
      "step": 787
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.625948515092631e-05,
      "loss": 0.7743,
      "step": 788
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6257553869972896e-05,
      "loss": 0.7331,
      "step": 789
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.625562018888731e-05,
      "loss": 0.7833,
      "step": 790
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6253684108267812e-05,
      "loss": 0.7677,
      "step": 791
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.625174562871343e-05,
      "loss": 0.7648,
      "step": 792
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.62498047508239e-05,
      "loss": 0.7649,
      "step": 793
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.624786147519972e-05,
      "loss": 0.7845,
      "step": 794
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6245915802442132e-05,
      "loss": 0.7709,
      "step": 795
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6243967733153107e-05,
      "loss": 0.7755,
      "step": 796
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6242017267935367e-05,
      "loss": 0.7298,
      "step": 797
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6240064407392367e-05,
      "loss": 0.783,
      "step": 798
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6238109152128314e-05,
      "loss": 0.7734,
      "step": 799
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6236151502748147e-05,
      "loss": 0.7882,
      "step": 800
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.623419145985755e-05,
      "loss": 0.7857,
      "step": 801
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6232229024062945e-05,
      "loss": 0.8048,
      "step": 802
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.62302641959715e-05,
      "loss": 0.7314,
      "step": 803
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6228296976191114e-05,
      "loss": 0.798,
      "step": 804
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6226327365330428e-05,
      "loss": 0.779,
      "step": 805
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6224355363998837e-05,
      "loss": 0.7682,
      "step": 806
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.622238097280645e-05,
      "loss": 0.7636,
      "step": 807
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6220404192364146e-05,
      "loss": 0.8095,
      "step": 808
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.621842502328351e-05,
      "loss": 0.778,
      "step": 809
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6216443466176895e-05,
      "loss": 0.7677,
      "step": 810
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6214459521657377e-05,
      "loss": 0.8022,
      "step": 811
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6212473190338773e-05,
      "loss": 0.7694,
      "step": 812
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.621048447283564e-05,
      "loss": 0.8016,
      "step": 813
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6208493369763275e-05,
      "loss": 0.7326,
      "step": 814
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.620649988173771e-05,
      "loss": 0.7697,
      "step": 815
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.620450400937571e-05,
      "loss": 0.7691,
      "step": 816
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6202505753294796e-05,
      "loss": 0.7768,
      "step": 817
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6200505114113206e-05,
      "loss": 0.7666,
      "step": 818
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6198502092449926e-05,
      "loss": 0.7609,
      "step": 819
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.619649668892468e-05,
      "loss": 0.774,
      "step": 820
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6194488904157912e-05,
      "loss": 0.7896,
      "step": 821
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6192478738770838e-05,
      "loss": 0.7622,
      "step": 822
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.619046619338537e-05,
      "loss": 0.7753,
      "step": 823
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6188451268624185e-05,
      "loss": 0.7604,
      "step": 824
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6186433965110683e-05,
      "loss": 0.7915,
      "step": 825
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6184414283469004e-05,
      "loss": 0.7568,
      "step": 826
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6182392224324026e-05,
      "loss": 0.755,
      "step": 827
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6180367788301355e-05,
      "loss": 0.7442,
      "step": 828
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.617834097602734e-05,
      "loss": 0.7793,
      "step": 829
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.617631178812906e-05,
      "loss": 0.7926,
      "step": 830
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6174280225234335e-05,
      "loss": 0.7663,
      "step": 831
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6172246287971713e-05,
      "loss": 0.767,
      "step": 832
    },
    {
      "epoch": 0.59,
      "eval_loss": 0.7476093769073486,
      "eval_runtime": 30.9205,
      "eval_samples_per_second": 264.906,
      "eval_steps_per_second": 16.559,
      "step": 832
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6170209976970482e-05,
      "loss": 0.805,
      "step": 833
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6168171292860665e-05,
      "loss": 0.7763,
      "step": 834
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6166130236273006e-05,
      "loss": 0.7736,
      "step": 835
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6164086807839003e-05,
      "loss": 0.7615,
      "step": 836
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6162041008190877e-05,
      "loss": 0.7947,
      "step": 837
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6159992837961583e-05,
      "loss": 0.7667,
      "step": 838
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6157942297784805e-05,
      "loss": 0.7648,
      "step": 839
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6155889388294976e-05,
      "loss": 0.7443,
      "step": 840
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.615383411012724e-05,
      "loss": 0.7575,
      "step": 841
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6151776463917497e-05,
      "loss": 0.7832,
      "step": 842
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.614971645030236e-05,
      "loss": 0.7752,
      "step": 843
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6147654069919184e-05,
      "loss": 0.7652,
      "step": 844
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6145589323406056e-05,
      "loss": 0.7501,
      "step": 845
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6143522211401794e-05,
      "loss": 0.7765,
      "step": 846
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6141452734545952e-05,
      "loss": 0.8037,
      "step": 847
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6139380893478798e-05,
      "loss": 0.8254,
      "step": 848
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6137306688841364e-05,
      "loss": 0.7546,
      "step": 849
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.613523012127538e-05,
      "loss": 0.7623,
      "step": 850
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6133151191423323e-05,
      "loss": 0.8199,
      "step": 851
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6131069899928407e-05,
      "loss": 0.7816,
      "step": 852
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.612898624743456e-05,
      "loss": 0.7618,
      "step": 853
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6126900234586452e-05,
      "loss": 0.7383,
      "step": 854
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6124811862029485e-05,
      "loss": 0.7794,
      "step": 855
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.612272113040978e-05,
      "loss": 0.7872,
      "step": 856
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.61206280403742e-05,
      "loss": 0.7446,
      "step": 857
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.611853259257033e-05,
      "loss": 0.8022,
      "step": 858
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6116434787646488e-05,
      "loss": 0.778,
      "step": 859
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6114334626251715e-05,
      "loss": 0.7587,
      "step": 860
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6112232109035792e-05,
      "loss": 0.8074,
      "step": 861
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6110127236649222e-05,
      "loss": 0.7602,
      "step": 862
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6108020009743234e-05,
      "loss": 0.7826,
      "step": 863
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.610591042896979e-05,
      "loss": 0.749,
      "step": 864
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6103798494981583e-05,
      "loss": 0.7584,
      "step": 865
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6101684208432025e-05,
      "loss": 0.7598,
      "step": 866
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6099567569975267e-05,
      "loss": 0.7882,
      "step": 867
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6097448580266176e-05,
      "loss": 0.7869,
      "step": 868
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.609532723996036e-05,
      "loss": 0.7727,
      "step": 869
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.609320354971414e-05,
      "loss": 0.7469,
      "step": 870
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6091077510184573e-05,
      "loss": 0.7447,
      "step": 871
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.608894912202944e-05,
      "loss": 0.8083,
      "step": 872
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6086818385907248e-05,
      "loss": 0.7555,
      "step": 873
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.608468530247724e-05,
      "loss": 0.7741,
      "step": 874
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6082549872399363e-05,
      "loss": 0.7673,
      "step": 875
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6080412096334316e-05,
      "loss": 0.8055,
      "step": 876
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6078271974943503e-05,
      "loss": 0.7671,
      "step": 877
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.607612950888907e-05,
      "loss": 0.7828,
      "step": 878
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6073984698833876e-05,
      "loss": 0.7975,
      "step": 879
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.607183754544151e-05,
      "loss": 0.7524,
      "step": 880
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6069688049376285e-05,
      "loss": 0.7725,
      "step": 881
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6067536211303245e-05,
      "loss": 0.8163,
      "step": 882
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.606538203188815e-05,
      "loss": 0.7943,
      "step": 883
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6063225511797487e-05,
      "loss": 0.7687,
      "step": 884
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6061066651698474e-05,
      "loss": 0.7983,
      "step": 885
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6058905452259036e-05,
      "loss": 0.7721,
      "step": 886
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6056741914147844e-05,
      "loss": 0.776,
      "step": 887
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.6054576038034274e-05,
      "loss": 0.7456,
      "step": 888
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.6052407824588436e-05,
      "loss": 0.7937,
      "step": 889
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.6050237274481157e-05,
      "loss": 0.7591,
      "step": 890
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.6048064388383994e-05,
      "loss": 0.777,
      "step": 891
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.6045889166969223e-05,
      "loss": 0.7634,
      "step": 892
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.6043711610909835e-05,
      "loss": 0.7584,
      "step": 893
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.6041531720879557e-05,
      "loss": 0.7533,
      "step": 894
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.603934949755283e-05,
      "loss": 0.7853,
      "step": 895
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.6037164941604817e-05,
      "loss": 0.757,
      "step": 896
    },
    {
      "epoch": 0.63,
      "eval_loss": 0.7455337047576904,
      "eval_runtime": 30.9476,
      "eval_samples_per_second": 264.673,
      "eval_steps_per_second": 16.544,
      "step": 896
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.6034978053711404e-05,
      "loss": 0.7509,
      "step": 897
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.60327888345492e-05,
      "loss": 0.7372,
      "step": 898
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.6030597284795537e-05,
      "loss": 0.7197,
      "step": 899
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.6028403405128454e-05,
      "loss": 0.7599,
      "step": 900
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.6026207196226737e-05,
      "loss": 0.7631,
      "step": 901
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.6024008658769866e-05,
      "loss": 0.788,
      "step": 902
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.6021807793438058e-05,
      "loss": 0.7633,
      "step": 903
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.6019604600912243e-05,
      "loss": 0.7821,
      "step": 904
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.6017399081874073e-05,
      "loss": 0.7674,
      "step": 905
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.601519123700592e-05,
      "loss": 0.7882,
      "step": 906
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.601298106699088e-05,
      "loss": 0.7599,
      "step": 907
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.601076857251276e-05,
      "loss": 0.7737,
      "step": 908
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.6008553754256086e-05,
      "loss": 0.7555,
      "step": 909
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.6006336612906113e-05,
      "loss": 0.7735,
      "step": 910
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.6004117149148812e-05,
      "loss": 0.7844,
      "step": 911
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.600189536367086e-05,
      "loss": 0.781,
      "step": 912
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.599967125715967e-05,
      "loss": 0.7522,
      "step": 913
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.599744483030336e-05,
      "loss": 0.7236,
      "step": 914
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.599521608379077e-05,
      "loss": 0.7585,
      "step": 915
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.599298501831146e-05,
      "loss": 0.781,
      "step": 916
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5990751634555707e-05,
      "loss": 0.7937,
      "step": 917
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5988515933214503e-05,
      "loss": 0.8059,
      "step": 918
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.598627791497956e-05,
      "loss": 0.7841,
      "step": 919
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5984037580543303e-05,
      "loss": 0.7675,
      "step": 920
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5981794930598874e-05,
      "loss": 0.7533,
      "step": 921
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5979549965840135e-05,
      "loss": 0.7644,
      "step": 922
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5977302686961663e-05,
      "loss": 0.7636,
      "step": 923
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5975053094658748e-05,
      "loss": 0.7694,
      "step": 924
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.59728011896274e-05,
      "loss": 0.7565,
      "step": 925
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.597054697256434e-05,
      "loss": 0.7552,
      "step": 926
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5968290444167012e-05,
      "loss": 0.7801,
      "step": 927
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5966031605133562e-05,
      "loss": 0.7776,
      "step": 928
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5963770456162864e-05,
      "loss": 0.7354,
      "step": 929
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5961506997954506e-05,
      "loss": 0.78,
      "step": 930
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5959241231208773e-05,
      "loss": 0.798,
      "step": 931
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5956973156626688e-05,
      "loss": 0.8015,
      "step": 932
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5954702774909973e-05,
      "loss": 0.7432,
      "step": 933
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.595243008676107e-05,
      "loss": 0.7701,
      "step": 934
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.595015509288313e-05,
      "loss": 0.7802,
      "step": 935
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5947877793980022e-05,
      "loss": 0.764,
      "step": 936
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5945598190756326e-05,
      "loss": 0.7863,
      "step": 937
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5943316283917333e-05,
      "loss": 0.7961,
      "step": 938
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.594103207416905e-05,
      "loss": 0.7844,
      "step": 939
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5938745562218197e-05,
      "loss": 0.7635,
      "step": 940
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5936456748772204e-05,
      "loss": 0.7308,
      "step": 941
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5934165634539213e-05,
      "loss": 0.7908,
      "step": 942
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5931872220228074e-05,
      "loss": 0.7505,
      "step": 943
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5929576506548364e-05,
      "loss": 0.7552,
      "step": 944
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.592727849421035e-05,
      "loss": 0.7873,
      "step": 945
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5924978183925026e-05,
      "loss": 0.7705,
      "step": 946
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.592267557640409e-05,
      "loss": 0.7576,
      "step": 947
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5920370672359953e-05,
      "loss": 0.7577,
      "step": 948
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5918063472505737e-05,
      "loss": 0.8096,
      "step": 949
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.591575397755527e-05,
      "loss": 0.7561,
      "step": 950
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5913442188223095e-05,
      "loss": 0.7722,
      "step": 951
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5911128105224468e-05,
      "loss": 0.7907,
      "step": 952
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5908811729275343e-05,
      "loss": 0.789,
      "step": 953
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5906493061092392e-05,
      "loss": 0.7557,
      "step": 954
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5904172101392997e-05,
      "loss": 0.7257,
      "step": 955
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5901848850895243e-05,
      "loss": 0.7936,
      "step": 956
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5899523310317932e-05,
      "loss": 0.7655,
      "step": 957
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5897195480380567e-05,
      "loss": 0.7625,
      "step": 958
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.589486536180336e-05,
      "loss": 0.7975,
      "step": 959
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5892532955307236e-05,
      "loss": 0.7765,
      "step": 960
    },
    {
      "epoch": 0.68,
      "eval_loss": 0.7440967559814453,
      "eval_runtime": 30.9185,
      "eval_samples_per_second": 264.922,
      "eval_steps_per_second": 16.56,
      "step": 960
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5890198261613824e-05,
      "loss": 0.7666,
      "step": 961
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.588786128144546e-05,
      "loss": 0.7521,
      "step": 962
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5885522015525195e-05,
      "loss": 0.7459,
      "step": 963
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5883180464576774e-05,
      "loss": 0.773,
      "step": 964
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.588083662932466e-05,
      "loss": 0.771,
      "step": 965
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.587849051049402e-05,
      "loss": 0.7604,
      "step": 966
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5876142108810723e-05,
      "loss": 0.8089,
      "step": 967
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5873791425001352e-05,
      "loss": 0.7721,
      "step": 968
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.587143845979319e-05,
      "loss": 0.7675,
      "step": 969
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5869083213914224e-05,
      "loss": 0.7474,
      "step": 970
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5866725688093158e-05,
      "loss": 0.7582,
      "step": 971
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5864365883059387e-05,
      "loss": 0.8078,
      "step": 972
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.586200379954302e-05,
      "loss": 0.7705,
      "step": 973
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5859639438274874e-05,
      "loss": 0.7952,
      "step": 974
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.585727279998646e-05,
      "loss": 0.7678,
      "step": 975
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.585490388541e-05,
      "loss": 0.773,
      "step": 976
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5852532695278418e-05,
      "loss": 0.772,
      "step": 977
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5850159230325342e-05,
      "loss": 0.7902,
      "step": 978
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5847783491285114e-05,
      "loss": 0.7701,
      "step": 979
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5845405478892763e-05,
      "loss": 0.7531,
      "step": 980
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.584302519388403e-05,
      "loss": 0.7844,
      "step": 981
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5840642636995363e-05,
      "loss": 0.745,
      "step": 982
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.58382578089639e-05,
      "loss": 0.7814,
      "step": 983
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.58358707105275e-05,
      "loss": 0.737,
      "step": 984
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5833481342424705e-05,
      "loss": 0.7522,
      "step": 985
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5831089705394776e-05,
      "loss": 0.7378,
      "step": 986
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5828695800177662e-05,
      "loss": 0.7564,
      "step": 987
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5826299627514027e-05,
      "loss": 0.7859,
      "step": 988
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.582390118814522e-05,
      "loss": 0.7524,
      "step": 989
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.582150048281331e-05,
      "loss": 0.7643,
      "step": 990
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.581909751226106e-05,
      "loss": 0.7796,
      "step": 991
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5816692277231927e-05,
      "loss": 0.7705,
      "step": 992
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.581428477847007e-05,
      "loss": 0.7766,
      "step": 993
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5811875016720358e-05,
      "loss": 0.7664,
      "step": 994
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5809462992728355e-05,
      "loss": 0.8073,
      "step": 995
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5807048707240318e-05,
      "loss": 0.7645,
      "step": 996
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5804632161003215e-05,
      "loss": 0.7644,
      "step": 997
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5802213354764705e-05,
      "loss": 0.7638,
      "step": 998
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5799792289273155e-05,
      "loss": 0.8022,
      "step": 999
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5797368965277623e-05,
      "loss": 0.7618,
      "step": 1000
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5794943383527863e-05,
      "loss": 0.7446,
      "step": 1001
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.579251554477434e-05,
      "loss": 0.746,
      "step": 1002
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5790085449768202e-05,
      "loss": 0.7866,
      "step": 1003
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.578765309926131e-05,
      "loss": 0.7351,
      "step": 1004
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5785218494006215e-05,
      "loss": 0.7338,
      "step": 1005
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5782781634756166e-05,
      "loss": 0.7978,
      "step": 1006
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5780342522265113e-05,
      "loss": 0.7842,
      "step": 1007
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5777901157287692e-05,
      "loss": 0.7867,
      "step": 1008
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.577545754057925e-05,
      "loss": 0.7876,
      "step": 1009
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5773011672895824e-05,
      "loss": 0.7595,
      "step": 1010
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5770563554994147e-05,
      "loss": 0.7498,
      "step": 1011
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.576811318763165e-05,
      "loss": 0.7598,
      "step": 1012
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.576566057156646e-05,
      "loss": 0.763,
      "step": 1013
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.57632057075574e-05,
      "loss": 0.7833,
      "step": 1014
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.576074859636398e-05,
      "loss": 0.7433,
      "step": 1015
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5758289238746425e-05,
      "loss": 0.7658,
      "step": 1016
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5755827635465635e-05,
      "loss": 0.7856,
      "step": 1017
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5753363787283212e-05,
      "loss": 0.754,
      "step": 1018
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5750897694961453e-05,
      "loss": 0.7534,
      "step": 1019
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5748429359263355e-05,
      "loss": 0.7926,
      "step": 1020
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5745958780952603e-05,
      "loss": 0.7787,
      "step": 1021
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.574348596079357e-05,
      "loss": 0.7716,
      "step": 1022
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5741010899551332e-05,
      "loss": 0.753,
      "step": 1023
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.573853359799166e-05,
      "loss": 0.7735,
      "step": 1024
    },
    {
      "epoch": 0.72,
      "eval_loss": 0.742942214012146,
      "eval_runtime": 30.9441,
      "eval_samples_per_second": 264.703,
      "eval_steps_per_second": 16.546,
      "step": 1024
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5736054056881004e-05,
      "loss": 0.7867,
      "step": 1025
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5733572276986524e-05,
      "loss": 0.7733,
      "step": 1026
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5731088259076065e-05,
      "loss": 0.7832,
      "step": 1027
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.572860200391816e-05,
      "loss": 0.7467,
      "step": 1028
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5726113512282044e-05,
      "loss": 0.7819,
      "step": 1029
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5723622784937636e-05,
      "loss": 0.7235,
      "step": 1030
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5721129822655546e-05,
      "loss": 0.762,
      "step": 1031
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5718634626207086e-05,
      "loss": 0.7589,
      "step": 1032
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5716137196364248e-05,
      "loss": 0.7873,
      "step": 1033
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.571363753389972e-05,
      "loss": 0.7631,
      "step": 1034
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5711135639586876e-05,
      "loss": 0.7769,
      "step": 1035
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5708631514199793e-05,
      "loss": 0.772,
      "step": 1036
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5706125158513218e-05,
      "loss": 0.8047,
      "step": 1037
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.570361657330261e-05,
      "loss": 0.7825,
      "step": 1038
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5701105759344103e-05,
      "loss": 0.7792,
      "step": 1039
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5698592717414525e-05,
      "loss": 0.7853,
      "step": 1040
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5696077448291392e-05,
      "loss": 0.7675,
      "step": 1041
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5693559952752914e-05,
      "loss": 0.7354,
      "step": 1042
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5691040231577987e-05,
      "loss": 0.7547,
      "step": 1043
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.568851828554619e-05,
      "loss": 0.7914,
      "step": 1044
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.56859941154378e-05,
      "loss": 0.7927,
      "step": 1045
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.568346772203377e-05,
      "loss": 0.763,
      "step": 1046
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5680939106115757e-05,
      "loss": 0.742,
      "step": 1047
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5678408268466094e-05,
      "loss": 0.7726,
      "step": 1048
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.56758752098678e-05,
      "loss": 0.7402,
      "step": 1049
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5673339931104594e-05,
      "loss": 0.7699,
      "step": 1050
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5670802432960862e-05,
      "loss": 0.7417,
      "step": 1051
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5668262716221695e-05,
      "loss": 0.7452,
      "step": 1052
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5665720781672866e-05,
      "loss": 0.7699,
      "step": 1053
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5663176630100824e-05,
      "loss": 0.7559,
      "step": 1054
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5660630262292716e-05,
      "loss": 0.8284,
      "step": 1055
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.565808167903637e-05,
      "loss": 0.7614,
      "step": 1056
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.56555308811203e-05,
      "loss": 0.7789,
      "step": 1057
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5652977869333703e-05,
      "loss": 0.7767,
      "step": 1058
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.565042264446646e-05,
      "loss": 0.7932,
      "step": 1059
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5647865207309146e-05,
      "loss": 0.7526,
      "step": 1060
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5645305558653012e-05,
      "loss": 0.7787,
      "step": 1061
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5642743699289994e-05,
      "loss": 0.7982,
      "step": 1062
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5640179630012714e-05,
      "loss": 0.7655,
      "step": 1063
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.563761335161447e-05,
      "loss": 0.7653,
      "step": 1064
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.563504486488926e-05,
      "loss": 0.8052,
      "step": 1065
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5632474170631754e-05,
      "loss": 0.7866,
      "step": 1066
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5629901269637303e-05,
      "loss": 0.7494,
      "step": 1067
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5627326162701944e-05,
      "loss": 0.7973,
      "step": 1068
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5624748850622403e-05,
      "loss": 0.778,
      "step": 1069
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5622169334196073e-05,
      "loss": 0.7512,
      "step": 1070
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5619587614221046e-05,
      "loss": 0.7716,
      "step": 1071
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5617003691496083e-05,
      "loss": 0.7761,
      "step": 1072
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5614417566820636e-05,
      "loss": 0.7751,
      "step": 1073
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.561182924099483e-05,
      "loss": 0.7593,
      "step": 1074
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5609238714819473e-05,
      "loss": 0.8079,
      "step": 1075
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5606645989096057e-05,
      "loss": 0.7579,
      "step": 1076
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.560405106462676e-05,
      "loss": 0.7425,
      "step": 1077
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.560145394221442e-05,
      "loss": 0.7771,
      "step": 1078
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5598854622662578e-05,
      "loss": 0.772,
      "step": 1079
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5596253106775442e-05,
      "loss": 0.7688,
      "step": 1080
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5593649395357903e-05,
      "loss": 0.7784,
      "step": 1081
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.559104348921553e-05,
      "loss": 0.7481,
      "step": 1082
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5588435389154577e-05,
      "loss": 0.7984,
      "step": 1083
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.558582509598196e-05,
      "loss": 0.7681,
      "step": 1084
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5583212610505295e-05,
      "loss": 0.7767,
      "step": 1085
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5580597933532866e-05,
      "loss": 0.7285,
      "step": 1086
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.5577981065873634e-05,
      "loss": 0.7667,
      "step": 1087
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.557536200833724e-05,
      "loss": 0.747,
      "step": 1088
    },
    {
      "epoch": 0.77,
      "eval_loss": 0.7416298389434814,
      "eval_runtime": 30.9371,
      "eval_samples_per_second": 264.763,
      "eval_steps_per_second": 16.55,
      "step": 1088
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.5572740761733995e-05,
      "loss": 0.7755,
      "step": 1089
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.5570117326874905e-05,
      "loss": 0.7874,
      "step": 1090
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.5567491704571636e-05,
      "loss": 0.788,
      "step": 1091
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.5564863895636534e-05,
      "loss": 0.7294,
      "step": 1092
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.556223390088263e-05,
      "loss": 0.7866,
      "step": 1093
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.5559601721123626e-05,
      "loss": 0.7325,
      "step": 1094
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.55569673571739e-05,
      "loss": 0.7685,
      "step": 1095
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.55543308098485e-05,
      "loss": 0.7643,
      "step": 1096
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.5551692079963156e-05,
      "loss": 0.737,
      "step": 1097
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.554905116833428e-05,
      "loss": 0.7611,
      "step": 1098
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.554640807577894e-05,
      "loss": 0.7438,
      "step": 1099
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.55437628031149e-05,
      "loss": 0.7549,
      "step": 1100
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5541115351160585e-05,
      "loss": 0.7641,
      "step": 1101
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5538465720735098e-05,
      "loss": 0.7883,
      "step": 1102
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5535813912658214e-05,
      "loss": 0.7416,
      "step": 1103
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5533159927750386e-05,
      "loss": 0.767,
      "step": 1104
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5530503766832735e-05,
      "loss": 0.7588,
      "step": 1105
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5527845430727062e-05,
      "loss": 0.729,
      "step": 1106
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5525184920255837e-05,
      "loss": 0.7533,
      "step": 1107
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5522522236242198e-05,
      "loss": 0.7515,
      "step": 1108
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5519857379509967e-05,
      "loss": 0.7532,
      "step": 1109
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5517190350883633e-05,
      "loss": 0.7177,
      "step": 1110
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5514521151188347e-05,
      "loss": 0.778,
      "step": 1111
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.551184978124995e-05,
      "loss": 0.7555,
      "step": 1112
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5509176241894935e-05,
      "loss": 0.761,
      "step": 1113
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5506500533950483e-05,
      "loss": 0.76,
      "step": 1114
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.550382265824444e-05,
      "loss": 0.7339,
      "step": 1115
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.550114261560532e-05,
      "loss": 0.7737,
      "step": 1116
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.5498460406862306e-05,
      "loss": 0.7627,
      "step": 1117
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.5495776032845257e-05,
      "loss": 0.7647,
      "step": 1118
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.54930894943847e-05,
      "loss": 0.773,
      "step": 1119
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.5490400792311834e-05,
      "loss": 0.7534,
      "step": 1120
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.5487709927458514e-05,
      "loss": 0.7642,
      "step": 1121
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.5485016900657284e-05,
      "loss": 0.8099,
      "step": 1122
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.5482321712741345e-05,
      "loss": 0.82,
      "step": 1123
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.547962436454457e-05,
      "loss": 0.7395,
      "step": 1124
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.5476924856901494e-05,
      "loss": 0.7895,
      "step": 1125
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.547422319064733e-05,
      "loss": 0.7751,
      "step": 1126
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.5471519366617955e-05,
      "loss": 0.7719,
      "step": 1127
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.5468813385649912e-05,
      "loss": 0.7486,
      "step": 1128
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.5466105248580414e-05,
      "loss": 0.7448,
      "step": 1129
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.5463394956247334e-05,
      "loss": 0.7438,
      "step": 1130
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.5460682509489223e-05,
      "loss": 0.7407,
      "step": 1131
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.545796790914529e-05,
      "loss": 0.7951,
      "step": 1132
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.5455251156055412e-05,
      "loss": 0.7281,
      "step": 1133
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.5452532251060138e-05,
      "loss": 0.7548,
      "step": 1134
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.5449811195000675e-05,
      "loss": 0.7778,
      "step": 1135
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.5447087988718894e-05,
      "loss": 0.7694,
      "step": 1136
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.544436263305734e-05,
      "loss": 0.7565,
      "step": 1137
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.5441635128859224e-05,
      "loss": 0.7659,
      "step": 1138
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.543890547696841e-05,
      "loss": 0.7681,
      "step": 1139
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.543617367822943e-05,
      "loss": 0.7886,
      "step": 1140
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.543343973348749e-05,
      "loss": 0.7674,
      "step": 1141
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.5430703643588447e-05,
      "loss": 0.7683,
      "step": 1142
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.5427965409378836e-05,
      "loss": 0.7911,
      "step": 1143
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.542522503170584e-05,
      "loss": 0.757,
      "step": 1144
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.5422482511417315e-05,
      "loss": 0.7759,
      "step": 1145
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.5419737849361778e-05,
      "loss": 0.7625,
      "step": 1146
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.5416991046388407e-05,
      "loss": 0.7591,
      "step": 1147
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.5414242103347046e-05,
      "loss": 0.7727,
      "step": 1148
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.541149102108819e-05,
      "loss": 0.7499,
      "step": 1149
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.5408737800463013e-05,
      "loss": 0.7472,
      "step": 1150
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.540598244232334e-05,
      "loss": 0.7735,
      "step": 1151
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.540322494752166e-05,
      "loss": 0.7658,
      "step": 1152
    },
    {
      "epoch": 0.81,
      "eval_loss": 0.7395820617675781,
      "eval_runtime": 30.9635,
      "eval_samples_per_second": 264.538,
      "eval_steps_per_second": 16.536,
      "step": 1152
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.5400465316911116e-05,
      "loss": 0.779,
      "step": 1153
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.5397703551345527e-05,
      "loss": 0.785,
      "step": 1154
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.539493965167936e-05,
      "loss": 0.7555,
      "step": 1155
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.5392173618767747e-05,
      "loss": 0.7693,
      "step": 1156
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.5389405453466474e-05,
      "loss": 0.7795,
      "step": 1157
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.5386635156631996e-05,
      "loss": 0.7757,
      "step": 1158
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.5383862729121424e-05,
      "loss": 0.7431,
      "step": 1159
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.5381088171792523e-05,
      "loss": 0.7735,
      "step": 1160
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.537831148550373e-05,
      "loss": 0.7583,
      "step": 1161
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.537553267111412e-05,
      "loss": 0.7359,
      "step": 1162
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.537275172948345e-05,
      "loss": 0.7678,
      "step": 1163
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.5369968661472117e-05,
      "loss": 0.7621,
      "step": 1164
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.5367183467941184e-05,
      "loss": 0.7466,
      "step": 1165
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.536439614975237e-05,
      "loss": 0.7888,
      "step": 1166
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.536160670776805e-05,
      "loss": 0.7436,
      "step": 1167
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.5358815142851263e-05,
      "loss": 0.7564,
      "step": 1168
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.5356021455865693e-05,
      "loss": 0.7707,
      "step": 1169
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.5353225647675692e-05,
      "loss": 0.7599,
      "step": 1170
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.5350427719146258e-05,
      "loss": 0.8026,
      "step": 1171
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.534762767114305e-05,
      "loss": 0.7614,
      "step": 1172
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.534482550453239e-05,
      "loss": 0.7996,
      "step": 1173
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.5342021220181248e-05,
      "loss": 0.783,
      "step": 1174
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.5339214818957243e-05,
      "loss": 0.7498,
      "step": 1175
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.5336406301728658e-05,
      "loss": 0.7557,
      "step": 1176
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.533359566936443e-05,
      "loss": 0.7639,
      "step": 1177
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.533078292273415e-05,
      "loss": 0.7866,
      "step": 1178
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.5327968062708057e-05,
      "loss": 0.764,
      "step": 1179
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.532515109015705e-05,
      "loss": 0.7647,
      "step": 1180
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.5322332005952685e-05,
      "loss": 0.7836,
      "step": 1181
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.5319510810967164e-05,
      "loss": 0.7727,
      "step": 1182
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.5316687506073342e-05,
      "loss": 0.7604,
      "step": 1183
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.5313862092144732e-05,
      "loss": 0.7265,
      "step": 1184
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.53110345700555e-05,
      "loss": 0.7905,
      "step": 1185
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.5308204940680454e-05,
      "loss": 0.7744,
      "step": 1186
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.5305373204895068e-05,
      "loss": 0.7764,
      "step": 1187
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.5302539363575457e-05,
      "loss": 0.7559,
      "step": 1188
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.5299703417598393e-05,
      "loss": 0.746,
      "step": 1189
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.52968653678413e-05,
      "loss": 0.7728,
      "step": 1190
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.5294025215182243e-05,
      "loss": 0.7596,
      "step": 1191
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.5291182960499954e-05,
      "loss": 0.773,
      "step": 1192
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.5288338604673803e-05,
      "loss": 0.7558,
      "step": 1193
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.528549214858381e-05,
      "loss": 0.7591,
      "step": 1194
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.528264359311065e-05,
      "loss": 0.7673,
      "step": 1195
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.527979293913565e-05,
      "loss": 0.7968,
      "step": 1196
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.527694018754078e-05,
      "loss": 0.7719,
      "step": 1197
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.5274085339208658e-05,
      "loss": 0.8009,
      "step": 1198
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.5271228395022557e-05,
      "loss": 0.7702,
      "step": 1199
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5268369355866395e-05,
      "loss": 0.7498,
      "step": 1200
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5265508222624738e-05,
      "loss": 0.7848,
      "step": 1201
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.52626449961828e-05,
      "loss": 0.7422,
      "step": 1202
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5259779677426443e-05,
      "loss": 0.7999,
      "step": 1203
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5256912267242175e-05,
      "loss": 0.7742,
      "step": 1204
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5254042766517156e-05,
      "loss": 0.747,
      "step": 1205
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5251171176139188e-05,
      "loss": 0.7729,
      "step": 1206
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5248297496996717e-05,
      "loss": 0.7563,
      "step": 1207
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5245421729978842e-05,
      "loss": 0.7896,
      "step": 1208
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5242543875975307e-05,
      "loss": 0.8003,
      "step": 1209
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5239663935876496e-05,
      "loss": 0.7586,
      "step": 1210
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5236781910573443e-05,
      "loss": 0.7826,
      "step": 1211
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5233897800957828e-05,
      "loss": 0.7683,
      "step": 1212
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5231011607921969e-05,
      "loss": 0.7809,
      "step": 1213
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5228123332358842e-05,
      "loss": 0.744,
      "step": 1214
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5225232975162053e-05,
      "loss": 0.7516,
      "step": 1215
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5222340537225859e-05,
      "loss": 0.7677,
      "step": 1216
    },
    {
      "epoch": 0.86,
      "eval_loss": 0.738717794418335,
      "eval_runtime": 30.9352,
      "eval_samples_per_second": 264.78,
      "eval_steps_per_second": 16.551,
      "step": 1216
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5219446019445159e-05,
      "loss": 0.7702,
      "step": 1217
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5216549422715499e-05,
      "loss": 0.7665,
      "step": 1218
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5213650747933062e-05,
      "loss": 0.7467,
      "step": 1219
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.521074999599468e-05,
      "loss": 0.7457,
      "step": 1220
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5207847167797825e-05,
      "loss": 0.7717,
      "step": 1221
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.520494226424061e-05,
      "loss": 0.7712,
      "step": 1222
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5202035286221795e-05,
      "loss": 0.7787,
      "step": 1223
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5199126234640772e-05,
      "loss": 0.8167,
      "step": 1224
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5196215110397587e-05,
      "loss": 0.7486,
      "step": 1225
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5193301914392917e-05,
      "loss": 0.7692,
      "step": 1226
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5190386647528088e-05,
      "loss": 0.7705,
      "step": 1227
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5187469310705061e-05,
      "loss": 0.7465,
      "step": 1228
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5184549904826438e-05,
      "loss": 0.7693,
      "step": 1229
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5181628430795465e-05,
      "loss": 0.7285,
      "step": 1230
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5178704889516025e-05,
      "loss": 0.7281,
      "step": 1231
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.517577928189264e-05,
      "loss": 0.7592,
      "step": 1232
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.517285160883047e-05,
      "loss": 0.7823,
      "step": 1233
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5169921871235322e-05,
      "loss": 0.7727,
      "step": 1234
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5166990070013631e-05,
      "loss": 0.79,
      "step": 1235
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5164056206072476e-05,
      "loss": 0.7492,
      "step": 1236
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5161120280319577e-05,
      "loss": 0.7551,
      "step": 1237
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5158182293663286e-05,
      "loss": 0.7599,
      "step": 1238
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5155242247012595e-05,
      "loss": 0.7615,
      "step": 1239
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5152300141277134e-05,
      "loss": 0.7595,
      "step": 1240
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5149355977367174e-05,
      "loss": 0.7663,
      "step": 1241
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5146409756193611e-05,
      "loss": 0.7659,
      "step": 1242
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.5143461478667988e-05,
      "loss": 0.7832,
      "step": 1243
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.5140511145702483e-05,
      "loss": 0.7669,
      "step": 1244
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.5137558758209905e-05,
      "loss": 0.7727,
      "step": 1245
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.5134604317103706e-05,
      "loss": 0.8,
      "step": 1246
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.5131647823297965e-05,
      "loss": 0.7773,
      "step": 1247
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.5128689277707403e-05,
      "loss": 0.7651,
      "step": 1248
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.5125728681247371e-05,
      "loss": 0.7126,
      "step": 1249
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.512276603483386e-05,
      "loss": 0.789,
      "step": 1250
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.5119801339383484e-05,
      "loss": 0.7547,
      "step": 1251
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.5116834595813507e-05,
      "loss": 0.7286,
      "step": 1252
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.5113865805041816e-05,
      "loss": 0.7601,
      "step": 1253
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.5110894967986933e-05,
      "loss": 0.7781,
      "step": 1254
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.5107922085568014e-05,
      "loss": 0.7552,
      "step": 1255
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.5104947158704847e-05,
      "loss": 0.7383,
      "step": 1256
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.5101970188317854e-05,
      "loss": 0.7656,
      "step": 1257
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.5098991175328091e-05,
      "loss": 0.7352,
      "step": 1258
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.509601012065724e-05,
      "loss": 0.8043,
      "step": 1259
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.5093027025227616e-05,
      "loss": 0.7615,
      "step": 1260
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.5090041889962173e-05,
      "loss": 0.7699,
      "step": 1261
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.5087054715784492e-05,
      "loss": 0.7376,
      "step": 1262
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.5084065503618776e-05,
      "loss": 0.7345,
      "step": 1263
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.5081074254389872e-05,
      "loss": 0.7755,
      "step": 1264
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.5078080969023251e-05,
      "loss": 0.762,
      "step": 1265
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.5075085648445009e-05,
      "loss": 0.7179,
      "step": 1266
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.5072088293581885e-05,
      "loss": 0.7536,
      "step": 1267
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.506908890536123e-05,
      "loss": 0.7626,
      "step": 1268
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.5066087484711044e-05,
      "loss": 0.7806,
      "step": 1269
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.5063084032559937e-05,
      "loss": 0.7394,
      "step": 1270
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.506007854983716e-05,
      "loss": 0.7842,
      "step": 1271
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.5057071037472583e-05,
      "loss": 0.7505,
      "step": 1272
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.5054061496396717e-05,
      "loss": 0.7538,
      "step": 1273
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.5051049927540686e-05,
      "loss": 0.7851,
      "step": 1274
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.5048036331836252e-05,
      "loss": 0.7584,
      "step": 1275
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.5045020710215797e-05,
      "loss": 0.7652,
      "step": 1276
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.5042003063612336e-05,
      "loss": 0.7776,
      "step": 1277
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.5038983392959506e-05,
      "loss": 0.7568,
      "step": 1278
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.5035961699191569e-05,
      "loss": 0.7909,
      "step": 1279
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.5032937983243419e-05,
      "loss": 0.7391,
      "step": 1280
    },
    {
      "epoch": 0.9,
      "eval_loss": 0.7373324036598206,
      "eval_runtime": 30.921,
      "eval_samples_per_second": 264.901,
      "eval_steps_per_second": 16.558,
      "step": 1280
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.5029912246050573e-05,
      "loss": 0.7359,
      "step": 1281
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.5026884488549171e-05,
      "loss": 0.7783,
      "step": 1282
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.502385471167598e-05,
      "loss": 0.7528,
      "step": 1283
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.5020822916368392e-05,
      "loss": 0.7926,
      "step": 1284
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.501778910356442e-05,
      "loss": 0.7608,
      "step": 1285
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.5014753274202707e-05,
      "loss": 0.7816,
      "step": 1286
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.5011715429222518e-05,
      "loss": 0.779,
      "step": 1287
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.500867556956374e-05,
      "loss": 0.7397,
      "step": 1288
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.500563369616688e-05,
      "loss": 0.7222,
      "step": 1289
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.5002589809973078e-05,
      "loss": 0.7781,
      "step": 1290
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4999543911924087e-05,
      "loss": 0.7795,
      "step": 1291
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4996496002962288e-05,
      "loss": 0.7484,
      "step": 1292
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4993446084030681e-05,
      "loss": 0.776,
      "step": 1293
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4990394156072893e-05,
      "loss": 0.7705,
      "step": 1294
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4987340220033163e-05,
      "loss": 0.8082,
      "step": 1295
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4984284276856364e-05,
      "loss": 0.7454,
      "step": 1296
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4981226327487975e-05,
      "loss": 0.7608,
      "step": 1297
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4978166372874112e-05,
      "loss": 0.7672,
      "step": 1298
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4975104413961496e-05,
      "loss": 0.736,
      "step": 1299
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.4972040451697481e-05,
      "loss": 0.7682,
      "step": 1300
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.4968974487030035e-05,
      "loss": 0.7344,
      "step": 1301
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.496590652090774e-05,
      "loss": 0.7422,
      "step": 1302
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.496283655427981e-05,
      "loss": 0.8123,
      "step": 1303
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.4959764588096063e-05,
      "loss": 0.7561,
      "step": 1304
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.4956690623306953e-05,
      "loss": 0.7487,
      "step": 1305
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.4953614660863535e-05,
      "loss": 0.7716,
      "step": 1306
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.4950536701717495e-05,
      "loss": 0.759,
      "step": 1307
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.494745674682113e-05,
      "loss": 0.7785,
      "step": 1308
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.4944374797127356e-05,
      "loss": 0.7535,
      "step": 1309
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.4941290853589705e-05,
      "loss": 0.7512,
      "step": 1310
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.4938204917162332e-05,
      "loss": 0.7594,
      "step": 1311
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.4935116988799997e-05,
      "loss": 0.7868,
      "step": 1312
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.4932027069458087e-05,
      "loss": 0.7503,
      "step": 1313
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.4928935160092603e-05,
      "loss": 0.7777,
      "step": 1314
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.4925841261660155e-05,
      "loss": 0.7714,
      "step": 1315
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.492274537511798e-05,
      "loss": 0.7192,
      "step": 1316
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.4919647501423917e-05,
      "loss": 0.7291,
      "step": 1317
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.4916547641536427e-05,
      "loss": 0.7434,
      "step": 1318
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.4913445796414587e-05,
      "loss": 0.7598,
      "step": 1319
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.4910341967018087e-05,
      "loss": 0.7504,
      "step": 1320
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.4907236154307226e-05,
      "loss": 0.7857,
      "step": 1321
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.4904128359242924e-05,
      "loss": 0.7863,
      "step": 1322
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.4901018582786706e-05,
      "loss": 0.7803,
      "step": 1323
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.4897906825900719e-05,
      "loss": 0.7892,
      "step": 1324
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.489479308954772e-05,
      "loss": 0.7488,
      "step": 1325
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.489167737469107e-05,
      "loss": 0.7601,
      "step": 1326
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.4888559682294756e-05,
      "loss": 0.7757,
      "step": 1327
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.4885440013323368e-05,
      "loss": 0.7887,
      "step": 1328
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.4882318368742108e-05,
      "loss": 0.7453,
      "step": 1329
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.487919474951679e-05,
      "loss": 0.7614,
      "step": 1330
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.4876069156613844e-05,
      "loss": 0.7669,
      "step": 1331
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.48729415910003e-05,
      "loss": 0.7778,
      "step": 1332
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.486981205364381e-05,
      "loss": 0.7609,
      "step": 1333
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.4866680545512628e-05,
      "loss": 0.7635,
      "step": 1334
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.4863547067575622e-05,
      "loss": 0.7776,
      "step": 1335
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.4860411620802268e-05,
      "loss": 0.7174,
      "step": 1336
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.4857274206162647e-05,
      "loss": 0.7696,
      "step": 1337
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.4854134824627457e-05,
      "loss": 0.7707,
      "step": 1338
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.4850993477168001e-05,
      "loss": 0.7602,
      "step": 1339
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.4847850164756188e-05,
      "loss": 0.7899,
      "step": 1340
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.4844704888364533e-05,
      "loss": 0.7696,
      "step": 1341
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.4841557648966167e-05,
      "loss": 0.7663,
      "step": 1342
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.4838408447534823e-05,
      "loss": 0.7569,
      "step": 1343
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.4835257285044841e-05,
      "loss": 0.7727,
      "step": 1344
    },
    {
      "epoch": 0.95,
      "eval_loss": 0.7359663248062134,
      "eval_runtime": 30.9431,
      "eval_samples_per_second": 264.712,
      "eval_steps_per_second": 16.547,
      "step": 1344
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.4832104162471167e-05,
      "loss": 0.7419,
      "step": 1345
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.4828949080789356e-05,
      "loss": 0.7498,
      "step": 1346
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.4825792040975569e-05,
      "loss": 0.7605,
      "step": 1347
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.482263304400657e-05,
      "loss": 0.7753,
      "step": 1348
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.4819472090859731e-05,
      "loss": 0.7625,
      "step": 1349
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.4816309182513027e-05,
      "loss": 0.7535,
      "step": 1350
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.4813144319945043e-05,
      "loss": 0.7564,
      "step": 1351
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.4809977504134961e-05,
      "loss": 0.7901,
      "step": 1352
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.4806808736062575e-05,
      "loss": 0.7733,
      "step": 1353
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.4803638016708276e-05,
      "loss": 0.7462,
      "step": 1354
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.4800465347053066e-05,
      "loss": 0.7648,
      "step": 1355
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.4797290728078541e-05,
      "loss": 0.779,
      "step": 1356
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.4794114160766909e-05,
      "loss": 0.7444,
      "step": 1357
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.4790935646100977e-05,
      "loss": 0.7897,
      "step": 1358
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.4787755185064156e-05,
      "loss": 0.7708,
      "step": 1359
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.4784572778640457e-05,
      "loss": 0.7691,
      "step": 1360
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.4781388427814494e-05,
      "loss": 0.7561,
      "step": 1361
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.477820213357148e-05,
      "loss": 0.7821,
      "step": 1362
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.477501389689724e-05,
      "loss": 0.7608,
      "step": 1363
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.4771823718778182e-05,
      "loss": 0.7276,
      "step": 1364
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.476863160020133e-05,
      "loss": 0.7703,
      "step": 1365
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.4765437542154304e-05,
      "loss": 0.77,
      "step": 1366
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.4762241545625319e-05,
      "loss": 0.7472,
      "step": 1367
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.4759043611603195e-05,
      "loss": 0.7811,
      "step": 1368
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.475584374107735e-05,
      "loss": 0.7864,
      "step": 1369
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.4752641935037802e-05,
      "loss": 0.758,
      "step": 1370
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4749438194475167e-05,
      "loss": 0.7536,
      "step": 1371
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.474623252038066e-05,
      "loss": 0.7913,
      "step": 1372
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.474302491374609e-05,
      "loss": 0.7833,
      "step": 1373
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4739815375563875e-05,
      "loss": 0.7722,
      "step": 1374
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4736603906827016e-05,
      "loss": 0.7747,
      "step": 1375
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4733390508529125e-05,
      "loss": 0.7522,
      "step": 1376
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.47301751816644e-05,
      "loss": 0.7673,
      "step": 1377
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4726957927227642e-05,
      "loss": 0.7666,
      "step": 1378
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4723738746214248e-05,
      "loss": 0.7663,
      "step": 1379
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4720517639620205e-05,
      "loss": 0.7726,
      "step": 1380
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4717294608442105e-05,
      "loss": 0.767,
      "step": 1381
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4714069653677134e-05,
      "loss": 0.7966,
      "step": 1382
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4710842776323063e-05,
      "loss": 0.7286,
      "step": 1383
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4707613977378269e-05,
      "loss": 0.7714,
      "step": 1384
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.4704383257841716e-05,
      "loss": 0.7653,
      "step": 1385
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.470115061871297e-05,
      "loss": 0.7597,
      "step": 1386
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.4697916060992187e-05,
      "loss": 0.7757,
      "step": 1387
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.4694679585680112e-05,
      "loss": 0.7752,
      "step": 1388
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.469144119377809e-05,
      "loss": 0.7614,
      "step": 1389
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.4688200886288058e-05,
      "loss": 0.7214,
      "step": 1390
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.468495866421254e-05,
      "loss": 0.7404,
      "step": 1391
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.4681714528554658e-05,
      "loss": 0.7835,
      "step": 1392
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.4678468480318131e-05,
      "loss": 0.7848,
      "step": 1393
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.4675220520507251e-05,
      "loss": 0.7783,
      "step": 1394
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.4671970650126924e-05,
      "loss": 0.763,
      "step": 1395
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.4668718870182635e-05,
      "loss": 0.7758,
      "step": 1396
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.4665465181680457e-05,
      "loss": 0.7603,
      "step": 1397
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.4662209585627063e-05,
      "loss": 0.7698,
      "step": 1398
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4658952083029707e-05,
      "loss": 0.7861,
      "step": 1399
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4655692674896241e-05,
      "loss": 0.7451,
      "step": 1400
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4652431362235102e-05,
      "loss": 0.7685,
      "step": 1401
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4649168146055319e-05,
      "loss": 0.7708,
      "step": 1402
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4645903027366501e-05,
      "loss": 0.7073,
      "step": 1403
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4642636007178858e-05,
      "loss": 0.7571,
      "step": 1404
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4639367086503183e-05,
      "loss": 0.7349,
      "step": 1405
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4636096266350856e-05,
      "loss": 0.781,
      "step": 1406
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4632823547733846e-05,
      "loss": 0.7294,
      "step": 1407
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4629548931664709e-05,
      "loss": 0.759,
      "step": 1408
    },
    {
      "epoch": 0.99,
      "eval_loss": 0.735153079032898,
      "eval_runtime": 30.9362,
      "eval_samples_per_second": 264.771,
      "eval_steps_per_second": 16.55,
      "step": 1408
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4626272419156584e-05,
      "loss": 0.7635,
      "step": 1409
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4622994011223208e-05,
      "loss": 0.7478,
      "step": 1410
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4619713708878891e-05,
      "loss": 0.7978,
      "step": 1411
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4616431513138537e-05,
      "loss": 0.7666,
      "step": 1412
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.4613147425017636e-05,
      "loss": 0.7677,
      "step": 1413
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.460986144553226e-05,
      "loss": 0.7751,
      "step": 1414
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.4606573575699068e-05,
      "loss": 0.76,
      "step": 1415
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.4603283816535302e-05,
      "loss": 0.721,
      "step": 1416
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.4599992169058793e-05,
      "loss": 0.7425,
      "step": 1417
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.459669863428795e-05,
      "loss": 0.7805,
      "step": 1418
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.4593403213241776e-05,
      "loss": 0.7788,
      "step": 1419
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.4590105906939842e-05,
      "loss": 0.7557,
      "step": 1420
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.4586806716402319e-05,
      "loss": 0.7327,
      "step": 1421
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.4583505642649947e-05,
      "loss": 0.7174,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.4580202686704056e-05,
      "loss": 0.7208,
      "step": 1423
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.4576897849586563e-05,
      "loss": 0.7335,
      "step": 1424
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.4573591132319958e-05,
      "loss": 0.7114,
      "step": 1425
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.457028253592731e-05,
      "loss": 0.7394,
      "step": 1426
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.4566972061432282e-05,
      "loss": 0.7307,
      "step": 1427
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.4563659709859108e-05,
      "loss": 0.744,
      "step": 1428
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.456034548223261e-05,
      "loss": 0.7848,
      "step": 1429
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.4557029379578182e-05,
      "loss": 0.7261,
      "step": 1430
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.4553711402921805e-05,
      "loss": 0.7453,
      "step": 1431
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.4550391553290038e-05,
      "loss": 0.7379,
      "step": 1432
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.4547069831710016e-05,
      "loss": 0.7098,
      "step": 1433
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.4543746239209462e-05,
      "loss": 0.7226,
      "step": 1434
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.4540420776816666e-05,
      "loss": 0.7414,
      "step": 1435
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.4537093445560502e-05,
      "loss": 0.7357,
      "step": 1436
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.4533764246470427e-05,
      "loss": 0.7199,
      "step": 1437
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.453043318057647e-05,
      "loss": 0.7417,
      "step": 1438
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.452710024890924e-05,
      "loss": 0.7442,
      "step": 1439
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.452376545249992e-05,
      "loss": 0.7039,
      "step": 1440
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.4520428792380275e-05,
      "loss": 0.7179,
      "step": 1441
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.4517090269582641e-05,
      "loss": 0.7088,
      "step": 1442
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.451374988513994e-05,
      "loss": 0.7108,
      "step": 1443
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.4510407640085653e-05,
      "loss": 0.7007,
      "step": 1444
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.4507063535453854e-05,
      "loss": 0.7475,
      "step": 1445
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.4503717572279184e-05,
      "loss": 0.739,
      "step": 1446
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.4500369751596862e-05,
      "loss": 0.7151,
      "step": 1447
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.4497020074442676e-05,
      "loss": 0.7338,
      "step": 1448
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.4493668541852993e-05,
      "loss": 0.751,
      "step": 1449
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.4490315154864753e-05,
      "loss": 0.6926,
      "step": 1450
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.4486959914515474e-05,
      "loss": 0.7324,
      "step": 1451
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.4483602821843241e-05,
      "loss": 0.7397,
      "step": 1452
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.4480243877886717e-05,
      "loss": 0.7404,
      "step": 1453
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.4476883083685133e-05,
      "loss": 0.732,
      "step": 1454
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.4473520440278294e-05,
      "loss": 0.7019,
      "step": 1455
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.4470155948706579e-05,
      "loss": 0.7366,
      "step": 1456
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.446678961001094e-05,
      "loss": 0.7292,
      "step": 1457
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.4463421425232896e-05,
      "loss": 0.7072,
      "step": 1458
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.446005139541454e-05,
      "loss": 0.6921,
      "step": 1459
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.4456679521598534e-05,
      "loss": 0.7461,
      "step": 1460
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.4453305804828114e-05,
      "loss": 0.7036,
      "step": 1461
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.4449930246147084e-05,
      "loss": 0.7495,
      "step": 1462
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.4446552846599817e-05,
      "loss": 0.7303,
      "step": 1463
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.4443173607231257e-05,
      "loss": 0.7273,
      "step": 1464
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.4439792529086918e-05,
      "loss": 0.6973,
      "step": 1465
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.4436409613212876e-05,
      "loss": 0.7949,
      "step": 1466
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.4433024860655788e-05,
      "loss": 0.7172,
      "step": 1467
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.442963827246287e-05,
      "loss": 0.7273,
      "step": 1468
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.4426249849681909e-05,
      "loss": 0.7262,
      "step": 1469
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.4422859593361253e-05,
      "loss": 0.7208,
      "step": 1470
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.4419467504549832e-05,
      "loss": 0.7337,
      "step": 1471
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.441607358429713e-05,
      "loss": 0.7359,
      "step": 1472
    },
    {
      "epoch": 1.04,
      "eval_loss": 0.7353258728981018,
      "eval_runtime": 30.95,
      "eval_samples_per_second": 264.653,
      "eval_steps_per_second": 16.543,
      "step": 1472
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.4412677833653202e-05,
      "loss": 0.7436,
      "step": 1473
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.440928025366867e-05,
      "loss": 0.754,
      "step": 1474
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.440588084539472e-05,
      "loss": 0.6811,
      "step": 1475
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.4402479609883106e-05,
      "loss": 0.709,
      "step": 1476
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.4399076548186147e-05,
      "loss": 0.7369,
      "step": 1477
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.439567166135672e-05,
      "loss": 0.7522,
      "step": 1478
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.4392264950448283e-05,
      "loss": 0.7426,
      "step": 1479
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.4388856416514843e-05,
      "loss": 0.7415,
      "step": 1480
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.4385446060610974e-05,
      "loss": 0.7156,
      "step": 1481
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.4382033883791818e-05,
      "loss": 0.7183,
      "step": 1482
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.437861988711308e-05,
      "loss": 0.7384,
      "step": 1483
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4375204071631025e-05,
      "loss": 0.7236,
      "step": 1484
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4371786438402478e-05,
      "loss": 0.7195,
      "step": 1485
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4368366988484839e-05,
      "loss": 0.7291,
      "step": 1486
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4364945722936054e-05,
      "loss": 0.7441,
      "step": 1487
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4361522642814643e-05,
      "loss": 0.7025,
      "step": 1488
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4358097749179679e-05,
      "loss": 0.7582,
      "step": 1489
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4354671043090803e-05,
      "loss": 0.7363,
      "step": 1490
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4351242525608212e-05,
      "loss": 0.7506,
      "step": 1491
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4347812197792667e-05,
      "loss": 0.7017,
      "step": 1492
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4344380060705484e-05,
      "loss": 0.7561,
      "step": 1493
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4340946115408542e-05,
      "loss": 0.7477,
      "step": 1494
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4337510362964282e-05,
      "loss": 0.7315,
      "step": 1495
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4334072804435702e-05,
      "loss": 0.7204,
      "step": 1496
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4330633440886355e-05,
      "loss": 0.7299,
      "step": 1497
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.4327192273380355e-05,
      "loss": 0.7064,
      "step": 1498
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.4323749302982378e-05,
      "loss": 0.7173,
      "step": 1499
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.4320304530757655e-05,
      "loss": 0.7515,
      "step": 1500
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.431685795777197e-05,
      "loss": 0.7349,
      "step": 1501
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.4313409585091675e-05,
      "loss": 0.7716,
      "step": 1502
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.4309959413783665e-05,
      "loss": 0.7732,
      "step": 1503
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.43065074449154e-05,
      "loss": 0.725,
      "step": 1504
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.4303053679554898e-05,
      "loss": 0.7292,
      "step": 1505
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.4299598118770725e-05,
      "loss": 0.717,
      "step": 1506
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.4296140763632014e-05,
      "loss": 0.7456,
      "step": 1507
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.4292681615208442e-05,
      "loss": 0.7466,
      "step": 1508
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.4289220674570243e-05,
      "loss": 0.7491,
      "step": 1509
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.4285757942788208e-05,
      "loss": 0.7339,
      "step": 1510
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.4282293420933683e-05,
      "loss": 0.7174,
      "step": 1511
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.4278827110078569e-05,
      "loss": 0.7415,
      "step": 1512
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.4275359011295315e-05,
      "loss": 0.7452,
      "step": 1513
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.4271889125656926e-05,
      "loss": 0.7613,
      "step": 1514
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.426841745423696e-05,
      "loss": 0.7451,
      "step": 1515
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.4264943998109529e-05,
      "loss": 0.7146,
      "step": 1516
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.4261468758349295e-05,
      "loss": 0.7496,
      "step": 1517
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.4257991736031473e-05,
      "loss": 0.7441,
      "step": 1518
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.4254512932231827e-05,
      "loss": 0.7582,
      "step": 1519
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.4251032348026677e-05,
      "loss": 0.7391,
      "step": 1520
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.424754998449289e-05,
      "loss": 0.7421,
      "step": 1521
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.4244065842707884e-05,
      "loss": 0.7364,
      "step": 1522
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.4240579923749629e-05,
      "loss": 0.7372,
      "step": 1523
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.4237092228696642e-05,
      "loss": 0.7741,
      "step": 1524
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.4233602758627992e-05,
      "loss": 0.7445,
      "step": 1525
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.4230111514623298e-05,
      "loss": 0.7362,
      "step": 1526
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.4226618497762721e-05,
      "loss": 0.7242,
      "step": 1527
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.422312370912698e-05,
      "loss": 0.7406,
      "step": 1528
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.4219627149797337e-05,
      "loss": 0.7435,
      "step": 1529
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.4216128820855607e-05,
      "loss": 0.6772,
      "step": 1530
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.4212628723384138e-05,
      "loss": 0.7126,
      "step": 1531
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.4209126858465843e-05,
      "loss": 0.7453,
      "step": 1532
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.4205623227184172e-05,
      "loss": 0.7098,
      "step": 1533
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.4202117830623124e-05,
      "loss": 0.7072,
      "step": 1534
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.4198610669867242e-05,
      "loss": 0.7344,
      "step": 1535
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.419510174600162e-05,
      "loss": 0.741,
      "step": 1536
    },
    {
      "epoch": 1.08,
      "eval_loss": 0.7347473502159119,
      "eval_runtime": 30.9203,
      "eval_samples_per_second": 264.907,
      "eval_steps_per_second": 16.559,
      "step": 1536
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.4191591060111894e-05,
      "loss": 0.7432,
      "step": 1537
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.4188078613284243e-05,
      "loss": 0.7425,
      "step": 1538
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.4184564406605393e-05,
      "loss": 0.7444,
      "step": 1539
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.4181048441162618e-05,
      "loss": 0.7267,
      "step": 1540
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.4177530718043729e-05,
      "loss": 0.724,
      "step": 1541
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.4174011238337086e-05,
      "loss": 0.786,
      "step": 1542
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.4170490003131591e-05,
      "loss": 0.74,
      "step": 1543
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.4166967013516689e-05,
      "loss": 0.7049,
      "step": 1544
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.416344227058237e-05,
      "loss": 0.7558,
      "step": 1545
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.4159915775419163e-05,
      "loss": 0.7307,
      "step": 1546
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.415638752911814e-05,
      "loss": 0.7255,
      "step": 1547
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.4152857532770914e-05,
      "loss": 0.7384,
      "step": 1548
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.4149325787469643e-05,
      "loss": 0.7087,
      "step": 1549
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.4145792294307026e-05,
      "loss": 0.7016,
      "step": 1550
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.4142257054376297e-05,
      "loss": 0.7153,
      "step": 1551
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.4138720068771236e-05,
      "loss": 0.7006,
      "step": 1552
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.4135181338586158e-05,
      "loss": 0.7475,
      "step": 1553
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.4131640864915928e-05,
      "loss": 0.7475,
      "step": 1554
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.4128098648855936e-05,
      "loss": 0.7439,
      "step": 1555
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.4124554691502126e-05,
      "loss": 0.7293,
      "step": 1556
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.4121008993950965e-05,
      "loss": 0.7664,
      "step": 1557
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.4117461557299473e-05,
      "loss": 0.7342,
      "step": 1558
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.41139123826452e-05,
      "loss": 0.7552,
      "step": 1559
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.4110361471086233e-05,
      "loss": 0.7461,
      "step": 1560
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.4106808823721202e-05,
      "loss": 0.7697,
      "step": 1561
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.4103254441649267e-05,
      "loss": 0.7762,
      "step": 1562
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.4099698325970136e-05,
      "loss": 0.7094,
      "step": 1563
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.409614047778404e-05,
      "loss": 0.7581,
      "step": 1564
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.4092580898191752e-05,
      "loss": 0.713,
      "step": 1565
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.4089019588294584e-05,
      "loss": 0.7366,
      "step": 1566
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.4085456549194376e-05,
      "loss": 0.7433,
      "step": 1567
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.408189178199351e-05,
      "loss": 0.7268,
      "step": 1568
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.4078325287794901e-05,
      "loss": 0.7328,
      "step": 1569
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.4074757067701992e-05,
      "loss": 0.7352,
      "step": 1570
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.4071187122818769e-05,
      "loss": 0.7105,
      "step": 1571
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.4067615454249746e-05,
      "loss": 0.7298,
      "step": 1572
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.4064042063099971e-05,
      "loss": 0.7334,
      "step": 1573
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.4060466950475032e-05,
      "loss": 0.7587,
      "step": 1574
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.4056890117481036e-05,
      "loss": 0.7219,
      "step": 1575
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.4053311565224633e-05,
      "loss": 0.7206,
      "step": 1576
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.4049731294813002e-05,
      "loss": 0.7311,
      "step": 1577
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.404614930735385e-05,
      "loss": 0.7501,
      "step": 1578
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.4042565603955424e-05,
      "loss": 0.7565,
      "step": 1579
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.4038980185726493e-05,
      "loss": 0.7422,
      "step": 1580
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.403539305377636e-05,
      "loss": 0.7858,
      "step": 1581
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.4031804209214859e-05,
      "loss": 0.7276,
      "step": 1582
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.4028213653152352e-05,
      "loss": 0.7316,
      "step": 1583
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.4024621386699731e-05,
      "loss": 0.7336,
      "step": 1584
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.4021027410968423e-05,
      "loss": 0.7383,
      "step": 1585
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.4017431727070373e-05,
      "loss": 0.7165,
      "step": 1586
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.4013834336118062e-05,
      "loss": 0.7435,
      "step": 1587
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.4010235239224497e-05,
      "loss": 0.7129,
      "step": 1588
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.4006634437503216e-05,
      "loss": 0.729,
      "step": 1589
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.4003031932068272e-05,
      "loss": 0.7321,
      "step": 1590
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.3999427724034267e-05,
      "loss": 0.7175,
      "step": 1591
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.399582181451631e-05,
      "loss": 0.7326,
      "step": 1592
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.3992214204630046e-05,
      "loss": 0.7015,
      "step": 1593
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.3988604895491644e-05,
      "loss": 0.7202,
      "step": 1594
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.3984993888217797e-05,
      "loss": 0.6687,
      "step": 1595
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.3981381183925728e-05,
      "loss": 0.7215,
      "step": 1596
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.3977766783733178e-05,
      "loss": 0.7489,
      "step": 1597
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.3974150688758421e-05,
      "loss": 0.7564,
      "step": 1598
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.397053290012025e-05,
      "loss": 0.7469,
      "step": 1599
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.3966913418937978e-05,
      "loss": 0.7174,
      "step": 1600
    },
    {
      "epoch": 1.13,
      "eval_loss": 0.7343421578407288,
      "eval_runtime": 30.9217,
      "eval_samples_per_second": 264.895,
      "eval_steps_per_second": 16.558,
      "step": 1600
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.3963292246331454e-05,
      "loss": 0.7465,
      "step": 1601
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.3959669383421042e-05,
      "loss": 0.7425,
      "step": 1602
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.3956044831327629e-05,
      "loss": 0.7107,
      "step": 1603
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.395241859117262e-05,
      "loss": 0.73,
      "step": 1604
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.3948790664077958e-05,
      "loss": 0.7531,
      "step": 1605
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.3945161051166092e-05,
      "loss": 0.7186,
      "step": 1606
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.3941529753560002e-05,
      "loss": 0.7432,
      "step": 1607
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.393789677238318e-05,
      "loss": 0.7324,
      "step": 1608
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.393426210875965e-05,
      "loss": 0.7211,
      "step": 1609
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.3930625763813946e-05,
      "loss": 0.7472,
      "step": 1610
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.3926987738671133e-05,
      "loss": 0.7134,
      "step": 1611
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.3923348034456788e-05,
      "loss": 0.7322,
      "step": 1612
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.391970665229701e-05,
      "loss": 0.736,
      "step": 1613
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.3916063593318413e-05,
      "loss": 0.7293,
      "step": 1614
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.391241885864814e-05,
      "loss": 0.7268,
      "step": 1615
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.3908772449413842e-05,
      "loss": 0.6909,
      "step": 1616
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.3905124366743691e-05,
      "loss": 0.6961,
      "step": 1617
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.3901474611766383e-05,
      "loss": 0.7616,
      "step": 1618
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.3897823185611121e-05,
      "loss": 0.7554,
      "step": 1619
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.3894170089407636e-05,
      "loss": 0.7469,
      "step": 1620
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.3890515324286164e-05,
      "loss": 0.7244,
      "step": 1621
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.3886858891377466e-05,
      "loss": 0.6978,
      "step": 1622
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.3883200791812817e-05,
      "loss": 0.761,
      "step": 1623
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.3879541026724008e-05,
      "loss": 0.7273,
      "step": 1624
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.387587959724334e-05,
      "loss": 0.7093,
      "step": 1625
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.387221650450364e-05,
      "loss": 0.741,
      "step": 1626
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.3868551749638237e-05,
      "loss": 0.7649,
      "step": 1627
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.3864885333780984e-05,
      "loss": 0.7194,
      "step": 1628
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.3861217258066241e-05,
      "loss": 0.7509,
      "step": 1629
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.3857547523628884e-05,
      "loss": 0.7258,
      "step": 1630
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.3853876131604311e-05,
      "loss": 0.7122,
      "step": 1631
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.3850203083128413e-05,
      "loss": 0.7428,
      "step": 1632
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.384652837933761e-05,
      "loss": 0.7365,
      "step": 1633
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.3842852021368835e-05,
      "loss": 0.7638,
      "step": 1634
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.3839174010359521e-05,
      "loss": 0.7226,
      "step": 1635
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.383549434744762e-05,
      "loss": 0.7393,
      "step": 1636
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.3831813033771592e-05,
      "loss": 0.7227,
      "step": 1637
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.382813007047041e-05,
      "loss": 0.7366,
      "step": 1638
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.3824445458683558e-05,
      "loss": 0.7273,
      "step": 1639
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.382075919955103e-05,
      "loss": 0.7429,
      "step": 1640
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.3817071294213326e-05,
      "loss": 0.7361,
      "step": 1641
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.3813381743811454e-05,
      "loss": 0.7375,
      "step": 1642
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.3809690549486944e-05,
      "loss": 0.7204,
      "step": 1643
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.3805997712381819e-05,
      "loss": 0.755,
      "step": 1644
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.3802303233638619e-05,
      "loss": 0.771,
      "step": 1645
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.3798607114400386e-05,
      "loss": 0.7188,
      "step": 1646
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.3794909355810677e-05,
      "loss": 0.7052,
      "step": 1647
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.3791209959013553e-05,
      "loss": 0.7455,
      "step": 1648
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.3787508925153578e-05,
      "loss": 0.7504,
      "step": 1649
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.3783806255375826e-05,
      "loss": 0.7338,
      "step": 1650
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.378010195082588e-05,
      "loss": 0.7524,
      "step": 1651
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.377639601264982e-05,
      "loss": 0.7219,
      "step": 1652
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.377268844199424e-05,
      "loss": 0.735,
      "step": 1653
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.3768979240006237e-05,
      "loss": 0.7043,
      "step": 1654
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.3765268407833412e-05,
      "loss": 0.7155,
      "step": 1655
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.3761555946623868e-05,
      "loss": 0.7331,
      "step": 1656
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.3757841857526216e-05,
      "loss": 0.7434,
      "step": 1657
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.375412614168957e-05,
      "loss": 0.7171,
      "step": 1658
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.3750408800263542e-05,
      "loss": 0.7517,
      "step": 1659
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.3746689834398253e-05,
      "loss": 0.7089,
      "step": 1660
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.3742969245244326e-05,
      "loss": 0.7388,
      "step": 1661
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.3739247033952884e-05,
      "loss": 0.7236,
      "step": 1662
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.3735523201675553e-05,
      "loss": 0.7015,
      "step": 1663
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.373179774956446e-05,
      "loss": 0.7541,
      "step": 1664
    },
    {
      "epoch": 1.17,
      "eval_loss": 0.7337671518325806,
      "eval_runtime": 30.9403,
      "eval_samples_per_second": 264.736,
      "eval_steps_per_second": 16.548,
      "step": 1664
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.3728070678772234e-05,
      "loss": 0.7205,
      "step": 1665
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.3724341990452002e-05,
      "loss": 0.7355,
      "step": 1666
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.3720611685757397e-05,
      "loss": 0.7118,
      "step": 1667
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.3716879765842549e-05,
      "loss": 0.736,
      "step": 1668
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.3713146231862084e-05,
      "loss": 0.7038,
      "step": 1669
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.3709411084971133e-05,
      "loss": 0.7277,
      "step": 1670
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.3705674326325323e-05,
      "loss": 0.7312,
      "step": 1671
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.3701935957080783e-05,
      "loss": 0.732,
      "step": 1672
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.3698195978394137e-05,
      "loss": 0.7416,
      "step": 1673
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.3694454391422505e-05,
      "loss": 0.7162,
      "step": 1674
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.3690711197323512e-05,
      "loss": 0.7528,
      "step": 1675
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.368696639725527e-05,
      "loss": 0.7173,
      "step": 1676
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.3683219992376401e-05,
      "loss": 0.7423,
      "step": 1677
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.3679471983846011e-05,
      "loss": 0.7343,
      "step": 1678
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.367572237282371e-05,
      "loss": 0.7068,
      "step": 1679
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.3671971160469597e-05,
      "loss": 0.7326,
      "step": 1680
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.3668218347944274e-05,
      "loss": 0.7741,
      "step": 1681
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.3664463936408833e-05,
      "loss": 0.6948,
      "step": 1682
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.3660707927024867e-05,
      "loss": 0.7297,
      "step": 1683
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.3656950320954451e-05,
      "loss": 0.7198,
      "step": 1684
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.3653191119360167e-05,
      "loss": 0.7257,
      "step": 1685
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.3649430323405081e-05,
      "loss": 0.7147,
      "step": 1686
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.364566793425276e-05,
      "loss": 0.7208,
      "step": 1687
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.364190395306726e-05,
      "loss": 0.7601,
      "step": 1688
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.3638138381013133e-05,
      "loss": 0.7322,
      "step": 1689
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.3634371219255414e-05,
      "loss": 0.7234,
      "step": 1690
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.3630602468959639e-05,
      "loss": 0.7568,
      "step": 1691
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.3626832131291835e-05,
      "loss": 0.7209,
      "step": 1692
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.3623060207418517e-05,
      "loss": 0.7212,
      "step": 1693
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.3619286698506686e-05,
      "loss": 0.722,
      "step": 1694
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.3615511605723844e-05,
      "loss": 0.7577,
      "step": 1695
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.361173493023798e-05,
      "loss": 0.7623,
      "step": 1696
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3607956673217565e-05,
      "loss": 0.7378,
      "step": 1697
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3604176835831567e-05,
      "loss": 0.7491,
      "step": 1698
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3600395419249444e-05,
      "loss": 0.7067,
      "step": 1699
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3596612424641138e-05,
      "loss": 0.7507,
      "step": 1700
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3592827853177079e-05,
      "loss": 0.722,
      "step": 1701
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3589041706028184e-05,
      "loss": 0.71,
      "step": 1702
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3585253984365867e-05,
      "loss": 0.7129,
      "step": 1703
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3581464689362015e-05,
      "loss": 0.7159,
      "step": 1704
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3577673822189013e-05,
      "loss": 0.7211,
      "step": 1705
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.357388138401973e-05,
      "loss": 0.769,
      "step": 1706
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3570087376027519e-05,
      "loss": 0.7358,
      "step": 1707
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3566291799386215e-05,
      "loss": 0.7579,
      "step": 1708
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3562494655270146e-05,
      "loss": 0.7385,
      "step": 1709
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.355869594485412e-05,
      "loss": 0.7321,
      "step": 1710
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3554895669313433e-05,
      "loss": 0.712,
      "step": 1711
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.3551093829823862e-05,
      "loss": 0.7327,
      "step": 1712
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.354729042756167e-05,
      "loss": 0.7253,
      "step": 1713
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.35434854637036e-05,
      "loss": 0.7329,
      "step": 1714
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.3539678939426883e-05,
      "loss": 0.7533,
      "step": 1715
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.353587085590923e-05,
      "loss": 0.7018,
      "step": 1716
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.3532061214328838e-05,
      "loss": 0.7504,
      "step": 1717
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.3528250015864379e-05,
      "loss": 0.7093,
      "step": 1718
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.352443726169501e-05,
      "loss": 0.768,
      "step": 1719
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.3520622953000378e-05,
      "loss": 0.6946,
      "step": 1720
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.3516807090960593e-05,
      "loss": 0.7349,
      "step": 1721
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.351298967675626e-05,
      "loss": 0.7072,
      "step": 1722
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.3509170711568464e-05,
      "loss": 0.7496,
      "step": 1723
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.3505350196578759e-05,
      "loss": 0.7723,
      "step": 1724
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.3501528132969186e-05,
      "loss": 0.7034,
      "step": 1725
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.349770452192227e-05,
      "loss": 0.7355,
      "step": 1726
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.3493879364621005e-05,
      "loss": 0.7469,
      "step": 1727
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.3490052662248868e-05,
      "loss": 0.7217,
      "step": 1728
    },
    {
      "epoch": 1.22,
      "eval_loss": 0.7331604361534119,
      "eval_runtime": 30.9467,
      "eval_samples_per_second": 264.681,
      "eval_steps_per_second": 16.545,
      "step": 1728
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.3486224415989812e-05,
      "loss": 0.7544,
      "step": 1729
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.3482394627028274e-05,
      "loss": 0.7431,
      "step": 1730
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.3478563296549158e-05,
      "loss": 0.7326,
      "step": 1731
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.3474730425737853e-05,
      "loss": 0.7325,
      "step": 1732
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.3470896015780225e-05,
      "loss": 0.7661,
      "step": 1733
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.3467060067862608e-05,
      "loss": 0.7519,
      "step": 1734
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.346322258317182e-05,
      "loss": 0.6935,
      "step": 1735
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.3459383562895151e-05,
      "loss": 0.7513,
      "step": 1736
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.3455543008220366e-05,
      "loss": 0.7252,
      "step": 1737
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.3451700920335706e-05,
      "loss": 0.7465,
      "step": 1738
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.3447857300429887e-05,
      "loss": 0.719,
      "step": 1739
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.3444012149692096e-05,
      "loss": 0.7042,
      "step": 1740
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.3440165469311997e-05,
      "loss": 0.7515,
      "step": 1741
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.3436317260479721e-05,
      "loss": 0.7172,
      "step": 1742
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.3432467524385884e-05,
      "loss": 0.7449,
      "step": 1743
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.342861626222156e-05,
      "loss": 0.7315,
      "step": 1744
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.342476347517831e-05,
      "loss": 0.6828,
      "step": 1745
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.3420909164448152e-05,
      "loss": 0.7148,
      "step": 1746
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.3417053331223581e-05,
      "loss": 0.714,
      "step": 1747
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.3413195976697573e-05,
      "loss": 0.7472,
      "step": 1748
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.3409337102063559e-05,
      "loss": 0.7323,
      "step": 1749
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.3405476708515452e-05,
      "loss": 0.6992,
      "step": 1750
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.3401614797247626e-05,
      "loss": 0.7068,
      "step": 1751
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.3397751369454931e-05,
      "loss": 0.7338,
      "step": 1752
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.3393886426332686e-05,
      "loss": 0.7024,
      "step": 1753
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.3390019969076674e-05,
      "loss": 0.7261,
      "step": 1754
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.3386151998883153e-05,
      "loss": 0.7286,
      "step": 1755
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.3382282516948844e-05,
      "loss": 0.7692,
      "step": 1756
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.3378411524470935e-05,
      "loss": 0.7649,
      "step": 1757
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.3374539022647083e-05,
      "loss": 0.7238,
      "step": 1758
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.337066501267542e-05,
      "loss": 0.7427,
      "step": 1759
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.3366789495754527e-05,
      "loss": 0.7371,
      "step": 1760
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.3362912473083468e-05,
      "loss": 0.7669,
      "step": 1761
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.3359033945861766e-05,
      "loss": 0.7184,
      "step": 1762
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.335515391528941e-05,
      "loss": 0.759,
      "step": 1763
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.3351272382566851e-05,
      "loss": 0.7364,
      "step": 1764
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.334738934889501e-05,
      "loss": 0.711,
      "step": 1765
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.3343504815475267e-05,
      "loss": 0.7075,
      "step": 1766
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.3339618783509474e-05,
      "loss": 0.7392,
      "step": 1767
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.3335731254199939e-05,
      "loss": 0.7374,
      "step": 1768
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.3331842228749436e-05,
      "loss": 0.7861,
      "step": 1769
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.3327951708361204e-05,
      "loss": 0.7538,
      "step": 1770
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.332405969423894e-05,
      "loss": 0.7369,
      "step": 1771
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.3320166187586803e-05,
      "loss": 0.7269,
      "step": 1772
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.3316271189609425e-05,
      "loss": 0.7612,
      "step": 1773
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.3312374701511883e-05,
      "loss": 0.7538,
      "step": 1774
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.3308476724499725e-05,
      "loss": 0.752,
      "step": 1775
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.3304577259778957e-05,
      "loss": 0.7339,
      "step": 1776
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.3300676308556045e-05,
      "loss": 0.7258,
      "step": 1777
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.3296773872037921e-05,
      "loss": 0.7184,
      "step": 1778
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.3292869951431963e-05,
      "loss": 0.7349,
      "step": 1779
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.3288964547946024e-05,
      "loss": 0.7246,
      "step": 1780
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.32850576627884e-05,
      "loss": 0.7274,
      "step": 1781
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.3281149297167863e-05,
      "loss": 0.7319,
      "step": 1782
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.3277239452293627e-05,
      "loss": 0.7755,
      "step": 1783
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.3273328129375376e-05,
      "loss": 0.7495,
      "step": 1784
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.3269415329623237e-05,
      "loss": 0.7168,
      "step": 1785
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.3265501054247808e-05,
      "loss": 0.6987,
      "step": 1786
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.3261585304460137e-05,
      "loss": 0.7625,
      "step": 1787
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.325766808147173e-05,
      "loss": 0.748,
      "step": 1788
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.3253749386494551e-05,
      "loss": 0.7411,
      "step": 1789
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.324982922074101e-05,
      "loss": 0.7272,
      "step": 1790
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.3245907585423981e-05,
      "loss": 0.6961,
      "step": 1791
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.3241984481756791e-05,
      "loss": 0.7449,
      "step": 1792
    },
    {
      "epoch": 1.26,
      "eval_loss": 0.7318360805511475,
      "eval_runtime": 30.9258,
      "eval_samples_per_second": 264.859,
      "eval_steps_per_second": 16.556,
      "step": 1792
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.323805991095322e-05,
      "loss": 0.7266,
      "step": 1793
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.3234133874227504e-05,
      "loss": 0.7202,
      "step": 1794
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.323020637279433e-05,
      "loss": 0.7356,
      "step": 1795
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.3226277407868837e-05,
      "loss": 0.7341,
      "step": 1796
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.3222346980666618e-05,
      "loss": 0.7517,
      "step": 1797
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.3218415092403727e-05,
      "loss": 0.7268,
      "step": 1798
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.3214481744296651e-05,
      "loss": 0.7272,
      "step": 1799
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.321054693756235e-05,
      "loss": 0.7066,
      "step": 1800
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.3206610673418222e-05,
      "loss": 0.7057,
      "step": 1801
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.3202672953082118e-05,
      "loss": 0.7493,
      "step": 1802
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.319873377777234e-05,
      "loss": 0.7281,
      "step": 1803
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.3194793148707642e-05,
      "loss": 0.7125,
      "step": 1804
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.319085106710723e-05,
      "loss": 0.6966,
      "step": 1805
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.3186907534190752e-05,
      "loss": 0.7363,
      "step": 1806
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.318296255117831e-05,
      "loss": 0.7168,
      "step": 1807
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.3179016119290453e-05,
      "loss": 0.7266,
      "step": 1808
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.3175068239748182e-05,
      "loss": 0.7432,
      "step": 1809
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.3171118913772939e-05,
      "loss": 0.7501,
      "step": 1810
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.3167168142586622e-05,
      "loss": 0.7038,
      "step": 1811
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.3163215927411567e-05,
      "loss": 0.7157,
      "step": 1812
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.3159262269470565e-05,
      "loss": 0.7508,
      "step": 1813
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.3155307169986844e-05,
      "loss": 0.73,
      "step": 1814
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.315135063018409e-05,
      "loss": 0.7366,
      "step": 1815
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.314739265128643e-05,
      "loss": 0.7069,
      "step": 1816
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.314343323451843e-05,
      "loss": 0.7633,
      "step": 1817
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.3139472381105102e-05,
      "loss": 0.7257,
      "step": 1818
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.3135510092271912e-05,
      "loss": 0.7424,
      "step": 1819
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.3131546369244764e-05,
      "loss": 0.7396,
      "step": 1820
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.3127581213250003e-05,
      "loss": 0.7405,
      "step": 1821
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.3123614625514423e-05,
      "loss": 0.7252,
      "step": 1822
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.3119646607265255e-05,
      "loss": 0.7194,
      "step": 1823
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.3115677159730177e-05,
      "loss": 0.7029,
      "step": 1824
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.3111706284137309e-05,
      "loss": 0.7505,
      "step": 1825
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.310773398171521e-05,
      "loss": 0.7352,
      "step": 1826
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.3103760253692883e-05,
      "loss": 0.7297,
      "step": 1827
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.3099785101299774e-05,
      "loss": 0.736,
      "step": 1828
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.3095808525765762e-05,
      "loss": 0.6791,
      "step": 1829
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.3091830528321174e-05,
      "loss": 0.7379,
      "step": 1830
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.3087851110196772e-05,
      "loss": 0.7206,
      "step": 1831
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.3083870272623763e-05,
      "loss": 0.7353,
      "step": 1832
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.3079888016833791e-05,
      "loss": 0.7418,
      "step": 1833
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.3075904344058934e-05,
      "loss": 0.7203,
      "step": 1834
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.3071919255531712e-05,
      "loss": 0.708,
      "step": 1835
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.3067932752485083e-05,
      "loss": 0.7708,
      "step": 1836
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.3063944836152446e-05,
      "loss": 0.7344,
      "step": 1837
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.3059955507767633e-05,
      "loss": 0.7196,
      "step": 1838
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.3055964768564914e-05,
      "loss": 0.7212,
      "step": 1839
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.3051972619778993e-05,
      "loss": 0.7192,
      "step": 1840
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.3047979062645012e-05,
      "loss": 0.7183,
      "step": 1841
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.3043984098398556e-05,
      "loss": 0.7671,
      "step": 1842
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.3039987728275631e-05,
      "loss": 0.7341,
      "step": 1843
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.303598995351269e-05,
      "loss": 0.7072,
      "step": 1844
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.3031990775346615e-05,
      "loss": 0.6992,
      "step": 1845
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.3027990195014724e-05,
      "loss": 0.7155,
      "step": 1846
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.3023988213754766e-05,
      "loss": 0.7358,
      "step": 1847
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.3019984832804925e-05,
      "loss": 0.7502,
      "step": 1848
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.3015980053403824e-05,
      "loss": 0.7385,
      "step": 1849
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.3011973876790509e-05,
      "loss": 0.7205,
      "step": 1850
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.3007966304204461e-05,
      "loss": 0.7258,
      "step": 1851
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.30039573368856e-05,
      "loss": 0.7245,
      "step": 1852
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.299994697607427e-05,
      "loss": 0.7582,
      "step": 1853
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.2995935223011245e-05,
      "loss": 0.7053,
      "step": 1854
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.2991922078937738e-05,
      "loss": 0.7404,
      "step": 1855
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.2987907545095384e-05,
      "loss": 0.749,
      "step": 1856
    },
    {
      "epoch": 1.31,
      "eval_loss": 0.7317308783531189,
      "eval_runtime": 30.9063,
      "eval_samples_per_second": 265.027,
      "eval_steps_per_second": 16.566,
      "step": 1856
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.2983891622726251e-05,
      "loss": 0.7511,
      "step": 1857
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.297987431307284e-05,
      "loss": 0.7396,
      "step": 1858
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.2975855617378074e-05,
      "loss": 0.7529,
      "step": 1859
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.2971835536885309e-05,
      "loss": 0.7303,
      "step": 1860
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.2967814072838335e-05,
      "loss": 0.7327,
      "step": 1861
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.2963791226481363e-05,
      "loss": 0.7633,
      "step": 1862
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.2959766999059028e-05,
      "loss": 0.7276,
      "step": 1863
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.29557413918164e-05,
      "loss": 0.7053,
      "step": 1864
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.2951714405998975e-05,
      "loss": 0.7327,
      "step": 1865
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.2947686042852674e-05,
      "loss": 0.7169,
      "step": 1866
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.2943656303623841e-05,
      "loss": 0.7138,
      "step": 1867
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.293962518955925e-05,
      "loss": 0.7134,
      "step": 1868
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.29355927019061e-05,
      "loss": 0.7388,
      "step": 1869
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.2931558841912015e-05,
      "loss": 0.7066,
      "step": 1870
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.2927523610825037e-05,
      "loss": 0.7364,
      "step": 1871
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.2923487009893648e-05,
      "loss": 0.7293,
      "step": 1872
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.2919449040366737e-05,
      "loss": 0.7576,
      "step": 1873
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.2915409703493622e-05,
      "loss": 0.7172,
      "step": 1874
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.2911369000524047e-05,
      "loss": 0.7207,
      "step": 1875
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.2907326932708179e-05,
      "loss": 0.7507,
      "step": 1876
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.2903283501296605e-05,
      "loss": 0.7554,
      "step": 1877
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.2899238707540327e-05,
      "loss": 0.7427,
      "step": 1878
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.2895192552690785e-05,
      "loss": 0.689,
      "step": 1879
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.2891145037999825e-05,
      "loss": 0.7149,
      "step": 1880
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.288709616471972e-05,
      "loss": 0.7143,
      "step": 1881
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.2883045934103165e-05,
      "loss": 0.7318,
      "step": 1882
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.2878994347403272e-05,
      "loss": 0.7454,
      "step": 1883
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.2874941405873572e-05,
      "loss": 0.6987,
      "step": 1884
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.2870887110768015e-05,
      "loss": 0.7364,
      "step": 1885
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.2866831463340973e-05,
      "loss": 0.7277,
      "step": 1886
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.2862774464847235e-05,
      "loss": 0.7229,
      "step": 1887
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.2858716116542006e-05,
      "loss": 0.7411,
      "step": 1888
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.2854656419680912e-05,
      "loss": 0.7327,
      "step": 1889
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.2850595375519994e-05,
      "loss": 0.7695,
      "step": 1890
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.284653298531571e-05,
      "loss": 0.7493,
      "step": 1891
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.2842469250324934e-05,
      "loss": 0.6988,
      "step": 1892
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.2838404171804958e-05,
      "loss": 0.7226,
      "step": 1893
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.2834337751013488e-05,
      "loss": 0.7353,
      "step": 1894
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.2830269989208645e-05,
      "loss": 0.721,
      "step": 1895
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.2826200887648967e-05,
      "loss": 0.7149,
      "step": 1896
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.2822130447593406e-05,
      "loss": 0.7408,
      "step": 1897
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.281805867030132e-05,
      "loss": 0.7814,
      "step": 1898
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.2813985557032499e-05,
      "loss": 0.7427,
      "step": 1899
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.2809911109047131e-05,
      "loss": 0.7553,
      "step": 1900
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.2805835327605818e-05,
      "loss": 0.72,
      "step": 1901
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.2801758213969579e-05,
      "loss": 0.7548,
      "step": 1902
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.2797679769399849e-05,
      "loss": 0.719,
      "step": 1903
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.2793599995158462e-05,
      "loss": 0.7454,
      "step": 1904
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.2789518892507682e-05,
      "loss": 0.7251,
      "step": 1905
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.2785436462710162e-05,
      "loss": 0.7597,
      "step": 1906
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.278135270702898e-05,
      "loss": 0.7063,
      "step": 1907
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.2777267626727625e-05,
      "loss": 0.748,
      "step": 1908
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.277318122306999e-05,
      "loss": 0.7187,
      "step": 1909
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.2769093497320379e-05,
      "loss": 0.6931,
      "step": 1910
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.2765004450743504e-05,
      "loss": 0.6961,
      "step": 1911
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.276091408460449e-05,
      "loss": 0.7443,
      "step": 1912
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.2756822400168864e-05,
      "loss": 0.7112,
      "step": 1913
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.2752729398702567e-05,
      "loss": 0.7508,
      "step": 1914
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.2748635081471941e-05,
      "loss": 0.7371,
      "step": 1915
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.2744539449743744e-05,
      "loss": 0.7188,
      "step": 1916
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.274044250478513e-05,
      "loss": 0.7439,
      "step": 1917
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.2736344247863667e-05,
      "loss": 0.7209,
      "step": 1918
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.2732244680247329e-05,
      "loss": 0.7466,
      "step": 1919
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.272814380320449e-05,
      "loss": 0.7301,
      "step": 1920
    },
    {
      "epoch": 1.35,
      "eval_loss": 0.7303444743156433,
      "eval_runtime": 30.9297,
      "eval_samples_per_second": 264.826,
      "eval_steps_per_second": 16.554,
      "step": 1920
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.2724041618003933e-05,
      "loss": 0.7007,
      "step": 1921
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.2719938125914845e-05,
      "loss": 0.7024,
      "step": 1922
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.271583332820682e-05,
      "loss": 0.742,
      "step": 1923
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.2711727226149848e-05,
      "loss": 0.7221,
      "step": 1924
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2707619821014332e-05,
      "loss": 0.7598,
      "step": 1925
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2703511114071072e-05,
      "loss": 0.6972,
      "step": 1926
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2699401106591275e-05,
      "loss": 0.7473,
      "step": 1927
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2695289799846546e-05,
      "loss": 0.7346,
      "step": 1928
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2691177195108896e-05,
      "loss": 0.744,
      "step": 1929
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2687063293650734e-05,
      "loss": 0.7458,
      "step": 1930
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2682948096744872e-05,
      "loss": 0.687,
      "step": 1931
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2678831605664523e-05,
      "loss": 0.7207,
      "step": 1932
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2674713821683304e-05,
      "loss": 0.7195,
      "step": 1933
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2670594746075221e-05,
      "loss": 0.7217,
      "step": 1934
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2666474380114694e-05,
      "loss": 0.7245,
      "step": 1935
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2662352725076528e-05,
      "loss": 0.7407,
      "step": 1936
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2658229782235936e-05,
      "loss": 0.7243,
      "step": 1937
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2654105552868535e-05,
      "loss": 0.689,
      "step": 1938
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.2649980038250322e-05,
      "loss": 0.7134,
      "step": 1939
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.2645853239657706e-05,
      "loss": 0.7322,
      "step": 1940
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.264172515836749e-05,
      "loss": 0.7009,
      "step": 1941
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.2637595795656874e-05,
      "loss": 0.7529,
      "step": 1942
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.2633465152803449e-05,
      "loss": 0.7403,
      "step": 1943
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.2629333231085211e-05,
      "loss": 0.7564,
      "step": 1944
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.2625200031780543e-05,
      "loss": 0.7281,
      "step": 1945
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.2621065556168234e-05,
      "loss": 0.6892,
      "step": 1946
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.2616929805527456e-05,
      "loss": 0.733,
      "step": 1947
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.2612792781137782e-05,
      "loss": 0.7395,
      "step": 1948
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.2608654484279183e-05,
      "loss": 0.7461,
      "step": 1949
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.2604514916232011e-05,
      "loss": 0.7415,
      "step": 1950
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.2600374078277023e-05,
      "loss": 0.7603,
      "step": 1951
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.2596231971695365e-05,
      "loss": 0.723,
      "step": 1952
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.2592088597768572e-05,
      "loss": 0.7416,
      "step": 1953
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.258794395777858e-05,
      "loss": 0.7163,
      "step": 1954
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.2583798053007707e-05,
      "loss": 0.7638,
      "step": 1955
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.2579650884738667e-05,
      "loss": 0.7287,
      "step": 1956
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.257550245425457e-05,
      "loss": 0.7338,
      "step": 1957
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.2571352762838902e-05,
      "loss": 0.7428,
      "step": 1958
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.2567201811775555e-05,
      "loss": 0.7276,
      "step": 1959
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.2563049602348802e-05,
      "loss": 0.7033,
      "step": 1960
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.2558896135843305e-05,
      "loss": 0.795,
      "step": 1961
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.255474141354412e-05,
      "loss": 0.7469,
      "step": 1962
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.2550585436736686e-05,
      "loss": 0.6942,
      "step": 1963
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.2546428206706833e-05,
      "loss": 0.7126,
      "step": 1964
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.254226972474078e-05,
      "loss": 0.7389,
      "step": 1965
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.2538109992125133e-05,
      "loss": 0.7111,
      "step": 1966
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.2533949010146882e-05,
      "loss": 0.741,
      "step": 1967
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.2529786780093405e-05,
      "loss": 0.7172,
      "step": 1968
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.2525623303252466e-05,
      "loss": 0.6985,
      "step": 1969
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.2521458580912218e-05,
      "loss": 0.7213,
      "step": 1970
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.2517292614361195e-05,
      "loss": 0.7096,
      "step": 1971
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.2513125404888317e-05,
      "loss": 0.7157,
      "step": 1972
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.250895695378289e-05,
      "loss": 0.7466,
      "step": 1973
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.2504787262334598e-05,
      "loss": 0.7209,
      "step": 1974
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.2500616331833521e-05,
      "loss": 0.7673,
      "step": 1975
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.2496444163570114e-05,
      "loss": 0.6716,
      "step": 1976
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.2492270758835217e-05,
      "loss": 0.7292,
      "step": 1977
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.2488096118920048e-05,
      "loss": 0.7391,
      "step": 1978
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.2483920245116213e-05,
      "loss": 0.7531,
      "step": 1979
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.2479743138715697e-05,
      "loss": 0.718,
      "step": 1980
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.2475564801010867e-05,
      "loss": 0.713,
      "step": 1981
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.2471385233294476e-05,
      "loss": 0.7526,
      "step": 1982
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.2467204436859644e-05,
      "loss": 0.7,
      "step": 1983
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.2463022412999886e-05,
      "loss": 0.7294,
      "step": 1984
    },
    {
      "epoch": 1.4,
      "eval_loss": 0.7304807901382446,
      "eval_runtime": 30.9366,
      "eval_samples_per_second": 264.767,
      "eval_steps_per_second": 16.55,
      "step": 1984
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.2458839163009089e-05,
      "loss": 0.7386,
      "step": 1985
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.2454654688181519e-05,
      "loss": 0.7349,
      "step": 1986
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.2450468989811827e-05,
      "loss": 0.7176,
      "step": 1987
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.2446282069195037e-05,
      "loss": 0.7491,
      "step": 1988
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.2442093927626551e-05,
      "loss": 0.7485,
      "step": 1989
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.2437904566402147e-05,
      "loss": 0.7282,
      "step": 1990
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.2433713986817991e-05,
      "loss": 0.725,
      "step": 1991
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.2429522190170615e-05,
      "loss": 0.7297,
      "step": 1992
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.242532917775693e-05,
      "loss": 0.7276,
      "step": 1993
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.2421134950874226e-05,
      "loss": 0.6914,
      "step": 1994
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.2416939510820166e-05,
      "loss": 0.748,
      "step": 1995
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.2412742858892791e-05,
      "loss": 0.7196,
      "step": 1996
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.2408544996390514e-05,
      "loss": 0.734,
      "step": 1997
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.2404345924612125e-05,
      "loss": 0.7567,
      "step": 1998
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.2400145644856789e-05,
      "loss": 0.7249,
      "step": 1999
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.2395944158424036e-05,
      "loss": 0.752,
      "step": 2000
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.2391741466613782e-05,
      "loss": 0.7395,
      "step": 2001
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.2387537570726308e-05,
      "loss": 0.7201,
      "step": 2002
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.2383332472062266e-05,
      "loss": 0.764,
      "step": 2003
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.2379126171922695e-05,
      "loss": 0.7345,
      "step": 2004
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.2374918671608983e-05,
      "loss": 0.719,
      "step": 2005
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.2370709972422905e-05,
      "loss": 0.7447,
      "step": 2006
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.2366500075666603e-05,
      "loss": 0.7081,
      "step": 2007
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.2362288982642589e-05,
      "loss": 0.7566,
      "step": 2008
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.2358076694653742e-05,
      "loss": 0.7088,
      "step": 2009
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.2353863213003322e-05,
      "loss": 0.7232,
      "step": 2010
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.2349648538994944e-05,
      "loss": 0.7332,
      "step": 2011
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.2345432673932596e-05,
      "loss": 0.7238,
      "step": 2012
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.2341215619120643e-05,
      "loss": 0.7611,
      "step": 2013
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.233699737586381e-05,
      "loss": 0.736,
      "step": 2014
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.2332777945467192e-05,
      "loss": 0.7583,
      "step": 2015
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.2328557329236248e-05,
      "loss": 0.7093,
      "step": 2016
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.232433552847681e-05,
      "loss": 0.7254,
      "step": 2017
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.232011254449507e-05,
      "loss": 0.7401,
      "step": 2018
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.231588837859759e-05,
      "loss": 0.7294,
      "step": 2019
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.23116630320913e-05,
      "loss": 0.7291,
      "step": 2020
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.2307436506283489e-05,
      "loss": 0.7565,
      "step": 2021
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.2303208802481816e-05,
      "loss": 0.7425,
      "step": 2022
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.2298979921994302e-05,
      "loss": 0.7561,
      "step": 2023
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.2294749866129332e-05,
      "loss": 0.7469,
      "step": 2024
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.2290518636195657e-05,
      "loss": 0.7007,
      "step": 2025
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.228628623350239e-05,
      "loss": 0.7268,
      "step": 2026
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.2282052659359002e-05,
      "loss": 0.7228,
      "step": 2027
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.2277817915075335e-05,
      "loss": 0.733,
      "step": 2028
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.2273582001961586e-05,
      "loss": 0.7429,
      "step": 2029
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.2269344921328318e-05,
      "loss": 0.7241,
      "step": 2030
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.2265106674486456e-05,
      "loss": 0.7375,
      "step": 2031
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.2260867262747279e-05,
      "loss": 0.7687,
      "step": 2032
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.2256626687422433e-05,
      "loss": 0.7296,
      "step": 2033
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.2252384949823923e-05,
      "loss": 0.7295,
      "step": 2034
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.2248142051264113e-05,
      "loss": 0.7501,
      "step": 2035
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.2243897993055723e-05,
      "loss": 0.7129,
      "step": 2036
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.2239652776511838e-05,
      "loss": 0.7241,
      "step": 2037
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.2235406402945895e-05,
      "loss": 0.7243,
      "step": 2038
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.223115887367169e-05,
      "loss": 0.7474,
      "step": 2039
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.2226910190003384e-05,
      "loss": 0.7455,
      "step": 2040
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.2222660353255483e-05,
      "loss": 0.6979,
      "step": 2041
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.2218409364742861e-05,
      "loss": 0.7213,
      "step": 2042
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.2214157225780747e-05,
      "loss": 0.7224,
      "step": 2043
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.2209903937684717e-05,
      "loss": 0.7107,
      "step": 2044
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.2205649501770708e-05,
      "loss": 0.7673,
      "step": 2045
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.2201393919355013e-05,
      "loss": 0.7469,
      "step": 2046
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.2197137191754283e-05,
      "loss": 0.7476,
      "step": 2047
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.2192879320285516e-05,
      "loss": 0.725,
      "step": 2048
    },
    {
      "epoch": 1.44,
      "eval_loss": 0.7302752137184143,
      "eval_runtime": 30.9138,
      "eval_samples_per_second": 264.962,
      "eval_steps_per_second": 16.562,
      "step": 2048
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.2188620306266065e-05,
      "loss": 0.7246,
      "step": 2049
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.2184360151013642e-05,
      "loss": 0.7223,
      "step": 2050
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.2180098855846308e-05,
      "loss": 0.7204,
      "step": 2051
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.2175836422082479e-05,
      "loss": 0.7281,
      "step": 2052
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.217157285104092e-05,
      "loss": 0.7195,
      "step": 2053
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.2167308144040753e-05,
      "loss": 0.7021,
      "step": 2054
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.2163042302401442e-05,
      "loss": 0.788,
      "step": 2055
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.2158775327442811e-05,
      "loss": 0.7252,
      "step": 2056
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.2154507220485032e-05,
      "loss": 0.7461,
      "step": 2057
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.2150237982848628e-05,
      "loss": 0.7311,
      "step": 2058
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.2145967615854471e-05,
      "loss": 0.7345,
      "step": 2059
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.214169612082378e-05,
      "loss": 0.73,
      "step": 2060
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.2137423499078128e-05,
      "loss": 0.7541,
      "step": 2061
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.2133149751939431e-05,
      "loss": 0.6983,
      "step": 2062
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.2128874880729958e-05,
      "loss": 0.7029,
      "step": 2063
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.2124598886772323e-05,
      "loss": 0.7393,
      "step": 2064
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.212032177138949e-05,
      "loss": 0.7296,
      "step": 2065
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.2116043535904769e-05,
      "loss": 0.7113,
      "step": 2066
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.211176418164181e-05,
      "loss": 0.7394,
      "step": 2067
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.210748370992462e-05,
      "loss": 0.7508,
      "step": 2068
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.2103202122077544e-05,
      "loss": 0.7161,
      "step": 2069
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.2098919419425273e-05,
      "loss": 0.7266,
      "step": 2070
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.2094635603292852e-05,
      "loss": 0.7271,
      "step": 2071
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.2090350675005658e-05,
      "loss": 0.7193,
      "step": 2072
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.208606463588942e-05,
      "loss": 0.6977,
      "step": 2073
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.2081777487270205e-05,
      "loss": 0.7366,
      "step": 2074
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.2077489230474428e-05,
      "loss": 0.7261,
      "step": 2075
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.2073199866828848e-05,
      "loss": 0.7149,
      "step": 2076
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.2068909397660558e-05,
      "loss": 0.7172,
      "step": 2077
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.2064617824297004e-05,
      "loss": 0.6975,
      "step": 2078
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.2060325148065962e-05,
      "loss": 0.7198,
      "step": 2079
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.2056031370295561e-05,
      "loss": 0.7279,
      "step": 2080
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.2051736492314265e-05,
      "loss": 0.7064,
      "step": 2081
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.204744051545088e-05,
      "loss": 0.7061,
      "step": 2082
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.2043143441034547e-05,
      "loss": 0.7151,
      "step": 2083
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.2038845270394752e-05,
      "loss": 0.7395,
      "step": 2084
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.2034546004861321e-05,
      "loss": 0.7431,
      "step": 2085
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.2030245645764409e-05,
      "loss": 0.7472,
      "step": 2086
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.2025944194434527e-05,
      "loss": 0.7136,
      "step": 2087
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.2021641652202502e-05,
      "loss": 0.7248,
      "step": 2088
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.2017338020399521e-05,
      "loss": 0.6927,
      "step": 2089
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.201303330035709e-05,
      "loss": 0.7328,
      "step": 2090
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.2008727493407062e-05,
      "loss": 0.7172,
      "step": 2091
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.2004420600881622e-05,
      "loss": 0.7319,
      "step": 2092
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.2000112624113293e-05,
      "loss": 0.7331,
      "step": 2093
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.1995803564434932e-05,
      "loss": 0.7129,
      "step": 2094
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.1991493423179732e-05,
      "loss": 0.7222,
      "step": 2095
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.1987182201681216e-05,
      "loss": 0.7157,
      "step": 2096
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.1982869901273251e-05,
      "loss": 0.702,
      "step": 2097
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.197855652329003e-05,
      "loss": 0.7116,
      "step": 2098
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.1974242069066082e-05,
      "loss": 0.6918,
      "step": 2099
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.1969926539936266e-05,
      "loss": 0.7522,
      "step": 2100
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.196560993723578e-05,
      "loss": 0.7086,
      "step": 2101
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.1961292262300147e-05,
      "loss": 0.7306,
      "step": 2102
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.1956973516465226e-05,
      "loss": 0.7272,
      "step": 2103
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.1952653701067212e-05,
      "loss": 0.76,
      "step": 2104
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.1948332817442617e-05,
      "loss": 0.714,
      "step": 2105
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.1944010866928294e-05,
      "loss": 0.7133,
      "step": 2106
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.1939687850861424e-05,
      "loss": 0.7241,
      "step": 2107
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.1935363770579518e-05,
      "loss": 0.7145,
      "step": 2108
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.1931038627420418e-05,
      "loss": 0.7141,
      "step": 2109
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.1926712422722292e-05,
      "loss": 0.7285,
      "step": 2110
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.1922385157823633e-05,
      "loss": 0.7144,
      "step": 2111
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.191805683406327e-05,
      "loss": 0.7586,
      "step": 2112
    },
    {
      "epoch": 1.49,
      "eval_loss": 0.7292432188987732,
      "eval_runtime": 30.9333,
      "eval_samples_per_second": 264.796,
      "eval_steps_per_second": 16.552,
      "step": 2112
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.1913727452780352e-05,
      "loss": 0.7219,
      "step": 2113
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.1909397015314366e-05,
      "loss": 0.727,
      "step": 2114
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.1905065523005111e-05,
      "loss": 0.7327,
      "step": 2115
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.1900732977192724e-05,
      "loss": 0.7099,
      "step": 2116
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.189639937921766e-05,
      "loss": 0.742,
      "step": 2117
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.1892064730420711e-05,
      "loss": 0.7628,
      "step": 2118
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.188772903214298e-05,
      "loss": 0.7174,
      "step": 2119
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.1883392285725902e-05,
      "loss": 0.7383,
      "step": 2120
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.187905449251124e-05,
      "loss": 0.7637,
      "step": 2121
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.1874715653841072e-05,
      "loss": 0.7253,
      "step": 2122
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.1870375771057806e-05,
      "loss": 0.7457,
      "step": 2123
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.1866034845504169e-05,
      "loss": 0.6733,
      "step": 2124
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.1861692878523213e-05,
      "loss": 0.7569,
      "step": 2125
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.1857349871458316e-05,
      "loss": 0.6971,
      "step": 2126
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.185300582565317e-05,
      "loss": 0.7329,
      "step": 2127
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.184866074245179e-05,
      "loss": 0.7424,
      "step": 2128
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.1844314623198515e-05,
      "loss": 0.7316,
      "step": 2129
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.183996746923801e-05,
      "loss": 0.6963,
      "step": 2130
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.1835619281915244e-05,
      "loss": 0.7294,
      "step": 2131
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.1831270062575522e-05,
      "loss": 0.6874,
      "step": 2132
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.1826919812564458e-05,
      "loss": 0.7447,
      "step": 2133
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.1822568533227991e-05,
      "loss": 0.7054,
      "step": 2134
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.1818216225912374e-05,
      "loss": 0.7388,
      "step": 2135
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.181386289196418e-05,
      "loss": 0.7181,
      "step": 2136
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.1809508532730303e-05,
      "loss": 0.7542,
      "step": 2137
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.1805153149557946e-05,
      "loss": 0.7306,
      "step": 2138
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.1800796743794637e-05,
      "loss": 0.7567,
      "step": 2139
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.1796439316788214e-05,
      "loss": 0.727,
      "step": 2140
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.179208086988684e-05,
      "loss": 0.7312,
      "step": 2141
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.1787721404438984e-05,
      "loss": 0.7297,
      "step": 2142
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.1783360921793429e-05,
      "loss": 0.7528,
      "step": 2143
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.1778999423299285e-05,
      "loss": 0.7281,
      "step": 2144
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.1774636910305964e-05,
      "loss": 0.6979,
      "step": 2145
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.1770273384163198e-05,
      "loss": 0.7244,
      "step": 2146
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.1765908846221031e-05,
      "loss": 0.7485,
      "step": 2147
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.1761543297829824e-05,
      "loss": 0.7016,
      "step": 2148
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.175717674034024e-05,
      "loss": 0.7275,
      "step": 2149
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.1752809175103264e-05,
      "loss": 0.7331,
      "step": 2150
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.1748440603470189e-05,
      "loss": 0.7232,
      "step": 2151
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.1744071026792622e-05,
      "loss": 0.755,
      "step": 2152
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.173970044642248e-05,
      "loss": 0.7219,
      "step": 2153
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.1735328863711982e-05,
      "loss": 0.6958,
      "step": 2154
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.1730956280013674e-05,
      "loss": 0.7381,
      "step": 2155
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.1726582696680396e-05,
      "loss": 0.7544,
      "step": 2156
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.1722208115065309e-05,
      "loss": 0.7239,
      "step": 2157
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.171783253652187e-05,
      "loss": 0.7093,
      "step": 2158
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.1713455962403863e-05,
      "loss": 0.6849,
      "step": 2159
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.170907839406536e-05,
      "loss": 0.7377,
      "step": 2160
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.1704699832860752e-05,
      "loss": 0.6892,
      "step": 2161
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.1700320280144735e-05,
      "loss": 0.7669,
      "step": 2162
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.1695939737272312e-05,
      "loss": 0.7178,
      "step": 2163
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.169155820559879e-05,
      "loss": 0.7364,
      "step": 2164
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.1687175686479787e-05,
      "loss": 0.7377,
      "step": 2165
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.1682792181271221e-05,
      "loss": 0.7177,
      "step": 2166
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.1678407691329319e-05,
      "loss": 0.7373,
      "step": 2167
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.167402221801061e-05,
      "loss": 0.6922,
      "step": 2168
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.1669635762671928e-05,
      "loss": 0.7509,
      "step": 2169
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.1665248326670414e-05,
      "loss": 0.7042,
      "step": 2170
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.1660859911363505e-05,
      "loss": 0.7315,
      "step": 2171
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.165647051810895e-05,
      "loss": 0.7124,
      "step": 2172
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.1652080148264791e-05,
      "loss": 0.7302,
      "step": 2173
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.1647688803189383e-05,
      "loss": 0.7421,
      "step": 2174
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.1643296484241373e-05,
      "loss": 0.7473,
      "step": 2175
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.1638903192779716e-05,
      "loss": 0.7444,
      "step": 2176
    },
    {
      "epoch": 1.53,
      "eval_loss": 0.7289339303970337,
      "eval_runtime": 30.9425,
      "eval_samples_per_second": 264.717,
      "eval_steps_per_second": 16.547,
      "step": 2176
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.1634508930163663e-05,
      "loss": 0.7388,
      "step": 2177
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.1630113697752772e-05,
      "loss": 0.7258,
      "step": 2178
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.1625717496906889e-05,
      "loss": 0.6965,
      "step": 2179
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1621320328986175e-05,
      "loss": 0.7276,
      "step": 2180
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1616922195351079e-05,
      "loss": 0.7317,
      "step": 2181
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1612523097362354e-05,
      "loss": 0.7601,
      "step": 2182
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1608123036381047e-05,
      "loss": 0.7194,
      "step": 2183
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1603722013768502e-05,
      "loss": 0.7568,
      "step": 2184
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1599320030886376e-05,
      "loss": 0.7564,
      "step": 2185
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1594917089096605e-05,
      "loss": 0.7205,
      "step": 2186
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1590513189761427e-05,
      "loss": 0.7104,
      "step": 2187
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1586108334243374e-05,
      "loss": 0.7361,
      "step": 2188
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1581702523905285e-05,
      "loss": 0.6955,
      "step": 2189
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1577295760110284e-05,
      "loss": 0.7389,
      "step": 2190
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1572888044221785e-05,
      "loss": 0.7185,
      "step": 2191
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1568479377603517e-05,
      "loss": 0.7241,
      "step": 2192
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.1564069761619478e-05,
      "loss": 0.6989,
      "step": 2193
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.155965919763398e-05,
      "loss": 0.7211,
      "step": 2194
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.1555247687011614e-05,
      "loss": 0.7116,
      "step": 2195
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.1550835231117277e-05,
      "loss": 0.7556,
      "step": 2196
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.1546421831316148e-05,
      "loss": 0.7301,
      "step": 2197
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.15420074889737e-05,
      "loss": 0.7247,
      "step": 2198
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.1537592205455703e-05,
      "loss": 0.6972,
      "step": 2199
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.153317598212821e-05,
      "loss": 0.7326,
      "step": 2200
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.1528758820357571e-05,
      "loss": 0.7224,
      "step": 2201
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.1524340721510429e-05,
      "loss": 0.7382,
      "step": 2202
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.1519921686953704e-05,
      "loss": 0.7351,
      "step": 2203
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.151550171805462e-05,
      "loss": 0.7126,
      "step": 2204
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.1511080816180683e-05,
      "loss": 0.6793,
      "step": 2205
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.1506658982699688e-05,
      "loss": 0.7255,
      "step": 2206
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.1502236218979718e-05,
      "loss": 0.7204,
      "step": 2207
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.1497812526389148e-05,
      "loss": 0.7499,
      "step": 2208
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.1493387906296636e-05,
      "loss": 0.7437,
      "step": 2209
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.1488962360071125e-05,
      "loss": 0.7198,
      "step": 2210
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.1484535889081849e-05,
      "loss": 0.7301,
      "step": 2211
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.1480108494698328e-05,
      "loss": 0.7017,
      "step": 2212
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.1475680178290364e-05,
      "loss": 0.7367,
      "step": 2213
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.1471250941228054e-05,
      "loss": 0.6997,
      "step": 2214
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.1466820784881764e-05,
      "loss": 0.7534,
      "step": 2215
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.1462389710622156e-05,
      "loss": 0.6981,
      "step": 2216
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.1457957719820173e-05,
      "loss": 0.7052,
      "step": 2217
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.145352481384704e-05,
      "loss": 0.7139,
      "step": 2218
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.1449090994074273e-05,
      "loss": 0.7412,
      "step": 2219
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.1444656261873658e-05,
      "loss": 0.776,
      "step": 2220
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.1440220618617268e-05,
      "loss": 0.7269,
      "step": 2221
    },
    {
      "epoch": 1.56,
      "learning_rate": 1.1435784065677464e-05,
      "loss": 0.7234,
      "step": 2222
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.1431346604426883e-05,
      "loss": 0.7154,
      "step": 2223
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.1426908236238447e-05,
      "loss": 0.748,
      "step": 2224
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.1422468962485351e-05,
      "loss": 0.7319,
      "step": 2225
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.1418028784541077e-05,
      "loss": 0.7339,
      "step": 2226
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.1413587703779382e-05,
      "loss": 0.7122,
      "step": 2227
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.140914572157431e-05,
      "loss": 0.7401,
      "step": 2228
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.1404702839300175e-05,
      "loss": 0.7473,
      "step": 2229
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.1400259058331574e-05,
      "loss": 0.76,
      "step": 2230
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.1395814380043379e-05,
      "loss": 0.7493,
      "step": 2231
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.1391368805810747e-05,
      "loss": 0.7593,
      "step": 2232
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.1386922337009103e-05,
      "loss": 0.7425,
      "step": 2233
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.1382474975014156e-05,
      "loss": 0.7577,
      "step": 2234
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.1378026721201886e-05,
      "loss": 0.7317,
      "step": 2235
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.1373577576948553e-05,
      "loss": 0.7461,
      "step": 2236
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.1369127543630689e-05,
      "loss": 0.7379,
      "step": 2237
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.1364676622625104e-05,
      "loss": 0.7227,
      "step": 2238
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.1360224815308879e-05,
      "loss": 0.7535,
      "step": 2239
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.1355772123059373e-05,
      "loss": 0.7237,
      "step": 2240
    },
    {
      "epoch": 1.58,
      "eval_loss": 0.7281379699707031,
      "eval_runtime": 30.9138,
      "eval_samples_per_second": 264.963,
      "eval_steps_per_second": 16.562,
      "step": 2240
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.1351318547254221e-05,
      "loss": 0.7468,
      "step": 2241
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.1346864089271323e-05,
      "loss": 0.7282,
      "step": 2242
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.1342408750488856e-05,
      "loss": 0.7029,
      "step": 2243
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.1337952532285275e-05,
      "loss": 0.749,
      "step": 2244
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.13334954360393e-05,
      "loss": 0.7066,
      "step": 2245
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.1329037463129923e-05,
      "loss": 0.7371,
      "step": 2246
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.1324578614936412e-05,
      "loss": 0.7074,
      "step": 2247
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.1320118892838303e-05,
      "loss": 0.7239,
      "step": 2248
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.1315658298215402e-05,
      "loss": 0.7409,
      "step": 2249
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.1311196832447785e-05,
      "loss": 0.7451,
      "step": 2250
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.1306734496915793e-05,
      "loss": 0.7419,
      "step": 2251
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.130227129300005e-05,
      "loss": 0.7516,
      "step": 2252
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.1297807222081434e-05,
      "loss": 0.7431,
      "step": 2253
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.12933422855411e-05,
      "loss": 0.7331,
      "step": 2254
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.1288876484760465e-05,
      "loss": 0.7483,
      "step": 2255
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.1284409821121216e-05,
      "loss": 0.7403,
      "step": 2256
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.1279942296005308e-05,
      "loss": 0.7357,
      "step": 2257
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.1275473910794965e-05,
      "loss": 0.7605,
      "step": 2258
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.1271004666872668e-05,
      "loss": 0.7347,
      "step": 2259
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.126653456562117e-05,
      "loss": 0.7348,
      "step": 2260
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.1262063608423492e-05,
      "loss": 0.7959,
      "step": 2261
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.1257591796662914e-05,
      "loss": 0.7352,
      "step": 2262
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.1253119131722986e-05,
      "loss": 0.7079,
      "step": 2263
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.1248645614987517e-05,
      "loss": 0.7464,
      "step": 2264
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.1244171247840579e-05,
      "loss": 0.7377,
      "step": 2265
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.1239696031666512e-05,
      "loss": 0.708,
      "step": 2266
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.1235219967849918e-05,
      "loss": 0.7207,
      "step": 2267
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.1230743057775651e-05,
      "loss": 0.699,
      "step": 2268
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.1226265302828845e-05,
      "loss": 0.7627,
      "step": 2269
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.1221786704394882e-05,
      "loss": 0.7257,
      "step": 2270
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.1217307263859409e-05,
      "loss": 0.7552,
      "step": 2271
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.1212826982608331e-05,
      "loss": 0.7338,
      "step": 2272
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.1208345862027818e-05,
      "loss": 0.7153,
      "step": 2273
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.1203863903504295e-05,
      "loss": 0.7387,
      "step": 2274
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.119938110842445e-05,
      "loss": 0.7215,
      "step": 2275
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.1194897478175227e-05,
      "loss": 0.7419,
      "step": 2276
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.119041301414383e-05,
      "loss": 0.7421,
      "step": 2277
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.1185927717717718e-05,
      "loss": 0.6994,
      "step": 2278
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.1181441590284611e-05,
      "loss": 0.7758,
      "step": 2279
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.1176954633232488e-05,
      "loss": 0.7468,
      "step": 2280
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.117246684794958e-05,
      "loss": 0.7576,
      "step": 2281
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.1167978235824373e-05,
      "loss": 0.7398,
      "step": 2282
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.1163488798245616e-05,
      "loss": 0.7308,
      "step": 2283
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.1158998536602305e-05,
      "loss": 0.7368,
      "step": 2284
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.11545074522837e-05,
      "loss": 0.7536,
      "step": 2285
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.1150015546679307e-05,
      "loss": 0.7452,
      "step": 2286
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.1145522821178892e-05,
      "loss": 0.7234,
      "step": 2287
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.114102927717247e-05,
      "loss": 0.7479,
      "step": 2288
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.1136534916050312e-05,
      "loss": 0.7209,
      "step": 2289
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.1132039739202942e-05,
      "loss": 0.74,
      "step": 2290
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.1127543748021139e-05,
      "loss": 0.7369,
      "step": 2291
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.112304694389593e-05,
      "loss": 0.735,
      "step": 2292
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.1118549328218588e-05,
      "loss": 0.7341,
      "step": 2293
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.1114050902380649e-05,
      "loss": 0.7271,
      "step": 2294
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.1109551667773893e-05,
      "loss": 0.754,
      "step": 2295
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.1105051625790351e-05,
      "loss": 0.7252,
      "step": 2296
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.1100550777822305e-05,
      "loss": 0.7147,
      "step": 2297
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.1096049125262282e-05,
      "loss": 0.7106,
      "step": 2298
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.1091546669503066e-05,
      "loss": 0.7432,
      "step": 2299
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.1087043411937684e-05,
      "loss": 0.7027,
      "step": 2300
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.108253935395941e-05,
      "loss": 0.7326,
      "step": 2301
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.107803449696177e-05,
      "loss": 0.7178,
      "step": 2302
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.1073528842338531e-05,
      "loss": 0.7462,
      "step": 2303
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.1069022391483714e-05,
      "loss": 0.735,
      "step": 2304
    },
    {
      "epoch": 1.62,
      "eval_loss": 0.727583646774292,
      "eval_runtime": 30.9791,
      "eval_samples_per_second": 264.404,
      "eval_steps_per_second": 16.527,
      "step": 2304
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.1064515145791582e-05,
      "loss": 0.7195,
      "step": 2305
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.1060007106656644e-05,
      "loss": 0.6962,
      "step": 2306
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.1055498275473657e-05,
      "loss": 0.7392,
      "step": 2307
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.105098865363762e-05,
      "loss": 0.7351,
      "step": 2308
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.1046478242543778e-05,
      "loss": 0.695,
      "step": 2309
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.104196704358762e-05,
      "loss": 0.7381,
      "step": 2310
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.1037455058164883e-05,
      "loss": 0.7161,
      "step": 2311
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.1032942287671536e-05,
      "loss": 0.7397,
      "step": 2312
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.1028428733503808e-05,
      "loss": 0.7347,
      "step": 2313
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.1023914397058149e-05,
      "loss": 0.7075,
      "step": 2314
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.101939927973127e-05,
      "loss": 0.7167,
      "step": 2315
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.1014883382920116e-05,
      "loss": 0.7287,
      "step": 2316
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.1010366708021872e-05,
      "loss": 0.7557,
      "step": 2317
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.1005849256433963e-05,
      "loss": 0.7517,
      "step": 2318
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.1001331029554063e-05,
      "loss": 0.7665,
      "step": 2319
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.0996812028780073e-05,
      "loss": 0.717,
      "step": 2320
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.0992292255510144e-05,
      "loss": 0.71,
      "step": 2321
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.0987771711142658e-05,
      "loss": 0.7681,
      "step": 2322
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.0983250397076242e-05,
      "loss": 0.7272,
      "step": 2323
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.097872831470976e-05,
      "loss": 0.7432,
      "step": 2324
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.0974205465442308e-05,
      "loss": 0.7582,
      "step": 2325
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.0969681850673226e-05,
      "loss": 0.754,
      "step": 2326
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.0965157471802085e-05,
      "loss": 0.7306,
      "step": 2327
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.0960632330228702e-05,
      "loss": 0.7414,
      "step": 2328
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.0956106427353121e-05,
      "loss": 0.71,
      "step": 2329
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.0951579764575624e-05,
      "loss": 0.7331,
      "step": 2330
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.0947052343296725e-05,
      "loss": 0.7302,
      "step": 2331
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.0942524164917182e-05,
      "loss": 0.7414,
      "step": 2332
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.0937995230837976e-05,
      "loss": 0.7143,
      "step": 2333
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.0933465542460332e-05,
      "loss": 0.7425,
      "step": 2334
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.09289351011857e-05,
      "loss": 0.7318,
      "step": 2335
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0924403908415766e-05,
      "loss": 0.7602,
      "step": 2336
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0919871965552451e-05,
      "loss": 0.7189,
      "step": 2337
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0915339273997905e-05,
      "loss": 0.716,
      "step": 2338
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0910805835154512e-05,
      "loss": 0.6962,
      "step": 2339
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0906271650424882e-05,
      "loss": 0.7306,
      "step": 2340
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0901736721211867e-05,
      "loss": 0.7141,
      "step": 2341
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0897201048918535e-05,
      "loss": 0.6955,
      "step": 2342
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0892664634948194e-05,
      "loss": 0.7143,
      "step": 2343
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0888127480704376e-05,
      "loss": 0.7142,
      "step": 2344
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0883589587590847e-05,
      "loss": 0.6998,
      "step": 2345
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0879050957011597e-05,
      "loss": 0.7233,
      "step": 2346
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0874511590370845e-05,
      "loss": 0.7579,
      "step": 2347
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0869971489073044e-05,
      "loss": 0.7477,
      "step": 2348
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0865430654522866e-05,
      "loss": 0.7225,
      "step": 2349
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.086088908812521e-05,
      "loss": 0.7574,
      "step": 2350
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.085634679128521e-05,
      "loss": 0.6888,
      "step": 2351
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.0851803765408217e-05,
      "loss": 0.7288,
      "step": 2352
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.0847260011899814e-05,
      "loss": 0.7274,
      "step": 2353
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.0842715532165801e-05,
      "loss": 0.7468,
      "step": 2354
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.0838170327612212e-05,
      "loss": 0.7008,
      "step": 2355
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.0833624399645298e-05,
      "loss": 0.698,
      "step": 2356
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.0829077749671541e-05,
      "loss": 0.7183,
      "step": 2357
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.0824530379097639e-05,
      "loss": 0.7008,
      "step": 2358
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.0819982289330518e-05,
      "loss": 0.719,
      "step": 2359
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.0815433481777323e-05,
      "loss": 0.7424,
      "step": 2360
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.0810883957845425e-05,
      "loss": 0.699,
      "step": 2361
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.0806333718942411e-05,
      "loss": 0.7407,
      "step": 2362
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.0801782766476099e-05,
      "loss": 0.7518,
      "step": 2363
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.0797231101854515e-05,
      "loss": 0.7529,
      "step": 2364
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.0792678726485909e-05,
      "loss": 0.7302,
      "step": 2365
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.0788125641778764e-05,
      "loss": 0.7107,
      "step": 2366
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.0783571849141766e-05,
      "loss": 0.7368,
      "step": 2367
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.0779017349983829e-05,
      "loss": 0.709,
      "step": 2368
    },
    {
      "epoch": 1.67,
      "eval_loss": 0.7275509238243103,
      "eval_runtime": 31.1259,
      "eval_samples_per_second": 263.157,
      "eval_steps_per_second": 16.449,
      "step": 2368
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.077446214571408e-05,
      "loss": 0.7354,
      "step": 2369
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.0769906237741867e-05,
      "loss": 0.7667,
      "step": 2370
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.0765349627476755e-05,
      "loss": 0.7573,
      "step": 2371
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.0760792316328529e-05,
      "loss": 0.7231,
      "step": 2372
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.0756234305707188e-05,
      "loss": 0.7059,
      "step": 2373
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.0751675597022947e-05,
      "loss": 0.7185,
      "step": 2374
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.0747116191686242e-05,
      "loss": 0.7214,
      "step": 2375
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.0742556091107717e-05,
      "loss": 0.7603,
      "step": 2376
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.0737995296698234e-05,
      "loss": 0.7182,
      "step": 2377
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.0733433809868873e-05,
      "loss": 0.716,
      "step": 2378
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.0728871632030927e-05,
      "loss": 0.7278,
      "step": 2379
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.0724308764595898e-05,
      "loss": 0.7526,
      "step": 2380
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.0719745208975506e-05,
      "loss": 0.7555,
      "step": 2381
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.0715180966581684e-05,
      "loss": 0.7315,
      "step": 2382
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.0710616038826573e-05,
      "loss": 0.7188,
      "step": 2383
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.0706050427122534e-05,
      "loss": 0.7428,
      "step": 2384
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.0701484132882131e-05,
      "loss": 0.7239,
      "step": 2385
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.0696917157518144e-05,
      "loss": 0.7276,
      "step": 2386
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.0692349502443564e-05,
      "loss": 0.7198,
      "step": 2387
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.068778116907159e-05,
      "loss": 0.7343,
      "step": 2388
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.0683212158815632e-05,
      "loss": 0.7498,
      "step": 2389
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.0678642473089312e-05,
      "loss": 0.6937,
      "step": 2390
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.0674072113306456e-05,
      "loss": 0.7463,
      "step": 2391
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.0669501080881099e-05,
      "loss": 0.6777,
      "step": 2392
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.0664929377227488e-05,
      "loss": 0.7381,
      "step": 2393
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.0660357003760074e-05,
      "loss": 0.6953,
      "step": 2394
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.0655783961893521e-05,
      "loss": 0.707,
      "step": 2395
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.0651210253042695e-05,
      "loss": 0.7148,
      "step": 2396
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.0646635878622666e-05,
      "loss": 0.7607,
      "step": 2397
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.0642060840048713e-05,
      "loss": 0.7032,
      "step": 2398
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.0637485138736324e-05,
      "loss": 0.7661,
      "step": 2399
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.0632908776101187e-05,
      "loss": 0.7304,
      "step": 2400
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.0628331753559195e-05,
      "loss": 0.7211,
      "step": 2401
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.0623754072526446e-05,
      "loss": 0.7264,
      "step": 2402
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.061917573441924e-05,
      "loss": 0.7642,
      "step": 2403
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.061459674065409e-05,
      "loss": 0.7285,
      "step": 2404
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.06100170926477e-05,
      "loss": 0.7274,
      "step": 2405
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.0605436791816979e-05,
      "loss": 0.7537,
      "step": 2406
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0600855839579041e-05,
      "loss": 0.735,
      "step": 2407
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0596274237351201e-05,
      "loss": 0.6972,
      "step": 2408
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0591691986550972e-05,
      "loss": 0.7341,
      "step": 2409
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0587109088596072e-05,
      "loss": 0.7498,
      "step": 2410
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0582525544904415e-05,
      "loss": 0.7422,
      "step": 2411
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0577941356894118e-05,
      "loss": 0.7536,
      "step": 2412
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0573356525983498e-05,
      "loss": 0.7072,
      "step": 2413
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0568771053591067e-05,
      "loss": 0.7349,
      "step": 2414
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0564184941135539e-05,
      "loss": 0.7253,
      "step": 2415
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0559598190035822e-05,
      "loss": 0.7129,
      "step": 2416
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0555010801711029e-05,
      "loss": 0.764,
      "step": 2417
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0550422777580464e-05,
      "loss": 0.7209,
      "step": 2418
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0545834119063625e-05,
      "loss": 0.7278,
      "step": 2419
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0541244827580213e-05,
      "loss": 0.735,
      "step": 2420
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0536654904550123e-05,
      "loss": 0.759,
      "step": 2421
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.0532064351393446e-05,
      "loss": 0.6917,
      "step": 2422
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.052747316953046e-05,
      "loss": 0.7422,
      "step": 2423
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.0522881360381655e-05,
      "loss": 0.7096,
      "step": 2424
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.0518288925367694e-05,
      "loss": 0.7322,
      "step": 2425
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.051369586590945e-05,
      "loss": 0.7052,
      "step": 2426
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.0509102183427981e-05,
      "loss": 0.7683,
      "step": 2427
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.0504507879344539e-05,
      "loss": 0.729,
      "step": 2428
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.049991295508057e-05,
      "loss": 0.7219,
      "step": 2429
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.0495317412057711e-05,
      "loss": 0.7218,
      "step": 2430
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.0490721251697789e-05,
      "loss": 0.7022,
      "step": 2431
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.0486124475422824e-05,
      "loss": 0.7354,
      "step": 2432
    },
    {
      "epoch": 1.71,
      "eval_loss": 0.7266652584075928,
      "eval_runtime": 30.9175,
      "eval_samples_per_second": 264.931,
      "eval_steps_per_second": 16.56,
      "step": 2432
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.0481527084655027e-05,
      "loss": 0.7049,
      "step": 2433
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.04769290808168e-05,
      "loss": 0.7096,
      "step": 2434
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.0472330465330731e-05,
      "loss": 0.7298,
      "step": 2435
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.0467731239619598e-05,
      "loss": 0.7248,
      "step": 2436
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.0463131405106368e-05,
      "loss": 0.7303,
      "step": 2437
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.04585309632142e-05,
      "loss": 0.7404,
      "step": 2438
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.0453929915366437e-05,
      "loss": 0.7275,
      "step": 2439
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.0449328262986613e-05,
      "loss": 0.7334,
      "step": 2440
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.0444726007498438e-05,
      "loss": 0.7408,
      "step": 2441
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.0440123150325827e-05,
      "loss": 0.7333,
      "step": 2442
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.0435519692892865e-05,
      "loss": 0.7358,
      "step": 2443
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.0430915636623832e-05,
      "loss": 0.7297,
      "step": 2444
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.0426310982943191e-05,
      "loss": 0.7083,
      "step": 2445
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.0421705733275592e-05,
      "loss": 0.7226,
      "step": 2446
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.041709988904586e-05,
      "loss": 0.7564,
      "step": 2447
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.0412493451679011e-05,
      "loss": 0.7274,
      "step": 2448
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.0407886422600249e-05,
      "loss": 0.7396,
      "step": 2449
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.0403278803234953e-05,
      "loss": 0.7612,
      "step": 2450
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.0398670595008689e-05,
      "loss": 0.761,
      "step": 2451
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.0394061799347202e-05,
      "loss": 0.7252,
      "step": 2452
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.0389452417676423e-05,
      "loss": 0.6821,
      "step": 2453
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.038484245142246e-05,
      "loss": 0.7384,
      "step": 2454
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.0380231902011605e-05,
      "loss": 0.7243,
      "step": 2455
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.037562077087033e-05,
      "loss": 0.7671,
      "step": 2456
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.0371009059425286e-05,
      "loss": 0.7342,
      "step": 2457
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.0366396769103304e-05,
      "loss": 0.7351,
      "step": 2458
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.0361783901331391e-05,
      "loss": 0.6922,
      "step": 2459
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.035717045753674e-05,
      "loss": 0.7043,
      "step": 2460
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.0352556439146711e-05,
      "loss": 0.7208,
      "step": 2461
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.034794184758886e-05,
      "loss": 0.7201,
      "step": 2462
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.0343326684290897e-05,
      "loss": 0.7239,
      "step": 2463
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.0338710950680729e-05,
      "loss": 0.7548,
      "step": 2464
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.0334094648186427e-05,
      "loss": 0.7372,
      "step": 2465
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.0329477778236242e-05,
      "loss": 0.7375,
      "step": 2466
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.0324860342258605e-05,
      "loss": 0.7699,
      "step": 2467
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.0320242341682117e-05,
      "loss": 0.759,
      "step": 2468
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.0315623777935548e-05,
      "loss": 0.7256,
      "step": 2469
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.0311004652447854e-05,
      "loss": 0.7519,
      "step": 2470
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.0306384966648161e-05,
      "loss": 0.7467,
      "step": 2471
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.0301764721965765e-05,
      "loss": 0.7123,
      "step": 2472
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.029714391983014e-05,
      "loss": 0.7088,
      "step": 2473
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.0292522561670926e-05,
      "loss": 0.715,
      "step": 2474
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.0287900648917936e-05,
      "loss": 0.7242,
      "step": 2475
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.0283278183001163e-05,
      "loss": 0.7196,
      "step": 2476
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.0278655165350763e-05,
      "loss": 0.7061,
      "step": 2477
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.0274031597397062e-05,
      "loss": 0.7513,
      "step": 2478
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.0269407480570564e-05,
      "loss": 0.7076,
      "step": 2479
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.0264782816301936e-05,
      "loss": 0.7204,
      "step": 2480
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.0260157606022014e-05,
      "loss": 0.7426,
      "step": 2481
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.0255531851161808e-05,
      "loss": 0.722,
      "step": 2482
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.025090555315249e-05,
      "loss": 0.7295,
      "step": 2483
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.0246278713425411e-05,
      "loss": 0.7247,
      "step": 2484
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.0241651333412077e-05,
      "loss": 0.7238,
      "step": 2485
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.0237023414544167e-05,
      "loss": 0.7577,
      "step": 2486
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.0232394958253527e-05,
      "loss": 0.7504,
      "step": 2487
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.0227765965972168e-05,
      "loss": 0.7508,
      "step": 2488
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.0223136439132269e-05,
      "loss": 0.7255,
      "step": 2489
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.0218506379166173e-05,
      "loss": 0.6862,
      "step": 2490
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.0213875787506383e-05,
      "loss": 0.7273,
      "step": 2491
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.0209244665585577e-05,
      "loss": 0.7285,
      "step": 2492
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.0204613014836588e-05,
      "loss": 0.7513,
      "step": 2493
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.0199980836692417e-05,
      "loss": 0.7571,
      "step": 2494
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.019534813258623e-05,
      "loss": 0.7297,
      "step": 2495
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.0190714903951346e-05,
      "loss": 0.7527,
      "step": 2496
    },
    {
      "epoch": 1.76,
      "eval_loss": 0.7260441184043884,
      "eval_runtime": 30.9237,
      "eval_samples_per_second": 264.878,
      "eval_steps_per_second": 16.557,
      "step": 2496
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.0186081152221259e-05,
      "loss": 0.7124,
      "step": 2497
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.0181446878829615e-05,
      "loss": 0.711,
      "step": 2498
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.0176812085210225e-05,
      "loss": 0.7088,
      "step": 2499
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.0172176772797067e-05,
      "loss": 0.7142,
      "step": 2500
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.0167540943024272e-05,
      "loss": 0.7246,
      "step": 2501
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.0162904597326127e-05,
      "loss": 0.7507,
      "step": 2502
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.0158267737137092e-05,
      "loss": 0.729,
      "step": 2503
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.0153630363891772e-05,
      "loss": 0.7353,
      "step": 2504
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.0148992479024942e-05,
      "loss": 0.7621,
      "step": 2505
    },
    {
      "epoch": 1.76,
      "learning_rate": 1.0144354083971531e-05,
      "loss": 0.7374,
      "step": 2506
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.0139715180166622e-05,
      "loss": 0.7586,
      "step": 2507
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.0135075769045459e-05,
      "loss": 0.7212,
      "step": 2508
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.0130435852043448e-05,
      "loss": 0.6939,
      "step": 2509
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.0125795430596141e-05,
      "loss": 0.7597,
      "step": 2510
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.0121154506139254e-05,
      "loss": 0.732,
      "step": 2511
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.011651308010866e-05,
      "loss": 0.7317,
      "step": 2512
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.0111871153940374e-05,
      "loss": 0.731,
      "step": 2513
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.0107228729070582e-05,
      "loss": 0.7238,
      "step": 2514
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.0102585806935617e-05,
      "loss": 0.732,
      "step": 2515
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.0097942388971961e-05,
      "loss": 0.7431,
      "step": 2516
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.0093298476616261e-05,
      "loss": 0.708,
      "step": 2517
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.0088654071305307e-05,
      "loss": 0.7172,
      "step": 2518
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.0084009174476046e-05,
      "loss": 0.7508,
      "step": 2519
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.0079363787565577e-05,
      "loss": 0.7259,
      "step": 2520
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.0074717912011147e-05,
      "loss": 0.7387,
      "step": 2521
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.0070071549250159e-05,
      "loss": 0.7101,
      "step": 2522
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.0065424700720168e-05,
      "loss": 0.7136,
      "step": 2523
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.0060777367858868e-05,
      "loss": 0.7066,
      "step": 2524
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.0056129552104118e-05,
      "loss": 0.7208,
      "step": 2525
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.0051481254893916e-05,
      "loss": 0.7681,
      "step": 2526
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.0046832477666412e-05,
      "loss": 0.7322,
      "step": 2527
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.0042183221859906e-05,
      "loss": 0.7282,
      "step": 2528
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.0037533488912845e-05,
      "loss": 0.7221,
      "step": 2529
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.003288328026382e-05,
      "loss": 0.7044,
      "step": 2530
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.002823259735158e-05,
      "loss": 0.7296,
      "step": 2531
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.0023581441615006e-05,
      "loss": 0.6937,
      "step": 2532
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.0018929814493134e-05,
      "loss": 0.7367,
      "step": 2533
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.001427771742515e-05,
      "loss": 0.7373,
      "step": 2534
    },
    {
      "epoch": 1.79,
      "learning_rate": 1.000962515185037e-05,
      "loss": 0.7159,
      "step": 2535
    },
    {
      "epoch": 1.79,
      "learning_rate": 1.0004972119208275e-05,
      "loss": 0.7352,
      "step": 2536
    },
    {
      "epoch": 1.79,
      "learning_rate": 1.0000318620938472e-05,
      "loss": 0.6997,
      "step": 2537
    },
    {
      "epoch": 1.79,
      "learning_rate": 9.995664658480724e-06,
      "loss": 0.7192,
      "step": 2538
    },
    {
      "epoch": 1.79,
      "learning_rate": 9.991010233274936e-06,
      "loss": 0.7351,
      "step": 2539
    },
    {
      "epoch": 1.79,
      "learning_rate": 9.986355346761153e-06,
      "loss": 0.697,
      "step": 2540
    },
    {
      "epoch": 1.79,
      "learning_rate": 9.98170000037956e-06,
      "loss": 0.7371,
      "step": 2541
    },
    {
      "epoch": 1.79,
      "learning_rate": 9.977044195570495e-06,
      "loss": 0.755,
      "step": 2542
    },
    {
      "epoch": 1.79,
      "learning_rate": 9.972387933774424e-06,
      "loss": 0.7186,
      "step": 2543
    },
    {
      "epoch": 1.79,
      "learning_rate": 9.967731216431967e-06,
      "loss": 0.6991,
      "step": 2544
    },
    {
      "epoch": 1.79,
      "learning_rate": 9.963074044983879e-06,
      "loss": 0.7169,
      "step": 2545
    },
    {
      "epoch": 1.79,
      "learning_rate": 9.958416420871046e-06,
      "loss": 0.7102,
      "step": 2546
    },
    {
      "epoch": 1.79,
      "learning_rate": 9.953758345534517e-06,
      "loss": 0.7143,
      "step": 2547
    },
    {
      "epoch": 1.79,
      "learning_rate": 9.949099820415462e-06,
      "loss": 0.7495,
      "step": 2548
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.944440846955195e-06,
      "loss": 0.7196,
      "step": 2549
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.939781426595171e-06,
      "loss": 0.7277,
      "step": 2550
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.93512156077698e-06,
      "loss": 0.7648,
      "step": 2551
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.930461250942353e-06,
      "loss": 0.7467,
      "step": 2552
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.925800498533156e-06,
      "loss": 0.715,
      "step": 2553
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.921139304991391e-06,
      "loss": 0.7203,
      "step": 2554
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.916477671759202e-06,
      "loss": 0.7251,
      "step": 2555
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.911815600278863e-06,
      "loss": 0.7486,
      "step": 2556
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.907153091992786e-06,
      "loss": 0.72,
      "step": 2557
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.902490148343517e-06,
      "loss": 0.7338,
      "step": 2558
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.89782677077374e-06,
      "loss": 0.7563,
      "step": 2559
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.893162960726271e-06,
      "loss": 0.7324,
      "step": 2560
    },
    {
      "epoch": 1.8,
      "eval_loss": 0.7253321409225464,
      "eval_runtime": 30.9177,
      "eval_samples_per_second": 264.929,
      "eval_steps_per_second": 16.56,
      "step": 2560
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.888498719644059e-06,
      "loss": 0.7208,
      "step": 2561
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.883834048970185e-06,
      "loss": 0.765,
      "step": 2562
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.879168950147868e-06,
      "loss": 0.7115,
      "step": 2563
    },
    {
      "epoch": 1.81,
      "learning_rate": 9.874503424620452e-06,
      "loss": 0.7522,
      "step": 2564
    },
    {
      "epoch": 1.81,
      "learning_rate": 9.869837473831424e-06,
      "loss": 0.709,
      "step": 2565
    },
    {
      "epoch": 1.81,
      "learning_rate": 9.86517109922439e-06,
      "loss": 0.7259,
      "step": 2566
    },
    {
      "epoch": 1.81,
      "learning_rate": 9.860504302243101e-06,
      "loss": 0.7356,
      "step": 2567
    },
    {
      "epoch": 1.81,
      "learning_rate": 9.855837084331423e-06,
      "loss": 0.7449,
      "step": 2568
    },
    {
      "epoch": 1.81,
      "learning_rate": 9.851169446933364e-06,
      "loss": 0.7371,
      "step": 2569
    },
    {
      "epoch": 1.81,
      "learning_rate": 9.846501391493055e-06,
      "loss": 0.7746,
      "step": 2570
    },
    {
      "epoch": 1.81,
      "learning_rate": 9.841832919454763e-06,
      "loss": 0.7352,
      "step": 2571
    },
    {
      "epoch": 1.81,
      "learning_rate": 9.83716403226288e-06,
      "loss": 0.7107,
      "step": 2572
    },
    {
      "epoch": 1.81,
      "learning_rate": 9.832494731361923e-06,
      "loss": 0.7157,
      "step": 2573
    },
    {
      "epoch": 1.81,
      "learning_rate": 9.827825018196544e-06,
      "loss": 0.7219,
      "step": 2574
    },
    {
      "epoch": 1.81,
      "learning_rate": 9.823154894211515e-06,
      "loss": 0.7438,
      "step": 2575
    },
    {
      "epoch": 1.81,
      "learning_rate": 9.818484360851745e-06,
      "loss": 0.729,
      "step": 2576
    },
    {
      "epoch": 1.81,
      "learning_rate": 9.81381341956226e-06,
      "loss": 0.7328,
      "step": 2577
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.809142071788216e-06,
      "loss": 0.7555,
      "step": 2578
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.8044703189749e-06,
      "loss": 0.7505,
      "step": 2579
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.799798162567712e-06,
      "loss": 0.7304,
      "step": 2580
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.795125604012187e-06,
      "loss": 0.7506,
      "step": 2581
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.790452644753984e-06,
      "loss": 0.7541,
      "step": 2582
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.78577928623888e-06,
      "loss": 0.7308,
      "step": 2583
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.781105529912779e-06,
      "loss": 0.7201,
      "step": 2584
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.776431377221708e-06,
      "loss": 0.7382,
      "step": 2585
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.771756829611822e-06,
      "loss": 0.7046,
      "step": 2586
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.767081888529388e-06,
      "loss": 0.74,
      "step": 2587
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.7624065554208e-06,
      "loss": 0.74,
      "step": 2588
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.757730831732573e-06,
      "loss": 0.7239,
      "step": 2589
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.753054718911344e-06,
      "loss": 0.7442,
      "step": 2590
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.748378218403867e-06,
      "loss": 0.6849,
      "step": 2591
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.74370133165702e-06,
      "loss": 0.7094,
      "step": 2592
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.739024060117798e-06,
      "loss": 0.7131,
      "step": 2593
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.734346405233317e-06,
      "loss": 0.7333,
      "step": 2594
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.729668368450808e-06,
      "loss": 0.7792,
      "step": 2595
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.724989951217625e-06,
      "loss": 0.7273,
      "step": 2596
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.720311154981233e-06,
      "loss": 0.7341,
      "step": 2597
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.715631981189221e-06,
      "loss": 0.7632,
      "step": 2598
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.710952431289293e-06,
      "loss": 0.7462,
      "step": 2599
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.70627250672927e-06,
      "loss": 0.7221,
      "step": 2600
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.701592208957082e-06,
      "loss": 0.7064,
      "step": 2601
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.696911539420785e-06,
      "loss": 0.7205,
      "step": 2602
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.69223049956854e-06,
      "loss": 0.719,
      "step": 2603
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.687549090848633e-06,
      "loss": 0.7265,
      "step": 2604
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.682867314709455e-06,
      "loss": 0.7331,
      "step": 2605
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.678185172599518e-06,
      "loss": 0.7306,
      "step": 2606
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.67350266596744e-06,
      "loss": 0.745,
      "step": 2607
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.668819796261955e-06,
      "loss": 0.7286,
      "step": 2608
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.664136564931914e-06,
      "loss": 0.7333,
      "step": 2609
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.65945297342627e-06,
      "loss": 0.7357,
      "step": 2610
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.6547690231941e-06,
      "loss": 0.7027,
      "step": 2611
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.650084715684577e-06,
      "loss": 0.7116,
      "step": 2612
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.645400052346998e-06,
      "loss": 0.6982,
      "step": 2613
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.640715034630761e-06,
      "loss": 0.7226,
      "step": 2614
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.63602966398538e-06,
      "loss": 0.738,
      "step": 2615
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.631343941860476e-06,
      "loss": 0.7391,
      "step": 2616
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.626657869705778e-06,
      "loss": 0.7452,
      "step": 2617
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.621971448971121e-06,
      "loss": 0.717,
      "step": 2618
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.617284681106452e-06,
      "loss": 0.7125,
      "step": 2619
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.612597567561823e-06,
      "loss": 0.7345,
      "step": 2620
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.607910109787395e-06,
      "loss": 0.7499,
      "step": 2621
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.603222309233433e-06,
      "loss": 0.703,
      "step": 2622
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.59853416735031e-06,
      "loss": 0.7269,
      "step": 2623
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.593845685588502e-06,
      "loss": 0.7124,
      "step": 2624
    },
    {
      "epoch": 1.85,
      "eval_loss": 0.724940836429596,
      "eval_runtime": 30.9396,
      "eval_samples_per_second": 264.742,
      "eval_steps_per_second": 16.548,
      "step": 2624
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.589156865398594e-06,
      "loss": 0.7005,
      "step": 2625
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.584467708231275e-06,
      "loss": 0.7332,
      "step": 2626
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.579778215537334e-06,
      "loss": 0.6968,
      "step": 2627
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.57508838876767e-06,
      "loss": 0.7417,
      "step": 2628
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.570398229373278e-06,
      "loss": 0.7399,
      "step": 2629
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.565707738805264e-06,
      "loss": 0.7379,
      "step": 2630
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.561016918514828e-06,
      "loss": 0.6968,
      "step": 2631
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.556325769953283e-06,
      "loss": 0.7237,
      "step": 2632
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.55163429457203e-06,
      "loss": 0.7344,
      "step": 2633
    },
    {
      "epoch": 1.85,
      "learning_rate": 9.546942493822583e-06,
      "loss": 0.7307,
      "step": 2634
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.54225036915655e-06,
      "loss": 0.7204,
      "step": 2635
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.537557922025645e-06,
      "loss": 0.7229,
      "step": 2636
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.53286515388167e-06,
      "loss": 0.7142,
      "step": 2637
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.528172066176541e-06,
      "loss": 0.7199,
      "step": 2638
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.523478660362265e-06,
      "loss": 0.7018,
      "step": 2639
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.518784937890948e-06,
      "loss": 0.7169,
      "step": 2640
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.514090900214792e-06,
      "loss": 0.7537,
      "step": 2641
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.509396548786099e-06,
      "loss": 0.7319,
      "step": 2642
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.504701885057271e-06,
      "loss": 0.766,
      "step": 2643
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.500006910480806e-06,
      "loss": 0.7087,
      "step": 2644
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.495311626509289e-06,
      "loss": 0.7372,
      "step": 2645
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.490616034595411e-06,
      "loss": 0.7412,
      "step": 2646
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.485920136191954e-06,
      "loss": 0.7307,
      "step": 2647
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.481223932751798e-06,
      "loss": 0.7649,
      "step": 2648
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.476527425727913e-06,
      "loss": 0.7534,
      "step": 2649
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.471830616573367e-06,
      "loss": 0.7612,
      "step": 2650
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.467133506741315e-06,
      "loss": 0.7344,
      "step": 2651
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.462436097685013e-06,
      "loss": 0.7098,
      "step": 2652
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.457738390857805e-06,
      "loss": 0.6883,
      "step": 2653
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.45304038771313e-06,
      "loss": 0.7166,
      "step": 2654
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.448342089704518e-06,
      "loss": 0.7548,
      "step": 2655
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.443643498285584e-06,
      "loss": 0.7202,
      "step": 2656
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.438944614910042e-06,
      "loss": 0.684,
      "step": 2657
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.434245441031692e-06,
      "loss": 0.7359,
      "step": 2658
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.429545978104425e-06,
      "loss": 0.7505,
      "step": 2659
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.424846227582223e-06,
      "loss": 0.6805,
      "step": 2660
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.420146190919157e-06,
      "loss": 0.7334,
      "step": 2661
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.41544586956938e-06,
      "loss": 0.7227,
      "step": 2662
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.410745264987142e-06,
      "loss": 0.7292,
      "step": 2663
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.406044378626775e-06,
      "loss": 0.7205,
      "step": 2664
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.4013432119427e-06,
      "loss": 0.7427,
      "step": 2665
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.396641766389429e-06,
      "loss": 0.7132,
      "step": 2666
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.391940043421548e-06,
      "loss": 0.7406,
      "step": 2667
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.387238044493742e-06,
      "loss": 0.7033,
      "step": 2668
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.382535771060776e-06,
      "loss": 0.7412,
      "step": 2669
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.377833224577498e-06,
      "loss": 0.7514,
      "step": 2670
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.373130406498842e-06,
      "loss": 0.7296,
      "step": 2671
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.368427318279832e-06,
      "loss": 0.7337,
      "step": 2672
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.363723961375566e-06,
      "loss": 0.7448,
      "step": 2673
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.359020337241227e-06,
      "loss": 0.7759,
      "step": 2674
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.354316447332086e-06,
      "loss": 0.714,
      "step": 2675
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.349612293103494e-06,
      "loss": 0.7242,
      "step": 2676
    },
    {
      "epoch": 1.89,
      "learning_rate": 9.344907876010881e-06,
      "loss": 0.7366,
      "step": 2677
    },
    {
      "epoch": 1.89,
      "learning_rate": 9.340203197509763e-06,
      "loss": 0.7478,
      "step": 2678
    },
    {
      "epoch": 1.89,
      "learning_rate": 9.33549825905573e-06,
      "loss": 0.7477,
      "step": 2679
    },
    {
      "epoch": 1.89,
      "learning_rate": 9.330793062104457e-06,
      "loss": 0.7059,
      "step": 2680
    },
    {
      "epoch": 1.89,
      "learning_rate": 9.326087608111704e-06,
      "loss": 0.7352,
      "step": 2681
    },
    {
      "epoch": 1.89,
      "learning_rate": 9.321381898533299e-06,
      "loss": 0.7208,
      "step": 2682
    },
    {
      "epoch": 1.89,
      "learning_rate": 9.316675934825161e-06,
      "loss": 0.7322,
      "step": 2683
    },
    {
      "epoch": 1.89,
      "learning_rate": 9.311969718443275e-06,
      "loss": 0.7362,
      "step": 2684
    },
    {
      "epoch": 1.89,
      "learning_rate": 9.307263250843715e-06,
      "loss": 0.6941,
      "step": 2685
    },
    {
      "epoch": 1.89,
      "learning_rate": 9.302556533482632e-06,
      "loss": 0.7357,
      "step": 2686
    },
    {
      "epoch": 1.89,
      "learning_rate": 9.297849567816243e-06,
      "loss": 0.708,
      "step": 2687
    },
    {
      "epoch": 1.89,
      "learning_rate": 9.293142355300857e-06,
      "loss": 0.7507,
      "step": 2688
    },
    {
      "epoch": 1.89,
      "eval_loss": 0.7243407368659973,
      "eval_runtime": 30.9463,
      "eval_samples_per_second": 264.685,
      "eval_steps_per_second": 16.545,
      "step": 2688
    },
    {
      "epoch": 1.89,
      "learning_rate": 9.288434897392846e-06,
      "loss": 0.6912,
      "step": 2689
    },
    {
      "epoch": 1.89,
      "learning_rate": 9.283727195548668e-06,
      "loss": 0.7572,
      "step": 2690
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.279019251224853e-06,
      "loss": 0.7217,
      "step": 2691
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.274311065878003e-06,
      "loss": 0.6811,
      "step": 2692
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.269602640964799e-06,
      "loss": 0.7128,
      "step": 2693
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.264893977941999e-06,
      "loss": 0.7278,
      "step": 2694
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.260185078266417e-06,
      "loss": 0.7184,
      "step": 2695
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.255475943394966e-06,
      "loss": 0.7615,
      "step": 2696
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.250766574784613e-06,
      "loss": 0.7397,
      "step": 2697
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.246056973892402e-06,
      "loss": 0.712,
      "step": 2698
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.241347142175455e-06,
      "loss": 0.717,
      "step": 2699
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.236637081090957e-06,
      "loss": 0.7218,
      "step": 2700
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.231926792096167e-06,
      "loss": 0.7364,
      "step": 2701
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.227216276648417e-06,
      "loss": 0.7508,
      "step": 2702
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.222505536205109e-06,
      "loss": 0.7223,
      "step": 2703
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.217794572223708e-06,
      "loss": 0.7263,
      "step": 2704
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.213083386161757e-06,
      "loss": 0.7078,
      "step": 2705
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.208371979476859e-06,
      "loss": 0.7391,
      "step": 2706
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.203660353626696e-06,
      "loss": 0.7089,
      "step": 2707
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.198948510069005e-06,
      "loss": 0.7267,
      "step": 2708
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.1942364502616e-06,
      "loss": 0.7295,
      "step": 2709
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.189524175662361e-06,
      "loss": 0.735,
      "step": 2710
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.18481168772923e-06,
      "loss": 0.7125,
      "step": 2711
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.180098987920217e-06,
      "loss": 0.7305,
      "step": 2712
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.175386077693398e-06,
      "loss": 0.7309,
      "step": 2713
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.170672958506913e-06,
      "loss": 0.7544,
      "step": 2714
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.165959631818973e-06,
      "loss": 0.7351,
      "step": 2715
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.161246099087846e-06,
      "loss": 0.7095,
      "step": 2716
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.156532361771858e-06,
      "loss": 0.7002,
      "step": 2717
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.151818421329413e-06,
      "loss": 0.7463,
      "step": 2718
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.147104279218972e-06,
      "loss": 0.7264,
      "step": 2719
    },
    {
      "epoch": 1.92,
      "learning_rate": 9.14238993689905e-06,
      "loss": 0.7164,
      "step": 2720
    },
    {
      "epoch": 1.92,
      "learning_rate": 9.137675395828243e-06,
      "loss": 0.741,
      "step": 2721
    },
    {
      "epoch": 1.92,
      "learning_rate": 9.132960657465186e-06,
      "loss": 0.7296,
      "step": 2722
    },
    {
      "epoch": 1.92,
      "learning_rate": 9.128245723268587e-06,
      "loss": 0.7335,
      "step": 2723
    },
    {
      "epoch": 1.92,
      "learning_rate": 9.123530594697216e-06,
      "loss": 0.7261,
      "step": 2724
    },
    {
      "epoch": 1.92,
      "learning_rate": 9.1188152732099e-06,
      "loss": 0.7122,
      "step": 2725
    },
    {
      "epoch": 1.92,
      "learning_rate": 9.114099760265525e-06,
      "loss": 0.7386,
      "step": 2726
    },
    {
      "epoch": 1.92,
      "learning_rate": 9.109384057323037e-06,
      "loss": 0.7453,
      "step": 2727
    },
    {
      "epoch": 1.92,
      "learning_rate": 9.104668165841442e-06,
      "loss": 0.7452,
      "step": 2728
    },
    {
      "epoch": 1.92,
      "learning_rate": 9.099952087279804e-06,
      "loss": 0.7463,
      "step": 2729
    },
    {
      "epoch": 1.92,
      "learning_rate": 9.095235823097244e-06,
      "loss": 0.7309,
      "step": 2730
    },
    {
      "epoch": 1.92,
      "learning_rate": 9.09051937475294e-06,
      "loss": 0.7029,
      "step": 2731
    },
    {
      "epoch": 1.92,
      "learning_rate": 9.08580274370613e-06,
      "loss": 0.7391,
      "step": 2732
    },
    {
      "epoch": 1.92,
      "learning_rate": 9.081085931416102e-06,
      "loss": 0.7499,
      "step": 2733
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.076368939342206e-06,
      "loss": 0.7239,
      "step": 2734
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.071651768943846e-06,
      "loss": 0.7538,
      "step": 2735
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.06693442168048e-06,
      "loss": 0.7526,
      "step": 2736
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.062216899011625e-06,
      "loss": 0.7038,
      "step": 2737
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.057499202396845e-06,
      "loss": 0.7173,
      "step": 2738
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.052781333295768e-06,
      "loss": 0.7236,
      "step": 2739
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.048063293168064e-06,
      "loss": 0.7455,
      "step": 2740
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.04334508347346e-06,
      "loss": 0.7162,
      "step": 2741
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.038626705671742e-06,
      "loss": 0.7157,
      "step": 2742
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.03390816122274e-06,
      "loss": 0.7093,
      "step": 2743
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.029189451586337e-06,
      "loss": 0.7049,
      "step": 2744
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.02447057822247e-06,
      "loss": 0.7274,
      "step": 2745
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.019751542591124e-06,
      "loss": 0.7337,
      "step": 2746
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.015032346152333e-06,
      "loss": 0.7117,
      "step": 2747
    },
    {
      "epoch": 1.94,
      "learning_rate": 9.010312990366186e-06,
      "loss": 0.7507,
      "step": 2748
    },
    {
      "epoch": 1.94,
      "learning_rate": 9.00559347669282e-06,
      "loss": 0.719,
      "step": 2749
    },
    {
      "epoch": 1.94,
      "learning_rate": 9.000873806592416e-06,
      "loss": 0.7178,
      "step": 2750
    },
    {
      "epoch": 1.94,
      "learning_rate": 8.996153981525208e-06,
      "loss": 0.7179,
      "step": 2751
    },
    {
      "epoch": 1.94,
      "learning_rate": 8.991434002951474e-06,
      "loss": 0.7014,
      "step": 2752
    },
    {
      "epoch": 1.94,
      "eval_loss": 0.7242591381072998,
      "eval_runtime": 30.946,
      "eval_samples_per_second": 264.687,
      "eval_steps_per_second": 16.545,
      "step": 2752
    },
    {
      "epoch": 1.94,
      "learning_rate": 8.986713872331543e-06,
      "loss": 0.7335,
      "step": 2753
    },
    {
      "epoch": 1.94,
      "learning_rate": 8.981993591125794e-06,
      "loss": 0.6946,
      "step": 2754
    },
    {
      "epoch": 1.94,
      "learning_rate": 8.977273160794642e-06,
      "loss": 0.7225,
      "step": 2755
    },
    {
      "epoch": 1.94,
      "learning_rate": 8.972552582798554e-06,
      "loss": 0.7117,
      "step": 2756
    },
    {
      "epoch": 1.94,
      "learning_rate": 8.967831858598051e-06,
      "loss": 0.7395,
      "step": 2757
    },
    {
      "epoch": 1.94,
      "learning_rate": 8.963110989653686e-06,
      "loss": 0.7101,
      "step": 2758
    },
    {
      "epoch": 1.94,
      "learning_rate": 8.958389977426067e-06,
      "loss": 0.723,
      "step": 2759
    },
    {
      "epoch": 1.94,
      "learning_rate": 8.953668823375838e-06,
      "loss": 0.7172,
      "step": 2760
    },
    {
      "epoch": 1.94,
      "learning_rate": 8.94894752896369e-06,
      "loss": 0.7198,
      "step": 2761
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.94422609565036e-06,
      "loss": 0.7096,
      "step": 2762
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.939504524896625e-06,
      "loss": 0.7273,
      "step": 2763
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.934782818163305e-06,
      "loss": 0.7393,
      "step": 2764
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.930060976911264e-06,
      "loss": 0.7092,
      "step": 2765
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.925339002601407e-06,
      "loss": 0.7215,
      "step": 2766
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.920616896694675e-06,
      "loss": 0.7164,
      "step": 2767
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.915894660652059e-06,
      "loss": 0.7288,
      "step": 2768
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.911172295934583e-06,
      "loss": 0.7355,
      "step": 2769
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.906449804003313e-06,
      "loss": 0.7277,
      "step": 2770
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.901727186319358e-06,
      "loss": 0.7203,
      "step": 2771
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.897004444343857e-06,
      "loss": 0.7219,
      "step": 2772
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.892281579537997e-06,
      "loss": 0.7376,
      "step": 2773
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.887558593362995e-06,
      "loss": 0.7415,
      "step": 2774
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.882835487280114e-06,
      "loss": 0.7022,
      "step": 2775
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.878112262750644e-06,
      "loss": 0.7311,
      "step": 2776
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.873388921235924e-06,
      "loss": 0.7698,
      "step": 2777
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.868665464197318e-06,
      "loss": 0.7084,
      "step": 2778
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.863941893096232e-06,
      "loss": 0.7242,
      "step": 2779
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.859218209394103e-06,
      "loss": 0.7479,
      "step": 2780
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.854494414552405e-06,
      "loss": 0.7199,
      "step": 2781
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.849770510032652e-06,
      "loss": 0.7401,
      "step": 2782
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.845046497296382e-06,
      "loss": 0.7454,
      "step": 2783
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.84032237780517e-06,
      "loss": 0.772,
      "step": 2784
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.835598153020625e-06,
      "loss": 0.7216,
      "step": 2785
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.830873824404394e-06,
      "loss": 0.7175,
      "step": 2786
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.826149393418146e-06,
      "loss": 0.7126,
      "step": 2787
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.821424861523591e-06,
      "loss": 0.6781,
      "step": 2788
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.81670023018246e-06,
      "loss": 0.7036,
      "step": 2789
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.811975500856524e-06,
      "loss": 0.7313,
      "step": 2790
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.807250675007579e-06,
      "loss": 0.7486,
      "step": 2791
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.802525754097452e-06,
      "loss": 0.7383,
      "step": 2792
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.797800739588006e-06,
      "loss": 0.691,
      "step": 2793
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.793075632941117e-06,
      "loss": 0.7191,
      "step": 2794
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.788350435618706e-06,
      "loss": 0.7495,
      "step": 2795
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.783625149082717e-06,
      "loss": 0.7153,
      "step": 2796
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.778899774795115e-06,
      "loss": 0.7326,
      "step": 2797
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.7741743142179e-06,
      "loss": 0.7207,
      "step": 2798
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.7694487688131e-06,
      "loss": 0.7412,
      "step": 2799
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.764723140042758e-06,
      "loss": 0.7178,
      "step": 2800
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.759997429368952e-06,
      "loss": 0.7321,
      "step": 2801
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.755271638253786e-06,
      "loss": 0.6909,
      "step": 2802
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.750545768159386e-06,
      "loss": 0.7434,
      "step": 2803
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.7458198205479e-06,
      "loss": 0.6691,
      "step": 2804
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.741093796881505e-06,
      "loss": 0.7323,
      "step": 2805
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.736367698622397e-06,
      "loss": 0.7612,
      "step": 2806
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.7316415272328e-06,
      "loss": 0.7303,
      "step": 2807
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.726915284174955e-06,
      "loss": 0.731,
      "step": 2808
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.722188970911133e-06,
      "loss": 0.7344,
      "step": 2809
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.717462588903617e-06,
      "loss": 0.7353,
      "step": 2810
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.712736139614717e-06,
      "loss": 0.7089,
      "step": 2811
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.708009624506765e-06,
      "loss": 0.7525,
      "step": 2812
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.703283045042109e-06,
      "loss": 0.7445,
      "step": 2813
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.698556402683117e-06,
      "loss": 0.7294,
      "step": 2814
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.693829698892186e-06,
      "loss": 0.7161,
      "step": 2815
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.68910293513172e-06,
      "loss": 0.7211,
      "step": 2816
    },
    {
      "epoch": 1.98,
      "eval_loss": 0.7237575054168701,
      "eval_runtime": 30.918,
      "eval_samples_per_second": 264.926,
      "eval_steps_per_second": 16.56,
      "step": 2816
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.684376112864144e-06,
      "loss": 0.74,
      "step": 2817
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.679649233551906e-06,
      "loss": 0.6996,
      "step": 2818
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.674922298657472e-06,
      "loss": 0.7071,
      "step": 2819
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.670195309643316e-06,
      "loss": 0.7009,
      "step": 2820
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.66546826797194e-06,
      "loss": 0.7079,
      "step": 2821
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.660741175105853e-06,
      "loss": 0.6829,
      "step": 2822
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.656014032507584e-06,
      "loss": 0.7372,
      "step": 2823
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.651286841639683e-06,
      "loss": 0.7073,
      "step": 2824
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.646559603964704e-06,
      "loss": 0.7142,
      "step": 2825
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.641832320945226e-06,
      "loss": 0.7225,
      "step": 2826
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.63710499404383e-06,
      "loss": 0.7313,
      "step": 2827
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.632377624723122e-06,
      "loss": 0.7197,
      "step": 2828
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.627650214445712e-06,
      "loss": 0.7201,
      "step": 2829
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.622922764674235e-06,
      "loss": 0.7491,
      "step": 2830
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.618195276871324e-06,
      "loss": 0.7168,
      "step": 2831
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.613467752499633e-06,
      "loss": 0.699,
      "step": 2832
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.608740193021823e-06,
      "loss": 0.7009,
      "step": 2833
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.604012599900569e-06,
      "loss": 0.7421,
      "step": 2834
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.599284974598554e-06,
      "loss": 0.7211,
      "step": 2835
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.594557318578471e-06,
      "loss": 0.6988,
      "step": 2836
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.589829633303026e-06,
      "loss": 0.7366,
      "step": 2837
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.585101920234927e-06,
      "loss": 0.7071,
      "step": 2838
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.580374180836898e-06,
      "loss": 0.7618,
      "step": 2839
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.575646416571668e-06,
      "loss": 0.7092,
      "step": 2840
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.57091862890197e-06,
      "loss": 0.7189,
      "step": 2841
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.566190819290554e-06,
      "loss": 0.7462,
      "step": 2842
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.561462989200165e-06,
      "loss": 0.7006,
      "step": 2843
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.556735140093566e-06,
      "loss": 0.722,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.552007273433516e-06,
      "loss": 0.6665,
      "step": 2845
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.547279390682783e-06,
      "loss": 0.713,
      "step": 2846
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.542551493304143e-06,
      "loss": 0.7181,
      "step": 2847
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.537823582760375e-06,
      "loss": 0.6954,
      "step": 2848
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.533095660514255e-06,
      "loss": 0.7009,
      "step": 2849
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.528367728028573e-06,
      "loss": 0.6977,
      "step": 2850
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.523639786766117e-06,
      "loss": 0.6856,
      "step": 2851
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.518911838189677e-06,
      "loss": 0.7207,
      "step": 2852
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.51418388376205e-06,
      "loss": 0.6823,
      "step": 2853
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.509455924946032e-06,
      "loss": 0.706,
      "step": 2854
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.504727963204416e-06,
      "loss": 0.687,
      "step": 2855
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.5e-06,
      "loss": 0.6898,
      "step": 2856
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.495272036795585e-06,
      "loss": 0.6945,
      "step": 2857
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.49054407505397e-06,
      "loss": 0.6941,
      "step": 2858
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.48581611623795e-06,
      "loss": 0.7026,
      "step": 2859
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.481088161810323e-06,
      "loss": 0.6954,
      "step": 2860
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.476360213233885e-06,
      "loss": 0.6798,
      "step": 2861
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.47163227197143e-06,
      "loss": 0.6744,
      "step": 2862
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.466904339485748e-06,
      "loss": 0.6856,
      "step": 2863
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.462176417239629e-06,
      "loss": 0.7064,
      "step": 2864
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.457448506695859e-06,
      "loss": 0.7061,
      "step": 2865
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.452720609317217e-06,
      "loss": 0.708,
      "step": 2866
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.447992726566486e-06,
      "loss": 0.6921,
      "step": 2867
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.443264859906434e-06,
      "loss": 0.7054,
      "step": 2868
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.438537010799835e-06,
      "loss": 0.715,
      "step": 2869
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.433809180709448e-06,
      "loss": 0.7012,
      "step": 2870
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.429081371098031e-06,
      "loss": 0.6823,
      "step": 2871
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.424353583428334e-06,
      "loss": 0.6893,
      "step": 2872
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.419625819163105e-06,
      "loss": 0.666,
      "step": 2873
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.414898079765075e-06,
      "loss": 0.698,
      "step": 2874
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.410170366696977e-06,
      "loss": 0.7028,
      "step": 2875
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.40544268142153e-06,
      "loss": 0.6758,
      "step": 2876
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.400715025401448e-06,
      "loss": 0.7161,
      "step": 2877
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.395987400099433e-06,
      "loss": 0.7106,
      "step": 2878
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.391259806978177e-06,
      "loss": 0.6844,
      "step": 2879
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.386532247500368e-06,
      "loss": 0.6773,
      "step": 2880
    },
    {
      "epoch": 2.03,
      "eval_loss": 0.7268651127815247,
      "eval_runtime": 30.9312,
      "eval_samples_per_second": 264.814,
      "eval_steps_per_second": 16.553,
      "step": 2880
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.381804723128676e-06,
      "loss": 0.6929,
      "step": 2881
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.377077235325765e-06,
      "loss": 0.6925,
      "step": 2882
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.372349785554286e-06,
      "loss": 0.7135,
      "step": 2883
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.367622375276883e-06,
      "loss": 0.6805,
      "step": 2884
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.362895005956174e-06,
      "loss": 0.7226,
      "step": 2885
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.358167679054778e-06,
      "loss": 0.7048,
      "step": 2886
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.353440396035297e-06,
      "loss": 0.7489,
      "step": 2887
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.348713158360318e-06,
      "loss": 0.6797,
      "step": 2888
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.343985967492415e-06,
      "loss": 0.6854,
      "step": 2889
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.339258824894147e-06,
      "loss": 0.7069,
      "step": 2890
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.334531732028061e-06,
      "loss": 0.6564,
      "step": 2891
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.329804690356684e-06,
      "loss": 0.7143,
      "step": 2892
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.325077701342528e-06,
      "loss": 0.6948,
      "step": 2893
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.320350766448092e-06,
      "loss": 0.6659,
      "step": 2894
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.315623887135858e-06,
      "loss": 0.7019,
      "step": 2895
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.310897064868284e-06,
      "loss": 0.6843,
      "step": 2896
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.306170301107815e-06,
      "loss": 0.695,
      "step": 2897
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.301443597316883e-06,
      "loss": 0.6848,
      "step": 2898
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.296716954957893e-06,
      "loss": 0.7163,
      "step": 2899
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.291990375493236e-06,
      "loss": 0.6601,
      "step": 2900
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.287263860385283e-06,
      "loss": 0.7242,
      "step": 2901
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.282537411096384e-06,
      "loss": 0.684,
      "step": 2902
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.277811029088867e-06,
      "loss": 0.7249,
      "step": 2903
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.273084715825043e-06,
      "loss": 0.6961,
      "step": 2904
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.2683584727672e-06,
      "loss": 0.695,
      "step": 2905
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.263632301377607e-06,
      "loss": 0.682,
      "step": 2906
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.258906203118497e-06,
      "loss": 0.6969,
      "step": 2907
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.254180179452102e-06,
      "loss": 0.6907,
      "step": 2908
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.249454231840616e-06,
      "loss": 0.7238,
      "step": 2909
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.244728361746214e-06,
      "loss": 0.6962,
      "step": 2910
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.240002570631048e-06,
      "loss": 0.6824,
      "step": 2911
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.235276859957244e-06,
      "loss": 0.7044,
      "step": 2912
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.230551231186902e-06,
      "loss": 0.676,
      "step": 2913
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.225825685782098e-06,
      "loss": 0.6739,
      "step": 2914
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.221100225204885e-06,
      "loss": 0.6898,
      "step": 2915
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.216374850917283e-06,
      "loss": 0.7064,
      "step": 2916
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.211649564381294e-06,
      "loss": 0.7093,
      "step": 2917
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.206924367058885e-06,
      "loss": 0.6969,
      "step": 2918
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.202199260411997e-06,
      "loss": 0.7022,
      "step": 2919
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.19747424590255e-06,
      "loss": 0.6852,
      "step": 2920
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.192749324992423e-06,
      "loss": 0.718,
      "step": 2921
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.188024499143477e-06,
      "loss": 0.6843,
      "step": 2922
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.18329976981754e-06,
      "loss": 0.7015,
      "step": 2923
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.178575138476409e-06,
      "loss": 0.7006,
      "step": 2924
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.173850606581854e-06,
      "loss": 0.6816,
      "step": 2925
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.169126175595606e-06,
      "loss": 0.7131,
      "step": 2926
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.164401846979374e-06,
      "loss": 0.6981,
      "step": 2927
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.159677622194832e-06,
      "loss": 0.6619,
      "step": 2928
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.154953502703623e-06,
      "loss": 0.7278,
      "step": 2929
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.15022948996735e-06,
      "loss": 0.673,
      "step": 2930
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.145505585447595e-06,
      "loss": 0.7194,
      "step": 2931
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.140781790605899e-06,
      "loss": 0.6963,
      "step": 2932
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.13605810690377e-06,
      "loss": 0.7196,
      "step": 2933
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.131334535802682e-06,
      "loss": 0.6849,
      "step": 2934
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.126611078764075e-06,
      "loss": 0.6895,
      "step": 2935
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.121887737249355e-06,
      "loss": 0.6796,
      "step": 2936
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.117164512719887e-06,
      "loss": 0.7027,
      "step": 2937
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.112441406637005e-06,
      "loss": 0.6988,
      "step": 2938
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.107718420462006e-06,
      "loss": 0.672,
      "step": 2939
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.102995555656144e-06,
      "loss": 0.6844,
      "step": 2940
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.098272813680644e-06,
      "loss": 0.6716,
      "step": 2941
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.093550195996688e-06,
      "loss": 0.7005,
      "step": 2942
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.088827704065417e-06,
      "loss": 0.6818,
      "step": 2943
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.084105339347941e-06,
      "loss": 0.6696,
      "step": 2944
    },
    {
      "epoch": 2.07,
      "eval_loss": 0.7266552448272705,
      "eval_runtime": 30.9289,
      "eval_samples_per_second": 264.834,
      "eval_steps_per_second": 16.554,
      "step": 2944
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.079383103305325e-06,
      "loss": 0.6877,
      "step": 2945
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.074660997398595e-06,
      "loss": 0.7012,
      "step": 2946
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.069939023088736e-06,
      "loss": 0.6785,
      "step": 2947
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.065217181836695e-06,
      "loss": 0.6919,
      "step": 2948
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.060495475103375e-06,
      "loss": 0.68,
      "step": 2949
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.055773904349644e-06,
      "loss": 0.7063,
      "step": 2950
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.051052471036313e-06,
      "loss": 0.6842,
      "step": 2951
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.046331176624165e-06,
      "loss": 0.6965,
      "step": 2952
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.041610022573935e-06,
      "loss": 0.7149,
      "step": 2953
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.036889010346313e-06,
      "loss": 0.711,
      "step": 2954
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.032168141401949e-06,
      "loss": 0.6911,
      "step": 2955
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.027447417201447e-06,
      "loss": 0.6867,
      "step": 2956
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.02272683920536e-06,
      "loss": 0.6982,
      "step": 2957
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.018006408874208e-06,
      "loss": 0.7282,
      "step": 2958
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.013286127668457e-06,
      "loss": 0.6865,
      "step": 2959
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.008565997048526e-06,
      "loss": 0.666,
      "step": 2960
    },
    {
      "epoch": 2.09,
      "learning_rate": 8.003846018474795e-06,
      "loss": 0.7157,
      "step": 2961
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.999126193407587e-06,
      "loss": 0.6929,
      "step": 2962
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.994406523307181e-06,
      "loss": 0.6901,
      "step": 2963
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.989687009633814e-06,
      "loss": 0.6988,
      "step": 2964
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.984967653847668e-06,
      "loss": 0.6828,
      "step": 2965
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.980248457408877e-06,
      "loss": 0.6815,
      "step": 2966
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.975529421777531e-06,
      "loss": 0.6987,
      "step": 2967
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.970810548413663e-06,
      "loss": 0.7002,
      "step": 2968
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.96609183877726e-06,
      "loss": 0.6943,
      "step": 2969
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.961373294328258e-06,
      "loss": 0.6624,
      "step": 2970
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.956654916526538e-06,
      "loss": 0.7427,
      "step": 2971
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.95193670683194e-06,
      "loss": 0.6844,
      "step": 2972
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.947218666704236e-06,
      "loss": 0.6889,
      "step": 2973
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.942500797603156e-06,
      "loss": 0.7216,
      "step": 2974
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.937783100988376e-06,
      "loss": 0.6802,
      "step": 2975
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.93306557831952e-06,
      "loss": 0.7035,
      "step": 2976
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.928348231056156e-06,
      "loss": 0.7018,
      "step": 2977
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.923631060657796e-06,
      "loss": 0.6711,
      "step": 2978
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.9189140685839e-06,
      "loss": 0.6988,
      "step": 2979
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.914197256293872e-06,
      "loss": 0.7062,
      "step": 2980
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.90948062524706e-06,
      "loss": 0.7506,
      "step": 2981
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.904764176902756e-06,
      "loss": 0.7016,
      "step": 2982
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.900047912720196e-06,
      "loss": 0.672,
      "step": 2983
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.895331834158558e-06,
      "loss": 0.6431,
      "step": 2984
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.890615942676963e-06,
      "loss": 0.698,
      "step": 2985
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.885900239734477e-06,
      "loss": 0.7229,
      "step": 2986
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.881184726790101e-06,
      "loss": 0.7263,
      "step": 2987
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.876469405302785e-06,
      "loss": 0.6993,
      "step": 2988
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.871754276731414e-06,
      "loss": 0.6849,
      "step": 2989
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.867039342534816e-06,
      "loss": 0.7149,
      "step": 2990
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.862324604171759e-06,
      "loss": 0.6952,
      "step": 2991
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.857610063100947e-06,
      "loss": 0.7199,
      "step": 2992
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.852895720781028e-06,
      "loss": 0.7084,
      "step": 2993
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.848181578670587e-06,
      "loss": 0.6872,
      "step": 2994
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.843467638228143e-06,
      "loss": 0.7111,
      "step": 2995
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.838753900912157e-06,
      "loss": 0.7087,
      "step": 2996
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.834040368181028e-06,
      "loss": 0.6714,
      "step": 2997
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.829327041493087e-06,
      "loss": 0.6857,
      "step": 2998
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.824613922306604e-06,
      "loss": 0.711,
      "step": 2999
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.819901012079785e-06,
      "loss": 0.6957,
      "step": 3000
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.815188312270773e-06,
      "loss": 0.6998,
      "step": 3001
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.81047582433764e-06,
      "loss": 0.6529,
      "step": 3002
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.805763549738399e-06,
      "loss": 0.6938,
      "step": 3003
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.801051489930995e-06,
      "loss": 0.686,
      "step": 3004
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.796339646373307e-06,
      "loss": 0.6924,
      "step": 3005
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.791628020523142e-06,
      "loss": 0.7003,
      "step": 3006
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.786916613838245e-06,
      "loss": 0.7042,
      "step": 3007
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.782205427776292e-06,
      "loss": 0.6876,
      "step": 3008
    },
    {
      "epoch": 2.12,
      "eval_loss": 0.7268000841140747,
      "eval_runtime": 30.9153,
      "eval_samples_per_second": 264.95,
      "eval_steps_per_second": 16.561,
      "step": 3008
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.777494463794893e-06,
      "loss": 0.6751,
      "step": 3009
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.772783723351583e-06,
      "loss": 0.7019,
      "step": 3010
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.768073207903834e-06,
      "loss": 0.6803,
      "step": 3011
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.763362918909045e-06,
      "loss": 0.7207,
      "step": 3012
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.758652857824546e-06,
      "loss": 0.7179,
      "step": 3013
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.753943026107598e-06,
      "loss": 0.7195,
      "step": 3014
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.749233425215388e-06,
      "loss": 0.6877,
      "step": 3015
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.744524056605037e-06,
      "loss": 0.6796,
      "step": 3016
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.739814921733584e-06,
      "loss": 0.7319,
      "step": 3017
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.735106022058006e-06,
      "loss": 0.7008,
      "step": 3018
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.730397359035201e-06,
      "loss": 0.6948,
      "step": 3019
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.725688934121998e-06,
      "loss": 0.7102,
      "step": 3020
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.720980748775148e-06,
      "loss": 0.6645,
      "step": 3021
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.716272804451333e-06,
      "loss": 0.6955,
      "step": 3022
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.711565102607156e-06,
      "loss": 0.6815,
      "step": 3023
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.706857644699144e-06,
      "loss": 0.7023,
      "step": 3024
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.702150432183756e-06,
      "loss": 0.7002,
      "step": 3025
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.697443466517368e-06,
      "loss": 0.711,
      "step": 3026
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.692736749156286e-06,
      "loss": 0.7319,
      "step": 3027
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.688030281556727e-06,
      "loss": 0.7144,
      "step": 3028
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.683324065174842e-06,
      "loss": 0.6555,
      "step": 3029
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.678618101466702e-06,
      "loss": 0.6961,
      "step": 3030
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.673912391888298e-06,
      "loss": 0.6581,
      "step": 3031
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.669206937895544e-06,
      "loss": 0.7036,
      "step": 3032
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.664501740944271e-06,
      "loss": 0.7227,
      "step": 3033
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.659796802490239e-06,
      "loss": 0.6756,
      "step": 3034
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.655092123989119e-06,
      "loss": 0.6856,
      "step": 3035
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.650387706896506e-06,
      "loss": 0.7002,
      "step": 3036
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.645683552667913e-06,
      "loss": 0.6831,
      "step": 3037
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.640979662758772e-06,
      "loss": 0.6603,
      "step": 3038
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.636276038624438e-06,
      "loss": 0.7158,
      "step": 3039
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.631572681720171e-06,
      "loss": 0.6959,
      "step": 3040
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.626869593501159e-06,
      "loss": 0.6868,
      "step": 3041
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.622166775422503e-06,
      "loss": 0.685,
      "step": 3042
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.617464228939225e-06,
      "loss": 0.7022,
      "step": 3043
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.612761955506258e-06,
      "loss": 0.7294,
      "step": 3044
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.608059956578452e-06,
      "loss": 0.7042,
      "step": 3045
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.603358233610572e-06,
      "loss": 0.7025,
      "step": 3046
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.5986567880572985e-06,
      "loss": 0.7122,
      "step": 3047
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.593955621373225e-06,
      "loss": 0.7116,
      "step": 3048
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.589254735012858e-06,
      "loss": 0.6959,
      "step": 3049
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.584554130430621e-06,
      "loss": 0.6701,
      "step": 3050
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.579853809080846e-06,
      "loss": 0.7336,
      "step": 3051
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.5751537724177765e-06,
      "loss": 0.6533,
      "step": 3052
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.570454021895575e-06,
      "loss": 0.696,
      "step": 3053
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.56575455896831e-06,
      "loss": 0.721,
      "step": 3054
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.56105538508996e-06,
      "loss": 0.6941,
      "step": 3055
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.556356501714417e-06,
      "loss": 0.7169,
      "step": 3056
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.551657910295483e-06,
      "loss": 0.7247,
      "step": 3057
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.5469596122868676e-06,
      "loss": 0.6939,
      "step": 3058
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.542261609142193e-06,
      "loss": 0.6918,
      "step": 3059
    },
    {
      "epoch": 2.15,
      "learning_rate": 7.5375639023149855e-06,
      "loss": 0.6818,
      "step": 3060
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.532866493258686e-06,
      "loss": 0.6839,
      "step": 3061
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.528169383426636e-06,
      "loss": 0.6753,
      "step": 3062
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.523472574272088e-06,
      "loss": 0.7084,
      "step": 3063
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.518776067248203e-06,
      "loss": 0.7164,
      "step": 3064
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.514079863808047e-06,
      "loss": 0.6808,
      "step": 3065
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.509383965404591e-06,
      "loss": 0.7021,
      "step": 3066
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.504688373490713e-06,
      "loss": 0.6883,
      "step": 3067
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.4999930895191965e-06,
      "loss": 0.7019,
      "step": 3068
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.495298114942728e-06,
      "loss": 0.7056,
      "step": 3069
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.490603451213901e-06,
      "loss": 0.6935,
      "step": 3070
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.485909099785209e-06,
      "loss": 0.6807,
      "step": 3071
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.481215062109055e-06,
      "loss": 0.6776,
      "step": 3072
    },
    {
      "epoch": 2.16,
      "eval_loss": 0.7261952757835388,
      "eval_runtime": 30.9393,
      "eval_samples_per_second": 264.744,
      "eval_steps_per_second": 16.549,
      "step": 3072
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.476521339637736e-06,
      "loss": 0.6879,
      "step": 3073
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.471827933823458e-06,
      "loss": 0.7099,
      "step": 3074
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.46713484611833e-06,
      "loss": 0.6932,
      "step": 3075
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.462442077974357e-06,
      "loss": 0.7127,
      "step": 3076
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.45774963084345e-06,
      "loss": 0.6811,
      "step": 3077
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.453057506177417e-06,
      "loss": 0.6579,
      "step": 3078
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.44836570542797e-06,
      "loss": 0.665,
      "step": 3079
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.443674230046719e-06,
      "loss": 0.7101,
      "step": 3080
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.438983081485172e-06,
      "loss": 0.7103,
      "step": 3081
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.434292261194739e-06,
      "loss": 0.6532,
      "step": 3082
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.429601770626724e-06,
      "loss": 0.6596,
      "step": 3083
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.424911611232333e-06,
      "loss": 0.6849,
      "step": 3084
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.420221784462667e-06,
      "loss": 0.67,
      "step": 3085
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.415532291768727e-06,
      "loss": 0.7153,
      "step": 3086
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.410843134601407e-06,
      "loss": 0.6555,
      "step": 3087
    },
    {
      "epoch": 2.17,
      "learning_rate": 7.406154314411499e-06,
      "loss": 0.7089,
      "step": 3088
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.401465832649692e-06,
      "loss": 0.6999,
      "step": 3089
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.3967776907665684e-06,
      "loss": 0.6904,
      "step": 3090
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.392089890212605e-06,
      "loss": 0.6877,
      "step": 3091
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.387402432438176e-06,
      "loss": 0.7105,
      "step": 3092
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.382715318893547e-06,
      "loss": 0.7192,
      "step": 3093
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.378028551028882e-06,
      "loss": 0.7138,
      "step": 3094
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.373342130294224e-06,
      "loss": 0.704,
      "step": 3095
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.368656058139524e-06,
      "loss": 0.6667,
      "step": 3096
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.36397033601462e-06,
      "loss": 0.6844,
      "step": 3097
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.35928496536924e-06,
      "loss": 0.6801,
      "step": 3098
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.354599947653004e-06,
      "loss": 0.7211,
      "step": 3099
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.349915284315423e-06,
      "loss": 0.721,
      "step": 3100
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.345230976805901e-06,
      "loss": 0.6896,
      "step": 3101
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.340547026573728e-06,
      "loss": 0.7109,
      "step": 3102
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.335863435068085e-06,
      "loss": 0.7124,
      "step": 3103
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.331180203738042e-06,
      "loss": 0.679,
      "step": 3104
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.326497334032563e-06,
      "loss": 0.6754,
      "step": 3105
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.321814827400484e-06,
      "loss": 0.6735,
      "step": 3106
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.317132685290546e-06,
      "loss": 0.6964,
      "step": 3107
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.3124509091513694e-06,
      "loss": 0.6847,
      "step": 3108
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.3077695004314596e-06,
      "loss": 0.7075,
      "step": 3109
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.303088460579217e-06,
      "loss": 0.6876,
      "step": 3110
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.298407791042918e-06,
      "loss": 0.7257,
      "step": 3111
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.293727493270732e-06,
      "loss": 0.7361,
      "step": 3112
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.289047568710706e-06,
      "loss": 0.7258,
      "step": 3113
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.284368018810778e-06,
      "loss": 0.6826,
      "step": 3114
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.279688845018767e-06,
      "loss": 0.7031,
      "step": 3115
    },
    {
      "epoch": 2.19,
      "learning_rate": 7.27501004878238e-06,
      "loss": 0.6919,
      "step": 3116
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.270331631549192e-06,
      "loss": 0.6922,
      "step": 3117
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.265653594766684e-06,
      "loss": 0.7083,
      "step": 3118
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.260975939882201e-06,
      "loss": 0.6813,
      "step": 3119
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.25629866834298e-06,
      "loss": 0.6854,
      "step": 3120
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.251621781596133e-06,
      "loss": 0.7343,
      "step": 3121
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.246945281088657e-06,
      "loss": 0.6925,
      "step": 3122
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.242269168267427e-06,
      "loss": 0.735,
      "step": 3123
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.237593444579201e-06,
      "loss": 0.7252,
      "step": 3124
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.232918111470612e-06,
      "loss": 0.675,
      "step": 3125
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.228243170388177e-06,
      "loss": 0.7419,
      "step": 3126
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.223568622778291e-06,
      "loss": 0.6854,
      "step": 3127
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.2188944700872225e-06,
      "loss": 0.7208,
      "step": 3128
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.214220713761122e-06,
      "loss": 0.7052,
      "step": 3129
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.209547355246018e-06,
      "loss": 0.6909,
      "step": 3130
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.204874395987813e-06,
      "loss": 0.6587,
      "step": 3131
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.200201837432289e-06,
      "loss": 0.6967,
      "step": 3132
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.195529681025102e-06,
      "loss": 0.6891,
      "step": 3133
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.190857928211783e-06,
      "loss": 0.7071,
      "step": 3134
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.186186580437741e-06,
      "loss": 0.7197,
      "step": 3135
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.181515639148255e-06,
      "loss": 0.683,
      "step": 3136
    },
    {
      "epoch": 2.21,
      "eval_loss": 0.7254009246826172,
      "eval_runtime": 30.9616,
      "eval_samples_per_second": 264.554,
      "eval_steps_per_second": 16.537,
      "step": 3136
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.176845105788484e-06,
      "loss": 0.657,
      "step": 3137
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.172174981803459e-06,
      "loss": 0.6987,
      "step": 3138
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.167505268638079e-06,
      "loss": 0.6887,
      "step": 3139
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.162835967737123e-06,
      "loss": 0.6796,
      "step": 3140
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.158167080545237e-06,
      "loss": 0.7088,
      "step": 3141
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.153498608506946e-06,
      "loss": 0.6865,
      "step": 3142
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.148830553066637e-06,
      "loss": 0.6937,
      "step": 3143
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.144162915668578e-06,
      "loss": 0.725,
      "step": 3144
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.139495697756901e-06,
      "loss": 0.6877,
      "step": 3145
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.134828900775608e-06,
      "loss": 0.7292,
      "step": 3146
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.130162526168576e-06,
      "loss": 0.7012,
      "step": 3147
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.125496575379548e-06,
      "loss": 0.6885,
      "step": 3148
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.120831049852135e-06,
      "loss": 0.7132,
      "step": 3149
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.116165951029818e-06,
      "loss": 0.7003,
      "step": 3150
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.1115012803559436e-06,
      "loss": 0.6778,
      "step": 3151
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.10683703927373e-06,
      "loss": 0.6953,
      "step": 3152
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.102173229226259e-06,
      "loss": 0.6766,
      "step": 3153
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.097509851656483e-06,
      "loss": 0.6823,
      "step": 3154
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.0928469080072154e-06,
      "loss": 0.6954,
      "step": 3155
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.088184399721138e-06,
      "loss": 0.7099,
      "step": 3156
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.0835223282407985e-06,
      "loss": 0.706,
      "step": 3157
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.078860695008609e-06,
      "loss": 0.6861,
      "step": 3158
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.074199501466844e-06,
      "loss": 0.7119,
      "step": 3159
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.069538749057649e-06,
      "loss": 0.6819,
      "step": 3160
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.064878439223021e-06,
      "loss": 0.7125,
      "step": 3161
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.060218573404831e-06,
      "loss": 0.712,
      "step": 3162
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.0555591530448066e-06,
      "loss": 0.7094,
      "step": 3163
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.0509001795845404e-06,
      "loss": 0.7191,
      "step": 3164
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.046241654465484e-06,
      "loss": 0.7141,
      "step": 3165
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.0415835791289535e-06,
      "loss": 0.7317,
      "step": 3166
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.036925955016124e-06,
      "loss": 0.7081,
      "step": 3167
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.0322687835680325e-06,
      "loss": 0.6957,
      "step": 3168
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.027612066225575e-06,
      "loss": 0.6965,
      "step": 3169
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.022955804429505e-06,
      "loss": 0.7079,
      "step": 3170
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.018299999620441e-06,
      "loss": 0.6905,
      "step": 3171
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.01364465323885e-06,
      "loss": 0.688,
      "step": 3172
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.008989766725066e-06,
      "loss": 0.7192,
      "step": 3173
    },
    {
      "epoch": 2.24,
      "learning_rate": 7.004335341519277e-06,
      "loss": 0.6995,
      "step": 3174
    },
    {
      "epoch": 2.24,
      "learning_rate": 6.9996813790615306e-06,
      "loss": 0.684,
      "step": 3175
    },
    {
      "epoch": 2.24,
      "learning_rate": 6.995027880791727e-06,
      "loss": 0.6986,
      "step": 3176
    },
    {
      "epoch": 2.24,
      "learning_rate": 6.9903748481496295e-06,
      "loss": 0.7132,
      "step": 3177
    },
    {
      "epoch": 2.24,
      "learning_rate": 6.9857222825748525e-06,
      "loss": 0.7093,
      "step": 3178
    },
    {
      "epoch": 2.24,
      "learning_rate": 6.981070185506864e-06,
      "loss": 0.6922,
      "step": 3179
    },
    {
      "epoch": 2.24,
      "learning_rate": 6.976418558384995e-06,
      "loss": 0.7203,
      "step": 3180
    },
    {
      "epoch": 2.24,
      "learning_rate": 6.97176740264842e-06,
      "loss": 0.6696,
      "step": 3181
    },
    {
      "epoch": 2.24,
      "learning_rate": 6.967116719736181e-06,
      "loss": 0.7004,
      "step": 3182
    },
    {
      "epoch": 2.24,
      "learning_rate": 6.962466511087158e-06,
      "loss": 0.689,
      "step": 3183
    },
    {
      "epoch": 2.24,
      "learning_rate": 6.957816778140096e-06,
      "loss": 0.6728,
      "step": 3184
    },
    {
      "epoch": 2.24,
      "learning_rate": 6.953167522333589e-06,
      "loss": 0.6711,
      "step": 3185
    },
    {
      "epoch": 2.24,
      "learning_rate": 6.948518745106085e-06,
      "loss": 0.6774,
      "step": 3186
    },
    {
      "epoch": 2.24,
      "learning_rate": 6.943870447895884e-06,
      "loss": 0.6524,
      "step": 3187
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.939222632141131e-06,
      "loss": 0.6924,
      "step": 3188
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.934575299279834e-06,
      "loss": 0.6481,
      "step": 3189
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.929928450749839e-06,
      "loss": 0.7137,
      "step": 3190
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.925282087988853e-06,
      "loss": 0.7203,
      "step": 3191
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.920636212434424e-06,
      "loss": 0.6602,
      "step": 3192
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.9159908255239555e-06,
      "loss": 0.6874,
      "step": 3193
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.911345928694695e-06,
      "loss": 0.699,
      "step": 3194
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.906701523383741e-06,
      "loss": 0.7307,
      "step": 3195
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.902057611028039e-06,
      "loss": 0.7018,
      "step": 3196
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.897414193064386e-06,
      "loss": 0.7018,
      "step": 3197
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.892771270929419e-06,
      "loss": 0.7152,
      "step": 3198
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.888128846059626e-06,
      "loss": 0.678,
      "step": 3199
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.883486919891342e-06,
      "loss": 0.705,
      "step": 3200
    },
    {
      "epoch": 2.25,
      "eval_loss": 0.7249176502227783,
      "eval_runtime": 30.9232,
      "eval_samples_per_second": 264.882,
      "eval_steps_per_second": 16.557,
      "step": 3200
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.8788454938607446e-06,
      "loss": 0.7308,
      "step": 3201
    },
    {
      "epoch": 2.25,
      "learning_rate": 6.874204569403858e-06,
      "loss": 0.7037,
      "step": 3202
    },
    {
      "epoch": 2.26,
      "learning_rate": 6.8695641479565525e-06,
      "loss": 0.7055,
      "step": 3203
    },
    {
      "epoch": 2.26,
      "learning_rate": 6.864924230954541e-06,
      "loss": 0.6671,
      "step": 3204
    },
    {
      "epoch": 2.26,
      "learning_rate": 6.860284819833381e-06,
      "loss": 0.679,
      "step": 3205
    },
    {
      "epoch": 2.26,
      "learning_rate": 6.855645916028472e-06,
      "loss": 0.687,
      "step": 3206
    },
    {
      "epoch": 2.26,
      "learning_rate": 6.851007520975059e-06,
      "loss": 0.6748,
      "step": 3207
    },
    {
      "epoch": 2.26,
      "learning_rate": 6.8463696361082285e-06,
      "loss": 0.6843,
      "step": 3208
    },
    {
      "epoch": 2.26,
      "learning_rate": 6.84173226286291e-06,
      "loss": 0.6877,
      "step": 3209
    },
    {
      "epoch": 2.26,
      "learning_rate": 6.837095402673873e-06,
      "loss": 0.6604,
      "step": 3210
    },
    {
      "epoch": 2.26,
      "learning_rate": 6.832459056975729e-06,
      "loss": 0.6715,
      "step": 3211
    },
    {
      "epoch": 2.26,
      "learning_rate": 6.827823227202932e-06,
      "loss": 0.7133,
      "step": 3212
    },
    {
      "epoch": 2.26,
      "learning_rate": 6.823187914789774e-06,
      "loss": 0.7116,
      "step": 3213
    },
    {
      "epoch": 2.26,
      "learning_rate": 6.8185531211703865e-06,
      "loss": 0.6949,
      "step": 3214
    },
    {
      "epoch": 2.26,
      "learning_rate": 6.813918847778745e-06,
      "loss": 0.6685,
      "step": 3215
    },
    {
      "epoch": 2.26,
      "learning_rate": 6.8092850960486564e-06,
      "loss": 0.6895,
      "step": 3216
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.804651867413773e-06,
      "loss": 0.6835,
      "step": 3217
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.800019163307583e-06,
      "loss": 0.7526,
      "step": 3218
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.795386985163413e-06,
      "loss": 0.7034,
      "step": 3219
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.790755334414424e-06,
      "loss": 0.6959,
      "step": 3220
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.786124212493617e-06,
      "loss": 0.7103,
      "step": 3221
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.78149362083383e-06,
      "loss": 0.6874,
      "step": 3222
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.776863560867731e-06,
      "loss": 0.6982,
      "step": 3223
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.7722340340278315e-06,
      "loss": 0.6999,
      "step": 3224
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.767605041746473e-06,
      "loss": 0.693,
      "step": 3225
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.762976585455835e-06,
      "loss": 0.6863,
      "step": 3226
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.758348666587925e-06,
      "loss": 0.6966,
      "step": 3227
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.7537212865745905e-06,
      "loss": 0.6762,
      "step": 3228
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.74909444684751e-06,
      "loss": 0.7095,
      "step": 3229
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.744468148838195e-06,
      "loss": 0.6993,
      "step": 3230
    },
    {
      "epoch": 2.28,
      "learning_rate": 6.739842393977989e-06,
      "loss": 0.6993,
      "step": 3231
    },
    {
      "epoch": 2.28,
      "learning_rate": 6.735217183698067e-06,
      "loss": 0.6757,
      "step": 3232
    },
    {
      "epoch": 2.28,
      "learning_rate": 6.730592519429437e-06,
      "loss": 0.6975,
      "step": 3233
    },
    {
      "epoch": 2.28,
      "learning_rate": 6.725968402602936e-06,
      "loss": 0.6985,
      "step": 3234
    },
    {
      "epoch": 2.28,
      "learning_rate": 6.721344834649237e-06,
      "loss": 0.6833,
      "step": 3235
    },
    {
      "epoch": 2.28,
      "learning_rate": 6.716721816998836e-06,
      "loss": 0.6991,
      "step": 3236
    },
    {
      "epoch": 2.28,
      "learning_rate": 6.712099351082066e-06,
      "loss": 0.7018,
      "step": 3237
    },
    {
      "epoch": 2.28,
      "learning_rate": 6.707477438329078e-06,
      "loss": 0.6729,
      "step": 3238
    },
    {
      "epoch": 2.28,
      "learning_rate": 6.702856080169863e-06,
      "loss": 0.7119,
      "step": 3239
    },
    {
      "epoch": 2.28,
      "learning_rate": 6.698235278034236e-06,
      "loss": 0.7262,
      "step": 3240
    },
    {
      "epoch": 2.28,
      "learning_rate": 6.69361503335184e-06,
      "loss": 0.6554,
      "step": 3241
    },
    {
      "epoch": 2.28,
      "learning_rate": 6.688995347552147e-06,
      "loss": 0.7369,
      "step": 3242
    },
    {
      "epoch": 2.28,
      "learning_rate": 6.6843762220644535e-06,
      "loss": 0.6737,
      "step": 3243
    },
    {
      "epoch": 2.28,
      "learning_rate": 6.679757658317886e-06,
      "loss": 0.7247,
      "step": 3244
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.675139657741395e-06,
      "loss": 0.7067,
      "step": 3245
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.670522221763756e-06,
      "loss": 0.6544,
      "step": 3246
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.665905351813573e-06,
      "loss": 0.7081,
      "step": 3247
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.661289049319274e-06,
      "loss": 0.7024,
      "step": 3248
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.656673315709105e-06,
      "loss": 0.6979,
      "step": 3249
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.652058152411144e-06,
      "loss": 0.7104,
      "step": 3250
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.64744356085329e-06,
      "loss": 0.7007,
      "step": 3251
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.642829542463263e-06,
      "loss": 0.7114,
      "step": 3252
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.63821609866861e-06,
      "loss": 0.7121,
      "step": 3253
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.633603230896697e-06,
      "loss": 0.6941,
      "step": 3254
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.628990940574714e-06,
      "loss": 0.6737,
      "step": 3255
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.624379229129669e-06,
      "loss": 0.7216,
      "step": 3256
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.619768097988394e-06,
      "loss": 0.712,
      "step": 3257
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.615157548577539e-06,
      "loss": 0.6793,
      "step": 3258
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.610547582323581e-06,
      "loss": 0.6897,
      "step": 3259
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.605938200652799e-06,
      "loss": 0.6879,
      "step": 3260
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.6013294049913125e-06,
      "loss": 0.708,
      "step": 3261
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.596721196765048e-06,
      "loss": 0.6901,
      "step": 3262
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.592113577399752e-06,
      "loss": 0.6937,
      "step": 3263
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.587506548320989e-06,
      "loss": 0.6834,
      "step": 3264
    },
    {
      "epoch": 2.3,
      "eval_loss": 0.7249485850334167,
      "eval_runtime": 30.9092,
      "eval_samples_per_second": 265.002,
      "eval_steps_per_second": 16.565,
      "step": 3264
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.582900110954142e-06,
      "loss": 0.7038,
      "step": 3265
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.578294266724409e-06,
      "loss": 0.685,
      "step": 3266
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.573689017056806e-06,
      "loss": 0.6821,
      "step": 3267
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.569084363376166e-06,
      "loss": 0.6968,
      "step": 3268
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.564480307107135e-06,
      "loss": 0.6974,
      "step": 3269
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.559876849674175e-06,
      "loss": 0.7224,
      "step": 3270
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.555273992501563e-06,
      "loss": 0.6898,
      "step": 3271
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.550671737013391e-06,
      "loss": 0.7505,
      "step": 3272
    },
    {
      "epoch": 2.3,
      "learning_rate": 6.546070084633564e-06,
      "loss": 0.7107,
      "step": 3273
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.541469036785801e-06,
      "loss": 0.6652,
      "step": 3274
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.536868594893631e-06,
      "loss": 0.7412,
      "step": 3275
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.532268760380402e-06,
      "loss": 0.7215,
      "step": 3276
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.527669534669269e-06,
      "loss": 0.6894,
      "step": 3277
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.523070919183198e-06,
      "loss": 0.7138,
      "step": 3278
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.518472915344971e-06,
      "loss": 0.7056,
      "step": 3279
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.513875524577175e-06,
      "loss": 0.6841,
      "step": 3280
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.509278748302214e-06,
      "loss": 0.688,
      "step": 3281
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.504682587942293e-06,
      "loss": 0.7139,
      "step": 3282
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.500087044919433e-06,
      "loss": 0.7289,
      "step": 3283
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.495492120655464e-06,
      "loss": 0.7196,
      "step": 3284
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.490897816572021e-06,
      "loss": 0.6988,
      "step": 3285
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.486304134090552e-06,
      "loss": 0.7015,
      "step": 3286
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.481711074632307e-06,
      "loss": 0.708,
      "step": 3287
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.477118639618347e-06,
      "loss": 0.7004,
      "step": 3288
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.472526830469539e-06,
      "loss": 0.6626,
      "step": 3289
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.467935648606556e-06,
      "loss": 0.7265,
      "step": 3290
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.463345095449877e-06,
      "loss": 0.6942,
      "step": 3291
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.458755172419789e-06,
      "loss": 0.7137,
      "step": 3292
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.454165880936378e-06,
      "loss": 0.7062,
      "step": 3293
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.4495772224195394e-06,
      "loss": 0.686,
      "step": 3294
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.444989198288972e-06,
      "loss": 0.699,
      "step": 3295
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.440401809964178e-06,
      "loss": 0.6998,
      "step": 3296
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.435815058864462e-06,
      "loss": 0.7086,
      "step": 3297
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.431228946408933e-06,
      "loss": 0.6956,
      "step": 3298
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.426643474016502e-06,
      "loss": 0.6922,
      "step": 3299
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.4220586431058816e-06,
      "loss": 0.7268,
      "step": 3300
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.417474455095586e-06,
      "loss": 0.7133,
      "step": 3301
    },
    {
      "epoch": 2.33,
      "learning_rate": 6.412890911403928e-06,
      "loss": 0.6928,
      "step": 3302
    },
    {
      "epoch": 2.33,
      "learning_rate": 6.40830801344903e-06,
      "loss": 0.6932,
      "step": 3303
    },
    {
      "epoch": 2.33,
      "learning_rate": 6.403725762648801e-06,
      "loss": 0.6925,
      "step": 3304
    },
    {
      "epoch": 2.33,
      "learning_rate": 6.3991441604209605e-06,
      "loss": 0.7124,
      "step": 3305
    },
    {
      "epoch": 2.33,
      "learning_rate": 6.394563208183022e-06,
      "loss": 0.7052,
      "step": 3306
    },
    {
      "epoch": 2.33,
      "learning_rate": 6.389982907352302e-06,
      "loss": 0.685,
      "step": 3307
    },
    {
      "epoch": 2.33,
      "learning_rate": 6.3854032593459096e-06,
      "loss": 0.6814,
      "step": 3308
    },
    {
      "epoch": 2.33,
      "learning_rate": 6.380824265580759e-06,
      "loss": 0.689,
      "step": 3309
    },
    {
      "epoch": 2.33,
      "learning_rate": 6.376245927473554e-06,
      "loss": 0.6529,
      "step": 3310
    },
    {
      "epoch": 2.33,
      "learning_rate": 6.371668246440806e-06,
      "loss": 0.7214,
      "step": 3311
    },
    {
      "epoch": 2.33,
      "learning_rate": 6.367091223898813e-06,
      "loss": 0.7015,
      "step": 3312
    },
    {
      "epoch": 2.33,
      "learning_rate": 6.3625148612636755e-06,
      "loss": 0.7063,
      "step": 3313
    },
    {
      "epoch": 2.33,
      "learning_rate": 6.357939159951288e-06,
      "loss": 0.702,
      "step": 3314
    },
    {
      "epoch": 2.33,
      "learning_rate": 6.353364121377337e-06,
      "loss": 0.7225,
      "step": 3315
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.348789746957307e-06,
      "loss": 0.6987,
      "step": 3316
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.34421603810648e-06,
      "loss": 0.7053,
      "step": 3317
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.339642996239926e-06,
      "loss": 0.691,
      "step": 3318
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.335070622772513e-06,
      "loss": 0.6811,
      "step": 3319
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.330498919118903e-06,
      "loss": 0.715,
      "step": 3320
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.325927886693545e-06,
      "loss": 0.6984,
      "step": 3321
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.321357526910688e-06,
      "loss": 0.7011,
      "step": 3322
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.3167878411843665e-06,
      "loss": 0.7093,
      "step": 3323
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.312218830928409e-06,
      "loss": 0.69,
      "step": 3324
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.307650497556439e-06,
      "loss": 0.6653,
      "step": 3325
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.303082842481859e-06,
      "loss": 0.6728,
      "step": 3326
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.298515867117873e-06,
      "loss": 0.6694,
      "step": 3327
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.293949572877468e-06,
      "loss": 0.7139,
      "step": 3328
    },
    {
      "epoch": 2.34,
      "eval_loss": 0.724527895450592,
      "eval_runtime": 30.9174,
      "eval_samples_per_second": 264.932,
      "eval_steps_per_second": 16.56,
      "step": 3328
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.2893839611734275e-06,
      "loss": 0.6939,
      "step": 3329
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.284819033418318e-06,
      "loss": 0.6917,
      "step": 3330
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.280254791024494e-06,
      "loss": 0.7097,
      "step": 3331
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.275691235404102e-06,
      "loss": 0.6931,
      "step": 3332
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.2711283679690735e-06,
      "loss": 0.673,
      "step": 3333
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.266566190131125e-06,
      "loss": 0.6647,
      "step": 3334
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.262004703301765e-06,
      "loss": 0.7136,
      "step": 3335
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.257443908892286e-06,
      "loss": 0.7013,
      "step": 3336
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.25288380831376e-06,
      "loss": 0.667,
      "step": 3337
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.248324402977053e-06,
      "loss": 0.7117,
      "step": 3338
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.243765694292813e-06,
      "loss": 0.7158,
      "step": 3339
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.239207683671472e-06,
      "loss": 0.6905,
      "step": 3340
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.234650372523246e-06,
      "loss": 0.7044,
      "step": 3341
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.230093762258136e-06,
      "loss": 0.6931,
      "step": 3342
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.225537854285921e-06,
      "loss": 0.7184,
      "step": 3343
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.220982650016171e-06,
      "loss": 0.6954,
      "step": 3344
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.2164281508582326e-06,
      "loss": 0.7554,
      "step": 3345
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.211874358221234e-06,
      "loss": 0.685,
      "step": 3346
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.207321273514091e-06,
      "loss": 0.6964,
      "step": 3347
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.2027688981454895e-06,
      "loss": 0.6981,
      "step": 3348
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.198217233523905e-06,
      "loss": 0.6798,
      "step": 3349
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.193666281057589e-06,
      "loss": 0.6729,
      "step": 3350
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.189116042154577e-06,
      "loss": 0.713,
      "step": 3351
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.184566518222678e-06,
      "loss": 0.6669,
      "step": 3352
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.180017710669482e-06,
      "loss": 0.7165,
      "step": 3353
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.175469620902361e-06,
      "loss": 0.6947,
      "step": 3354
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.17092225032846e-06,
      "loss": 0.713,
      "step": 3355
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.166375600354702e-06,
      "loss": 0.7026,
      "step": 3356
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.161829672387789e-06,
      "loss": 0.6607,
      "step": 3357
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.157284467834201e-06,
      "loss": 0.6861,
      "step": 3358
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.152739988100189e-06,
      "loss": 0.688,
      "step": 3359
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.148196234591784e-06,
      "loss": 0.6688,
      "step": 3360
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.143653208714791e-06,
      "loss": 0.6953,
      "step": 3361
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.13911091187479e-06,
      "loss": 0.7129,
      "step": 3362
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.134569345477136e-06,
      "loss": 0.6666,
      "step": 3363
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.130028510926957e-06,
      "loss": 0.6855,
      "step": 3364
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.125488409629154e-06,
      "loss": 0.7105,
      "step": 3365
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.120949042988404e-06,
      "loss": 0.7016,
      "step": 3366
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.116410412409155e-06,
      "loss": 0.6689,
      "step": 3367
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.111872519295625e-06,
      "loss": 0.6871,
      "step": 3368
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.10733536505181e-06,
      "loss": 0.7203,
      "step": 3369
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.102798951081467e-06,
      "loss": 0.7183,
      "step": 3370
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.098263278788134e-06,
      "loss": 0.7254,
      "step": 3371
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.093728349575117e-06,
      "loss": 0.6444,
      "step": 3372
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.089194164845489e-06,
      "loss": 0.6839,
      "step": 3373
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.084660726002096e-06,
      "loss": 0.7138,
      "step": 3374
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.0801280344475505e-06,
      "loss": 0.6669,
      "step": 3375
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.075596091584235e-06,
      "loss": 0.6997,
      "step": 3376
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.0710648988143e-06,
      "loss": 0.7127,
      "step": 3377
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.066534457539668e-06,
      "loss": 0.6876,
      "step": 3378
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.0620047691620226e-06,
      "loss": 0.7035,
      "step": 3379
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.057475835082821e-06,
      "loss": 0.7,
      "step": 3380
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.052947656703276e-06,
      "loss": 0.6831,
      "step": 3381
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.048420235424379e-06,
      "loss": 0.7569,
      "step": 3382
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.043893572646881e-06,
      "loss": 0.6826,
      "step": 3383
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.0393676697712985e-06,
      "loss": 0.6859,
      "step": 3384
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.034842528197915e-06,
      "loss": 0.6942,
      "step": 3385
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.0303181493267745e-06,
      "loss": 0.7272,
      "step": 3386
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.025794534557692e-06,
      "loss": 0.6799,
      "step": 3387
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.021271685290241e-06,
      "loss": 0.6928,
      "step": 3388
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.016749602923757e-06,
      "loss": 0.6788,
      "step": 3389
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.012228288857341e-06,
      "loss": 0.7161,
      "step": 3390
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.007707744489856e-06,
      "loss": 0.6999,
      "step": 3391
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.003187971219929e-06,
      "loss": 0.6834,
      "step": 3392
    },
    {
      "epoch": 2.39,
      "eval_loss": 0.7247565388679504,
      "eval_runtime": 30.9208,
      "eval_samples_per_second": 264.902,
      "eval_steps_per_second": 16.558,
      "step": 3392
    },
    {
      "epoch": 2.39,
      "learning_rate": 5.99866897044594e-06,
      "loss": 0.6786,
      "step": 3393
    },
    {
      "epoch": 2.39,
      "learning_rate": 5.994150743566037e-06,
      "loss": 0.6732,
      "step": 3394
    },
    {
      "epoch": 2.39,
      "learning_rate": 5.989633291978128e-06,
      "loss": 0.6809,
      "step": 3395
    },
    {
      "epoch": 2.39,
      "learning_rate": 5.9851166170798855e-06,
      "loss": 0.6742,
      "step": 3396
    },
    {
      "epoch": 2.39,
      "learning_rate": 5.9806007202687306e-06,
      "loss": 0.7152,
      "step": 3397
    },
    {
      "epoch": 2.39,
      "learning_rate": 5.976085602941852e-06,
      "loss": 0.7064,
      "step": 3398
    },
    {
      "epoch": 2.39,
      "learning_rate": 5.971571266496195e-06,
      "loss": 0.6765,
      "step": 3399
    },
    {
      "epoch": 2.39,
      "learning_rate": 5.967057712328463e-06,
      "loss": 0.6783,
      "step": 3400
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.962544941835118e-06,
      "loss": 0.6662,
      "step": 3401
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.958032956412378e-06,
      "loss": 0.6769,
      "step": 3402
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.953521757456224e-06,
      "loss": 0.7119,
      "step": 3403
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.949011346362382e-06,
      "loss": 0.7138,
      "step": 3404
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.944501724526345e-06,
      "loss": 0.6733,
      "step": 3405
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.939992893343356e-06,
      "loss": 0.7231,
      "step": 3406
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.93548485420842e-06,
      "loss": 0.7122,
      "step": 3407
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.930977608516287e-06,
      "loss": 0.7004,
      "step": 3408
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.9264711576614694e-06,
      "loss": 0.7346,
      "step": 3409
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.921965503038232e-06,
      "loss": 0.7245,
      "step": 3410
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.91746064604059e-06,
      "loss": 0.7177,
      "step": 3411
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.912956588062315e-06,
      "loss": 0.7047,
      "step": 3412
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.9084533304969315e-06,
      "loss": 0.735,
      "step": 3413
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.903950874737718e-06,
      "loss": 0.6565,
      "step": 3414
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.899449222177696e-06,
      "loss": 0.6966,
      "step": 3415
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.89494837420965e-06,
      "loss": 0.6894,
      "step": 3416
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.890448332226108e-06,
      "loss": 0.6984,
      "step": 3417
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.885949097619352e-06,
      "loss": 0.6962,
      "step": 3418
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.881450671781413e-06,
      "loss": 0.7005,
      "step": 3419
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.876953056104072e-06,
      "loss": 0.7154,
      "step": 3420
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.87245625197886e-06,
      "loss": 0.707,
      "step": 3421
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.867960260797055e-06,
      "loss": 0.7189,
      "step": 3422
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.863465083949688e-06,
      "loss": 0.7214,
      "step": 3423
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.8589707228275305e-06,
      "loss": 0.7074,
      "step": 3424
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.854477178821111e-06,
      "loss": 0.7057,
      "step": 3425
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.8499844533206944e-06,
      "loss": 0.6975,
      "step": 3426
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.845492547716301e-06,
      "loss": 0.6606,
      "step": 3427
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.841001463397695e-06,
      "loss": 0.7316,
      "step": 3428
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.836511201754386e-06,
      "loss": 0.7183,
      "step": 3429
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.832021764175628e-06,
      "loss": 0.6913,
      "step": 3430
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.827533152050423e-06,
      "loss": 0.6871,
      "step": 3431
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.823045366767513e-06,
      "loss": 0.7255,
      "step": 3432
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.818558409715389e-06,
      "loss": 0.7078,
      "step": 3433
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.814072282282283e-06,
      "loss": 0.6624,
      "step": 3434
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.809586985856172e-06,
      "loss": 0.7053,
      "step": 3435
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.805102521824775e-06,
      "loss": 0.7377,
      "step": 3436
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.800618891575552e-06,
      "loss": 0.6844,
      "step": 3437
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.7961360964957065e-06,
      "loss": 0.7273,
      "step": 3438
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.791654137972184e-06,
      "loss": 0.7159,
      "step": 3439
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.78717301739167e-06,
      "loss": 0.7077,
      "step": 3440
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.782692736140593e-06,
      "loss": 0.7119,
      "step": 3441
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.778213295605119e-06,
      "loss": 0.7023,
      "step": 3442
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.773734697171154e-06,
      "loss": 0.6956,
      "step": 3443
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.769256942224347e-06,
      "loss": 0.6844,
      "step": 3444
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.764780032150083e-06,
      "loss": 0.704,
      "step": 3445
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.760303968333486e-06,
      "loss": 0.7122,
      "step": 3446
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.755828752159423e-06,
      "loss": 0.6451,
      "step": 3447
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.751354385012485e-06,
      "loss": 0.7003,
      "step": 3448
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.746880868277014e-06,
      "loss": 0.6983,
      "step": 3449
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.742408203337085e-06,
      "loss": 0.6967,
      "step": 3450
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.7379363915765085e-06,
      "loss": 0.7141,
      "step": 3451
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.73346543437883e-06,
      "loss": 0.6968,
      "step": 3452
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.728995333127333e-06,
      "loss": 0.6798,
      "step": 3453
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.724526089205035e-06,
      "loss": 0.7149,
      "step": 3454
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.7200577039946895e-06,
      "loss": 0.7006,
      "step": 3455
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.715590178878783e-06,
      "loss": 0.7269,
      "step": 3456
    },
    {
      "epoch": 2.43,
      "eval_loss": 0.7251064777374268,
      "eval_runtime": 30.9235,
      "eval_samples_per_second": 264.879,
      "eval_steps_per_second": 16.557,
      "step": 3456
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.711123515239534e-06,
      "loss": 0.66,
      "step": 3457
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.7066577144589035e-06,
      "loss": 0.6641,
      "step": 3458
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.702192777918567e-06,
      "loss": 0.7361,
      "step": 3459
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.6977287069999514e-06,
      "loss": 0.7106,
      "step": 3460
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.693265503084208e-06,
      "loss": 0.7144,
      "step": 3461
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.688803167552217e-06,
      "loss": 0.7199,
      "step": 3462
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.6843417017845985e-06,
      "loss": 0.6811,
      "step": 3463
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.679881107161696e-06,
      "loss": 0.7049,
      "step": 3464
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.675421385063587e-06,
      "loss": 0.6918,
      "step": 3465
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.6709625368700765e-06,
      "loss": 0.6859,
      "step": 3466
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.6665045639607e-06,
      "loss": 0.7093,
      "step": 3467
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.662047467714724e-06,
      "loss": 0.7103,
      "step": 3468
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.657591249511146e-06,
      "loss": 0.7211,
      "step": 3469
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.65313591072868e-06,
      "loss": 0.6987,
      "step": 3470
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.648681452745781e-06,
      "loss": 0.6867,
      "step": 3471
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.644227876940627e-06,
      "loss": 0.7207,
      "step": 3472
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.6397751846911225e-06,
      "loss": 0.7018,
      "step": 3473
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.635323377374898e-06,
      "loss": 0.7371,
      "step": 3474
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.630872456369311e-06,
      "loss": 0.7139,
      "step": 3475
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.626422423051447e-06,
      "loss": 0.6907,
      "step": 3476
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.621973278798113e-06,
      "loss": 0.7056,
      "step": 3477
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.617525024985842e-06,
      "loss": 0.7006,
      "step": 3478
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.613077662990896e-06,
      "loss": 0.7152,
      "step": 3479
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.608631194189254e-06,
      "loss": 0.692,
      "step": 3480
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.6041856199566205e-06,
      "loss": 0.697,
      "step": 3481
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.599740941668428e-06,
      "loss": 0.6814,
      "step": 3482
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.5952971606998266e-06,
      "loss": 0.6709,
      "step": 3483
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.5908542784256905e-06,
      "loss": 0.6849,
      "step": 3484
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.5864122962206175e-06,
      "loss": 0.7036,
      "step": 3485
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.581971215458924e-06,
      "loss": 0.726,
      "step": 3486
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.577531037514649e-06,
      "loss": 0.6889,
      "step": 3487
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.573091763761554e-06,
      "loss": 0.6959,
      "step": 3488
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.568653395573115e-06,
      "loss": 0.7178,
      "step": 3489
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.564215934322535e-06,
      "loss": 0.6971,
      "step": 3490
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.559779381382734e-06,
      "loss": 0.6924,
      "step": 3491
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.555343738126346e-06,
      "loss": 0.7151,
      "step": 3492
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.55090900592573e-06,
      "loss": 0.7043,
      "step": 3493
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.5464751861529595e-06,
      "loss": 0.7326,
      "step": 3494
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.542042280179829e-06,
      "loss": 0.6974,
      "step": 3495
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.537610289377846e-06,
      "loss": 0.6876,
      "step": 3496
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.5331792151182385e-06,
      "loss": 0.6847,
      "step": 3497
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.528749058771948e-06,
      "loss": 0.6788,
      "step": 3498
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.524319821709634e-06,
      "loss": 0.6703,
      "step": 3499
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.519891505301673e-06,
      "loss": 0.692,
      "step": 3500
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.515464110918152e-06,
      "loss": 0.6731,
      "step": 3501
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.5110376399288785e-06,
      "loss": 0.6785,
      "step": 3502
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.506612093703367e-06,
      "loss": 0.7202,
      "step": 3503
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.5021874736108524e-06,
      "loss": 0.6983,
      "step": 3504
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.497763781020282e-06,
      "loss": 0.718,
      "step": 3505
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.4933410173003134e-06,
      "loss": 0.6668,
      "step": 3506
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.488919183819318e-06,
      "loss": 0.7298,
      "step": 3507
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.484498281945381e-06,
      "loss": 0.7136,
      "step": 3508
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.480078313046297e-06,
      "loss": 0.6685,
      "step": 3509
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.475659278489574e-06,
      "loss": 0.6773,
      "step": 3510
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.4712411796424285e-06,
      "loss": 0.6765,
      "step": 3511
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.466824017871789e-06,
      "loss": 0.6906,
      "step": 3512
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.462407794544301e-06,
      "loss": 0.673,
      "step": 3513
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.457992511026303e-06,
      "loss": 0.7211,
      "step": 3514
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.453578168683854e-06,
      "loss": 0.7195,
      "step": 3515
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.449164768882725e-06,
      "loss": 0.7114,
      "step": 3516
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.4447523129883856e-06,
      "loss": 0.7215,
      "step": 3517
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.440340802366021e-06,
      "loss": 0.685,
      "step": 3518
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.4359302383805225e-06,
      "loss": 0.7505,
      "step": 3519
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.4315206223964835e-06,
      "loss": 0.7201,
      "step": 3520
    },
    {
      "epoch": 2.48,
      "eval_loss": 0.7247292399406433,
      "eval_runtime": 30.9117,
      "eval_samples_per_second": 264.981,
      "eval_steps_per_second": 16.563,
      "step": 3520
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.427111955778212e-06,
      "loss": 0.6838,
      "step": 3521
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.422704239889717e-06,
      "loss": 0.7251,
      "step": 3522
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.418297476094713e-06,
      "loss": 0.6723,
      "step": 3523
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.413891665756626e-06,
      "loss": 0.7275,
      "step": 3524
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.409486810238576e-06,
      "loss": 0.7068,
      "step": 3525
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.405082910903397e-06,
      "loss": 0.6771,
      "step": 3526
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.400679969113624e-06,
      "loss": 0.682,
      "step": 3527
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.396277986231497e-06,
      "loss": 0.6825,
      "step": 3528
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.391876963618955e-06,
      "loss": 0.6858,
      "step": 3529
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.387476902637648e-06,
      "loss": 0.7116,
      "step": 3530
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.383077804648921e-06,
      "loss": 0.6874,
      "step": 3531
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.378679671013825e-06,
      "loss": 0.6917,
      "step": 3532
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.374282503093109e-06,
      "loss": 0.7088,
      "step": 3533
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.369886302247228e-06,
      "loss": 0.6712,
      "step": 3534
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.3654910698363385e-06,
      "loss": 0.6874,
      "step": 3535
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.361096807220286e-06,
      "loss": 0.6762,
      "step": 3536
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.356703515758629e-06,
      "loss": 0.7167,
      "step": 3537
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.352311196810618e-06,
      "loss": 0.6959,
      "step": 3538
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.3479198517352095e-06,
      "loss": 0.6743,
      "step": 3539
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.343529481891052e-06,
      "loss": 0.6729,
      "step": 3540
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.339140088636495e-06,
      "loss": 0.6902,
      "step": 3541
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.334751673329588e-06,
      "loss": 0.6893,
      "step": 3542
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.3303642373280715e-06,
      "loss": 0.7181,
      "step": 3543
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.32597778198939e-06,
      "loss": 0.6849,
      "step": 3544
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.32159230867068e-06,
      "loss": 0.702,
      "step": 3545
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.31720781872878e-06,
      "loss": 0.6956,
      "step": 3546
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.312824313520214e-06,
      "loss": 0.6851,
      "step": 3547
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.30844179440121e-06,
      "loss": 0.7254,
      "step": 3548
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.30406026272769e-06,
      "loss": 0.7089,
      "step": 3549
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.299679719855267e-06,
      "loss": 0.7287,
      "step": 3550
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.295300167139249e-06,
      "loss": 0.694,
      "step": 3551
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.290921605934641e-06,
      "loss": 0.7145,
      "step": 3552
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.286544037596137e-06,
      "loss": 0.7185,
      "step": 3553
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.2821674634781275e-06,
      "loss": 0.687,
      "step": 3554
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.277791884934691e-06,
      "loss": 0.7292,
      "step": 3555
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.2734173033196024e-06,
      "loss": 0.6952,
      "step": 3556
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.269043719986328e-06,
      "loss": 0.6774,
      "step": 3557
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.264671136288019e-06,
      "loss": 0.6843,
      "step": 3558
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.260299553577524e-06,
      "loss": 0.6903,
      "step": 3559
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.255928973207379e-06,
      "loss": 0.712,
      "step": 3560
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.251559396529812e-06,
      "loss": 0.7044,
      "step": 3561
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.247190824896737e-06,
      "loss": 0.7023,
      "step": 3562
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.242823259659761e-06,
      "loss": 0.6829,
      "step": 3563
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.238456702170178e-06,
      "loss": 0.6901,
      "step": 3564
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.234091153778968e-06,
      "loss": 0.6711,
      "step": 3565
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.229726615836802e-06,
      "loss": 0.6858,
      "step": 3566
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.225363089694036e-06,
      "loss": 0.6877,
      "step": 3567
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.221000576700718e-06,
      "loss": 0.6694,
      "step": 3568
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.2166390782065725e-06,
      "loss": 0.6851,
      "step": 3569
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.21227859556102e-06,
      "loss": 0.6957,
      "step": 3570
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.207919130113162e-06,
      "loss": 0.6791,
      "step": 3571
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.203560683211786e-06,
      "loss": 0.6918,
      "step": 3572
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.199203256205365e-06,
      "loss": 0.6954,
      "step": 3573
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.194846850442056e-06,
      "loss": 0.6873,
      "step": 3574
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.1904914672697e-06,
      "loss": 0.6971,
      "step": 3575
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.18613710803582e-06,
      "loss": 0.7113,
      "step": 3576
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.181783774087627e-06,
      "loss": 0.6911,
      "step": 3577
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.17743146677201e-06,
      "loss": 0.7053,
      "step": 3578
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.173080187435544e-06,
      "loss": 0.7019,
      "step": 3579
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.16872993742448e-06,
      "loss": 0.6766,
      "step": 3580
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.164380718084757e-06,
      "loss": 0.668,
      "step": 3581
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.160032530761992e-06,
      "loss": 0.7198,
      "step": 3582
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.1556853768014845e-06,
      "loss": 0.7132,
      "step": 3583
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.151339257548211e-06,
      "loss": 0.7107,
      "step": 3584
    },
    {
      "epoch": 2.52,
      "eval_loss": 0.7242535948753357,
      "eval_runtime": 30.8971,
      "eval_samples_per_second": 265.106,
      "eval_steps_per_second": 16.571,
      "step": 3584
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.146994174346832e-06,
      "loss": 0.6895,
      "step": 3585
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.1426501285416855e-06,
      "loss": 0.6937,
      "step": 3586
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.138307121476784e-06,
      "loss": 0.7538,
      "step": 3587
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.133965154495831e-06,
      "loss": 0.6518,
      "step": 3588
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.129624228942194e-06,
      "loss": 0.7026,
      "step": 3589
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.125284346158931e-06,
      "loss": 0.6815,
      "step": 3590
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.120945507488762e-06,
      "loss": 0.6968,
      "step": 3591
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.116607714274099e-06,
      "loss": 0.7351,
      "step": 3592
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.112270967857022e-06,
      "loss": 0.7281,
      "step": 3593
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.1079352695792915e-06,
      "loss": 0.6842,
      "step": 3594
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.10360062078234e-06,
      "loss": 0.7022,
      "step": 3595
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.099267022807276e-06,
      "loss": 0.7159,
      "step": 3596
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.094934476994889e-06,
      "loss": 0.7098,
      "step": 3597
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.090602984685634e-06,
      "loss": 0.6697,
      "step": 3598
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.086272547219645e-06,
      "loss": 0.6857,
      "step": 3599
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.08194316593673e-06,
      "loss": 0.7208,
      "step": 3600
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.077614842176369e-06,
      "loss": 0.6634,
      "step": 3601
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.073287577277711e-06,
      "loss": 0.7154,
      "step": 3602
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.068961372579583e-06,
      "loss": 0.7127,
      "step": 3603
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.064636229420482e-06,
      "loss": 0.6809,
      "step": 3604
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.060312149138576e-06,
      "loss": 0.6774,
      "step": 3605
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.055989133071707e-06,
      "loss": 0.693,
      "step": 3606
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.051667182557385e-06,
      "loss": 0.6834,
      "step": 3607
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.047346298932788e-06,
      "loss": 0.7024,
      "step": 3608
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.043026483534771e-06,
      "loss": 0.6747,
      "step": 3609
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.038707737699851e-06,
      "loss": 0.677,
      "step": 3610
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.0343900627642196e-06,
      "loss": 0.6828,
      "step": 3611
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.030073460063736e-06,
      "loss": 0.7086,
      "step": 3612
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.02575793093392e-06,
      "loss": 0.6982,
      "step": 3613
    },
    {
      "epoch": 2.55,
      "learning_rate": 5.021443476709972e-06,
      "loss": 0.6739,
      "step": 3614
    },
    {
      "epoch": 2.55,
      "learning_rate": 5.01713009872675e-06,
      "loss": 0.7035,
      "step": 3615
    },
    {
      "epoch": 2.55,
      "learning_rate": 5.012817798318785e-06,
      "loss": 0.6999,
      "step": 3616
    },
    {
      "epoch": 2.55,
      "learning_rate": 5.008506576820271e-06,
      "loss": 0.6931,
      "step": 3617
    },
    {
      "epoch": 2.55,
      "learning_rate": 5.004196435565069e-06,
      "loss": 0.6916,
      "step": 3618
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.999887375886706e-06,
      "loss": 0.686,
      "step": 3619
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.9955793991183775e-06,
      "loss": 0.7162,
      "step": 3620
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.9912725065929374e-06,
      "loss": 0.703,
      "step": 3621
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.9869666996429085e-06,
      "loss": 0.7053,
      "step": 3622
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.98266197960048e-06,
      "loss": 0.7009,
      "step": 3623
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.978358347797498e-06,
      "loss": 0.6993,
      "step": 3624
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.974055805565475e-06,
      "loss": 0.7035,
      "step": 3625
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.96975435423559e-06,
      "loss": 0.6798,
      "step": 3626
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.965453995138681e-06,
      "loss": 0.6571,
      "step": 3627
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.9611547296052495e-06,
      "loss": 0.6965,
      "step": 3628
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.956856558965455e-06,
      "loss": 0.692,
      "step": 3629
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.9525594845491186e-06,
      "loss": 0.7045,
      "step": 3630
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.948263507685732e-06,
      "loss": 0.6923,
      "step": 3631
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.943968629704437e-06,
      "loss": 0.7264,
      "step": 3632
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.939674851934037e-06,
      "loss": 0.6948,
      "step": 3633
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.9353821757030005e-06,
      "loss": 0.7015,
      "step": 3634
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.931090602339445e-06,
      "loss": 0.687,
      "step": 3635
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.926800133171156e-06,
      "loss": 0.689,
      "step": 3636
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.922510769525573e-06,
      "loss": 0.696,
      "step": 3637
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.918222512729796e-06,
      "loss": 0.6956,
      "step": 3638
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.9139353641105815e-06,
      "loss": 0.6969,
      "step": 3639
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.909649324994342e-06,
      "loss": 0.6871,
      "step": 3640
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.905364396707148e-06,
      "loss": 0.7262,
      "step": 3641
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.901080580574725e-06,
      "loss": 0.7194,
      "step": 3642
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.896797877922458e-06,
      "loss": 0.7169,
      "step": 3643
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.892516290075382e-06,
      "loss": 0.6669,
      "step": 3644
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.888235818358194e-06,
      "loss": 0.7539,
      "step": 3645
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.883956464095236e-06,
      "loss": 0.7104,
      "step": 3646
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.879678228610509e-06,
      "loss": 0.7207,
      "step": 3647
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.875401113227675e-06,
      "loss": 0.715,
      "step": 3648
    },
    {
      "epoch": 2.57,
      "eval_loss": 0.7238507866859436,
      "eval_runtime": 30.9124,
      "eval_samples_per_second": 264.974,
      "eval_steps_per_second": 16.563,
      "step": 3648
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.87112511927004e-06,
      "loss": 0.6971,
      "step": 3649
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.866850248060567e-06,
      "loss": 0.7023,
      "step": 3650
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.862576500921871e-06,
      "loss": 0.7049,
      "step": 3651
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.858303879176218e-06,
      "loss": 0.67,
      "step": 3652
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.854032384145529e-06,
      "loss": 0.7147,
      "step": 3653
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.849762017151369e-06,
      "loss": 0.7104,
      "step": 3654
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.845492779514966e-06,
      "loss": 0.6917,
      "step": 3655
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.841224672557193e-06,
      "loss": 0.6916,
      "step": 3656
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.8369576975985615e-06,
      "loss": 0.715,
      "step": 3657
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.832691855959251e-06,
      "loss": 0.6695,
      "step": 3658
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.82842714895908e-06,
      "loss": 0.6859,
      "step": 3659
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.824163577917521e-06,
      "loss": 0.6931,
      "step": 3660
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.819901144153692e-06,
      "loss": 0.6681,
      "step": 3661
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.815639848986359e-06,
      "loss": 0.7008,
      "step": 3662
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.8113796937339356e-06,
      "loss": 0.7216,
      "step": 3663
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.807120679714487e-06,
      "loss": 0.6893,
      "step": 3664
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.802862808245719e-06,
      "loss": 0.6691,
      "step": 3665
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.798606080644987e-06,
      "loss": 0.6865,
      "step": 3666
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.7943504982292935e-06,
      "loss": 0.704,
      "step": 3667
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.790096062315285e-06,
      "loss": 0.6544,
      "step": 3668
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.785842774219254e-06,
      "loss": 0.6658,
      "step": 3669
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.781590635257137e-06,
      "loss": 0.7047,
      "step": 3670
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.7773396467445166e-06,
      "loss": 0.7142,
      "step": 3671
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.773089809996618e-06,
      "loss": 0.6745,
      "step": 3672
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.768841126328311e-06,
      "loss": 0.6982,
      "step": 3673
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.764593597054107e-06,
      "loss": 0.6943,
      "step": 3674
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.760347223488163e-06,
      "loss": 0.7103,
      "step": 3675
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.7561020069442765e-06,
      "loss": 0.7138,
      "step": 3676
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.751857948735886e-06,
      "loss": 0.6993,
      "step": 3677
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.747615050176079e-06,
      "loss": 0.7117,
      "step": 3678
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.743373312577568e-06,
      "loss": 0.6884,
      "step": 3679
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.739132737252723e-06,
      "loss": 0.7115,
      "step": 3680
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.734893325513547e-06,
      "loss": 0.6838,
      "step": 3681
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.730655078671683e-06,
      "loss": 0.7048,
      "step": 3682
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.726417998038416e-06,
      "loss": 0.6555,
      "step": 3683
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.722182084924669e-06,
      "loss": 0.7188,
      "step": 3684
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.717947340641001e-06,
      "loss": 0.707,
      "step": 3685
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.713713766497614e-06,
      "loss": 0.6822,
      "step": 3686
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.709481363804344e-06,
      "loss": 0.7187,
      "step": 3687
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.705250133870664e-06,
      "loss": 0.7238,
      "step": 3688
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.701020078005699e-06,
      "loss": 0.6913,
      "step": 3689
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.696791197518185e-06,
      "loss": 0.6973,
      "step": 3690
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.692563493716512e-06,
      "loss": 0.715,
      "step": 3691
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.688336967908702e-06,
      "loss": 0.7134,
      "step": 3692
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.68411162140241e-06,
      "loss": 0.6879,
      "step": 3693
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.679887455504931e-06,
      "loss": 0.6902,
      "step": 3694
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.675664471523192e-06,
      "loss": 0.6881,
      "step": 3695
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.671442670763753e-06,
      "loss": 0.6811,
      "step": 3696
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.667222054532809e-06,
      "loss": 0.7009,
      "step": 3697
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.66300262413619e-06,
      "loss": 0.6742,
      "step": 3698
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.658784380879356e-06,
      "loss": 0.7104,
      "step": 3699
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.654567326067406e-06,
      "loss": 0.6794,
      "step": 3700
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.650351461005061e-06,
      "loss": 0.6708,
      "step": 3701
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.646136786996681e-06,
      "loss": 0.7014,
      "step": 3702
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.641923305346259e-06,
      "loss": 0.696,
      "step": 3703
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.637711017357415e-06,
      "loss": 0.7063,
      "step": 3704
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.633499924333396e-06,
      "loss": 0.688,
      "step": 3705
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.629290027577095e-06,
      "loss": 0.6686,
      "step": 3706
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.625081328391017e-06,
      "loss": 0.6851,
      "step": 3707
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.620873828077305e-06,
      "loss": 0.7006,
      "step": 3708
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.61666752793773e-06,
      "loss": 0.6957,
      "step": 3709
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.612462429273692e-06,
      "loss": 0.6832,
      "step": 3710
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.608258533386221e-06,
      "loss": 0.6714,
      "step": 3711
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.604055841575965e-06,
      "loss": 0.6906,
      "step": 3712
    },
    {
      "epoch": 2.61,
      "eval_loss": 0.7237129807472229,
      "eval_runtime": 30.9049,
      "eval_samples_per_second": 265.039,
      "eval_steps_per_second": 16.567,
      "step": 3712
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.599854355143215e-06,
      "loss": 0.6875,
      "step": 3713
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.5956540753878744e-06,
      "loss": 0.7236,
      "step": 3714
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.591455003609486e-06,
      "loss": 0.6994,
      "step": 3715
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.5872571411072085e-06,
      "loss": 0.7102,
      "step": 3716
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.583060489179834e-06,
      "loss": 0.6796,
      "step": 3717
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.578865049125774e-06,
      "loss": 0.7074,
      "step": 3718
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.574670822243071e-06,
      "loss": 0.6857,
      "step": 3719
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.570477809829387e-06,
      "loss": 0.7186,
      "step": 3720
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.56628601318201e-06,
      "loss": 0.7149,
      "step": 3721
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.562095433597856e-06,
      "loss": 0.7192,
      "step": 3722
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.557906072373451e-06,
      "loss": 0.7172,
      "step": 3723
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.553717930804964e-06,
      "loss": 0.7194,
      "step": 3724
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.549531010188171e-06,
      "loss": 0.7121,
      "step": 3725
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.545345311818479e-06,
      "loss": 0.7236,
      "step": 3726
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.5411608369909104e-06,
      "loss": 0.6893,
      "step": 3727
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.5369775870001135e-06,
      "loss": 0.7062,
      "step": 3728
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.532795563140356e-06,
      "loss": 0.7141,
      "step": 3729
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.528614766705525e-06,
      "loss": 0.7,
      "step": 3730
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.524435198989131e-06,
      "loss": 0.6832,
      "step": 3731
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.520256861284302e-06,
      "loss": 0.6902,
      "step": 3732
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.516079754883791e-06,
      "loss": 0.6852,
      "step": 3733
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.511903881079955e-06,
      "loss": 0.6903,
      "step": 3734
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.507729241164787e-06,
      "loss": 0.6921,
      "step": 3735
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.503555836429886e-06,
      "loss": 0.6872,
      "step": 3736
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.4993836681664785e-06,
      "loss": 0.6711,
      "step": 3737
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.4952127376654025e-06,
      "loss": 0.7025,
      "step": 3738
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.491043046217114e-06,
      "loss": 0.6803,
      "step": 3739
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.486874595111686e-06,
      "loss": 0.726,
      "step": 3740
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.482707385638807e-06,
      "loss": 0.6956,
      "step": 3741
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.478541419087782e-06,
      "loss": 0.6783,
      "step": 3742
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.474376696747533e-06,
      "loss": 0.6959,
      "step": 3743
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.470213219906595e-06,
      "loss": 0.6792,
      "step": 3744
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.466050989853118e-06,
      "loss": 0.6915,
      "step": 3745
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.461890007874867e-06,
      "loss": 0.7085,
      "step": 3746
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.4577302752592195e-06,
      "loss": 0.6657,
      "step": 3747
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.453571793293167e-06,
      "loss": 0.7037,
      "step": 3748
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.449414563263316e-06,
      "loss": 0.6775,
      "step": 3749
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.445258586455882e-06,
      "loss": 0.7303,
      "step": 3750
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.441103864156695e-06,
      "loss": 0.6989,
      "step": 3751
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.436950397651199e-06,
      "loss": 0.6819,
      "step": 3752
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.432798188224445e-06,
      "loss": 0.7353,
      "step": 3753
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.428647237161097e-06,
      "loss": 0.7023,
      "step": 3754
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.424497545745431e-06,
      "loss": 0.6906,
      "step": 3755
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.420349115261334e-06,
      "loss": 0.6957,
      "step": 3756
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.416201946992295e-06,
      "loss": 0.6888,
      "step": 3757
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.412056042221424e-06,
      "loss": 0.7038,
      "step": 3758
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.40791140223143e-06,
      "loss": 0.6817,
      "step": 3759
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.4037680283046385e-06,
      "loss": 0.7009,
      "step": 3760
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.399625921722979e-06,
      "loss": 0.6919,
      "step": 3761
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.395485083767992e-06,
      "loss": 0.729,
      "step": 3762
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.39134551572082e-06,
      "loss": 0.6856,
      "step": 3763
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.387207218862214e-06,
      "loss": 0.7115,
      "step": 3764
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.383070194472541e-06,
      "loss": 0.7291,
      "step": 3765
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.378934443831764e-06,
      "loss": 0.7185,
      "step": 3766
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.374799968219456e-06,
      "loss": 0.6877,
      "step": 3767
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.370666768914791e-06,
      "loss": 0.695,
      "step": 3768
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.366534847196552e-06,
      "loss": 0.7319,
      "step": 3769
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.362404204343128e-06,
      "loss": 0.6898,
      "step": 3770
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.358274841632509e-06,
      "loss": 0.7113,
      "step": 3771
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.354146760342294e-06,
      "loss": 0.6824,
      "step": 3772
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.350019961749678e-06,
      "loss": 0.6915,
      "step": 3773
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.3458944471314655e-06,
      "loss": 0.7264,
      "step": 3774
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.341770217764061e-06,
      "loss": 0.7017,
      "step": 3775
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.3376472749234725e-06,
      "loss": 0.6819,
      "step": 3776
    },
    {
      "epoch": 2.66,
      "eval_loss": 0.7233282327651978,
      "eval_runtime": 30.9255,
      "eval_samples_per_second": 264.862,
      "eval_steps_per_second": 16.556,
      "step": 3776
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.333525619885308e-06,
      "loss": 0.7049,
      "step": 3777
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.329405253924782e-06,
      "loss": 0.7352,
      "step": 3778
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.3252861783167e-06,
      "loss": 0.7102,
      "step": 3779
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.321168394335479e-06,
      "loss": 0.7076,
      "step": 3780
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.3170519032551274e-06,
      "loss": 0.7107,
      "step": 3781
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.312936706349266e-06,
      "loss": 0.7016,
      "step": 3782
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.308822804891104e-06,
      "loss": 0.7101,
      "step": 3783
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.304710200153453e-06,
      "loss": 0.7366,
      "step": 3784
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.300598893408725e-06,
      "loss": 0.7116,
      "step": 3785
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.296488885928926e-06,
      "loss": 0.7193,
      "step": 3786
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.2923801789856665e-06,
      "loss": 0.696,
      "step": 3787
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.28827277385015e-06,
      "loss": 0.6922,
      "step": 3788
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.284166671793183e-06,
      "loss": 0.6982,
      "step": 3789
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.280061874085156e-06,
      "loss": 0.6896,
      "step": 3790
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.275958381996068e-06,
      "loss": 0.6857,
      "step": 3791
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.271856196795511e-06,
      "loss": 0.708,
      "step": 3792
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.267755319752672e-06,
      "loss": 0.7055,
      "step": 3793
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.2636557521363324e-06,
      "loss": 0.675,
      "step": 3794
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.259557495214871e-06,
      "loss": 0.6816,
      "step": 3795
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.255460550256258e-06,
      "loss": 0.6702,
      "step": 3796
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.251364918528058e-06,
      "loss": 0.7235,
      "step": 3797
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.2472706012974335e-06,
      "loss": 0.7141,
      "step": 3798
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.243177599831137e-06,
      "loss": 0.7213,
      "step": 3799
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.239085915395511e-06,
      "loss": 0.684,
      "step": 3800
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.234995549256495e-06,
      "loss": 0.692,
      "step": 3801
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.23090650267962e-06,
      "loss": 0.7217,
      "step": 3802
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.226818776930009e-06,
      "loss": 0.7058,
      "step": 3803
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.2227323732723734e-06,
      "loss": 0.6719,
      "step": 3804
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.218647292971018e-06,
      "loss": 0.7259,
      "step": 3805
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.214563537289839e-06,
      "loss": 0.7358,
      "step": 3806
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.21048110749232e-06,
      "loss": 0.6575,
      "step": 3807
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.206400004841535e-06,
      "loss": 0.6846,
      "step": 3808
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.202320230600152e-06,
      "loss": 0.6953,
      "step": 3809
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.19824178603042e-06,
      "loss": 0.6977,
      "step": 3810
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.194164672394186e-06,
      "loss": 0.7134,
      "step": 3811
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.190088890952873e-06,
      "loss": 0.6968,
      "step": 3812
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.186014442967502e-06,
      "loss": 0.6944,
      "step": 3813
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.18194132969868e-06,
      "loss": 0.7172,
      "step": 3814
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.177869552406598e-06,
      "loss": 0.6883,
      "step": 3815
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.173799112351035e-06,
      "loss": 0.6862,
      "step": 3816
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.169730010791356e-06,
      "loss": 0.6861,
      "step": 3817
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.165662248986514e-06,
      "loss": 0.7123,
      "step": 3818
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.1615958281950435e-06,
      "loss": 0.6993,
      "step": 3819
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.157530749675067e-06,
      "loss": 0.6976,
      "step": 3820
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.153467014684291e-06,
      "loss": 0.7159,
      "step": 3821
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.149404624480006e-06,
      "loss": 0.6736,
      "step": 3822
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.145343580319088e-06,
      "loss": 0.694,
      "step": 3823
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.141283883457995e-06,
      "loss": 0.6828,
      "step": 3824
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.1372255351527666e-06,
      "loss": 0.7014,
      "step": 3825
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.133168536659028e-06,
      "loss": 0.6892,
      "step": 3826
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.129112889231986e-06,
      "loss": 0.711,
      "step": 3827
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.12505859412643e-06,
      "loss": 0.6653,
      "step": 3828
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.1210056525967285e-06,
      "loss": 0.6926,
      "step": 3829
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.1169540658968345e-06,
      "loss": 0.6793,
      "step": 3830
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.112903835280279e-06,
      "loss": 0.6845,
      "step": 3831
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.108854962000174e-06,
      "loss": 0.6988,
      "step": 3832
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.1048074473092175e-06,
      "loss": 0.7165,
      "step": 3833
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.100761292459675e-06,
      "loss": 0.7212,
      "step": 3834
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.0967164987033995e-06,
      "loss": 0.6899,
      "step": 3835
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.092673067291823e-06,
      "loss": 0.6609,
      "step": 3836
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.088630999475954e-06,
      "loss": 0.6943,
      "step": 3837
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.08459029650638e-06,
      "loss": 0.7311,
      "step": 3838
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.080550959633266e-06,
      "loss": 0.6705,
      "step": 3839
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.07651299010635e-06,
      "loss": 0.7169,
      "step": 3840
    },
    {
      "epoch": 2.7,
      "eval_loss": 0.7229313254356384,
      "eval_runtime": 30.9076,
      "eval_samples_per_second": 265.016,
      "eval_steps_per_second": 16.566,
      "step": 3840
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.072476389174959e-06,
      "loss": 0.7119,
      "step": 3841
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.068441158087984e-06,
      "loss": 0.7188,
      "step": 3842
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.064407298093898e-06,
      "loss": 0.7174,
      "step": 3843
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.060374810440752e-06,
      "loss": 0.6934,
      "step": 3844
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.056343696376162e-06,
      "loss": 0.7355,
      "step": 3845
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.0523139571473285e-06,
      "loss": 0.7051,
      "step": 3846
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.048285594001026e-06,
      "loss": 0.6904,
      "step": 3847
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.0442586081836005e-06,
      "loss": 0.7008,
      "step": 3848
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.040233000940974e-06,
      "loss": 0.688,
      "step": 3849
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.036208773518639e-06,
      "loss": 0.6913,
      "step": 3850
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.032185927161664e-06,
      "loss": 0.6887,
      "step": 3851
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.028164463114689e-06,
      "loss": 0.7101,
      "step": 3852
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.024144382621928e-06,
      "loss": 0.7127,
      "step": 3853
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.020125686927162e-06,
      "loss": 0.6868,
      "step": 3854
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.016108377273753e-06,
      "loss": 0.7257,
      "step": 3855
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.01209245490462e-06,
      "loss": 0.6883,
      "step": 3856
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.0080779210622625e-06,
      "loss": 0.6556,
      "step": 3857
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.004064776988754e-06,
      "loss": 0.709,
      "step": 3858
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.00005302392573e-06,
      "loss": 0.6995,
      "step": 3859
    },
    {
      "epoch": 2.72,
      "learning_rate": 3.996042663114399e-06,
      "loss": 0.7122,
      "step": 3860
    },
    {
      "epoch": 2.72,
      "learning_rate": 3.992033695795538e-06,
      "loss": 0.6968,
      "step": 3861
    },
    {
      "epoch": 2.72,
      "learning_rate": 3.988026123209491e-06,
      "loss": 0.6969,
      "step": 3862
    },
    {
      "epoch": 2.72,
      "learning_rate": 3.984019946596177e-06,
      "loss": 0.6707,
      "step": 3863
    },
    {
      "epoch": 2.72,
      "learning_rate": 3.980015167195074e-06,
      "loss": 0.6789,
      "step": 3864
    },
    {
      "epoch": 2.72,
      "learning_rate": 3.9760117862452346e-06,
      "loss": 0.7,
      "step": 3865
    },
    {
      "epoch": 2.72,
      "learning_rate": 3.9720098049852796e-06,
      "loss": 0.6997,
      "step": 3866
    },
    {
      "epoch": 2.72,
      "learning_rate": 3.968009224653387e-06,
      "loss": 0.6867,
      "step": 3867
    },
    {
      "epoch": 2.72,
      "learning_rate": 3.9640100464873115e-06,
      "loss": 0.6888,
      "step": 3868
    },
    {
      "epoch": 2.72,
      "learning_rate": 3.960012271724369e-06,
      "loss": 0.6971,
      "step": 3869
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.956015901601446e-06,
      "loss": 0.7357,
      "step": 3870
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.952020937354987e-06,
      "loss": 0.7329,
      "step": 3871
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.948027380221009e-06,
      "loss": 0.7103,
      "step": 3872
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.944035231435089e-06,
      "loss": 0.7037,
      "step": 3873
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.940044492232368e-06,
      "loss": 0.6922,
      "step": 3874
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.936055163847554e-06,
      "loss": 0.6933,
      "step": 3875
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.9320672475149176e-06,
      "loss": 0.6978,
      "step": 3876
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.92808074446829e-06,
      "loss": 0.7034,
      "step": 3877
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.924095655941068e-06,
      "loss": 0.6973,
      "step": 3878
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.92011198316621e-06,
      "loss": 0.6974,
      "step": 3879
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.916129727376236e-06,
      "loss": 0.6809,
      "step": 3880
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.912148889803228e-06,
      "loss": 0.7047,
      "step": 3881
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.9081694716788275e-06,
      "loss": 0.6866,
      "step": 3882
    },
    {
      "epoch": 2.73,
      "learning_rate": 3.9041914742342385e-06,
      "loss": 0.7035,
      "step": 3883
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.900214898700227e-06,
      "loss": 0.6972,
      "step": 3884
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.8962397463071156e-06,
      "loss": 0.7253,
      "step": 3885
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.89226601828479e-06,
      "loss": 0.6557,
      "step": 3886
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.888293715862691e-06,
      "loss": 0.6832,
      "step": 3887
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.884322840269826e-06,
      "loss": 0.6897,
      "step": 3888
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.880353392734747e-06,
      "loss": 0.7144,
      "step": 3889
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.87638537448558e-06,
      "loss": 0.6916,
      "step": 3890
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.872418786749998e-06,
      "loss": 0.7259,
      "step": 3891
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.868453630755238e-06,
      "loss": 0.6872,
      "step": 3892
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.86448990772809e-06,
      "loss": 0.7051,
      "step": 3893
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.860527618894899e-06,
      "loss": 0.6985,
      "step": 3894
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.856566765481575e-06,
      "loss": 0.6971,
      "step": 3895
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.852607348713572e-06,
      "loss": 0.6548,
      "step": 3896
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.848649369815909e-06,
      "loss": 0.7097,
      "step": 3897
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.844692830013153e-06,
      "loss": 0.6773,
      "step": 3898
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.840737730529438e-06,
      "loss": 0.6683,
      "step": 3899
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.836784072588435e-06,
      "loss": 0.7142,
      "step": 3900
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.83283185741338e-06,
      "loss": 0.7078,
      "step": 3901
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.828881086227061e-06,
      "loss": 0.7046,
      "step": 3902
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.824931760251819e-06,
      "loss": 0.7074,
      "step": 3903
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.820983880709546e-06,
      "loss": 0.7236,
      "step": 3904
    },
    {
      "epoch": 2.75,
      "eval_loss": 0.7227227091789246,
      "eval_runtime": 30.9263,
      "eval_samples_per_second": 264.856,
      "eval_steps_per_second": 16.556,
      "step": 3904
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.81703744882169e-06,
      "loss": 0.6907,
      "step": 3905
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.8130924658092486e-06,
      "loss": 0.6879,
      "step": 3906
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.80914893289277e-06,
      "loss": 0.6853,
      "step": 3907
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.805206851292356e-06,
      "loss": 0.6954,
      "step": 3908
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.8012662222276587e-06,
      "loss": 0.7021,
      "step": 3909
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.797327046917885e-06,
      "loss": 0.7153,
      "step": 3910
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.7933893265817808e-06,
      "loss": 0.6664,
      "step": 3911
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.7894530624376514e-06,
      "loss": 0.6917,
      "step": 3912
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.78551825570335e-06,
      "loss": 0.709,
      "step": 3913
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.7815849075962772e-06,
      "loss": 0.6892,
      "step": 3914
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.7776530193333835e-06,
      "loss": 0.691,
      "step": 3915
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.7737225921311637e-06,
      "loss": 0.6917,
      "step": 3916
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.7697936272056708e-06,
      "loss": 0.7007,
      "step": 3917
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.7658661257724945e-06,
      "loss": 0.6827,
      "step": 3918
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.7619400890467774e-06,
      "loss": 0.7335,
      "step": 3919
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.758015518243207e-06,
      "loss": 0.6932,
      "step": 3920
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.754092414576021e-06,
      "loss": 0.6748,
      "step": 3921
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.750170779258992e-06,
      "loss": 0.658,
      "step": 3922
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.746250613505451e-06,
      "loss": 0.7216,
      "step": 3923
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.7423319185282696e-06,
      "loss": 0.7035,
      "step": 3924
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.7384146955398625e-06,
      "loss": 0.7127,
      "step": 3925
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.7344989457521927e-06,
      "loss": 0.7024,
      "step": 3926
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.7305846703767645e-06,
      "loss": 0.6893,
      "step": 3927
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.7266718706246267e-06,
      "loss": 0.6885,
      "step": 3928
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.722760547706372e-06,
      "loss": 0.6979,
      "step": 3929
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.718850702832137e-06,
      "loss": 0.6894,
      "step": 3930
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.714942337211598e-06,
      "loss": 0.6993,
      "step": 3931
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.711035452053981e-06,
      "loss": 0.7149,
      "step": 3932
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.707130048568037e-06,
      "loss": 0.7127,
      "step": 3933
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.7032261279620803e-06,
      "loss": 0.6548,
      "step": 3934
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.6993236914439536e-06,
      "loss": 0.7187,
      "step": 3935
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.6954227402210444e-06,
      "loss": 0.6895,
      "step": 3936
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.691523275500276e-06,
      "loss": 0.6918,
      "step": 3937
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.687625298488118e-06,
      "loss": 0.683,
      "step": 3938
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.683728810390576e-06,
      "loss": 0.6806,
      "step": 3939
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.6798338124131947e-06,
      "loss": 0.7244,
      "step": 3940
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.67594030576106e-06,
      "loss": 0.7095,
      "step": 3941
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.6720482916387947e-06,
      "loss": 0.7033,
      "step": 3942
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.6681577712505647e-06,
      "loss": 0.6803,
      "step": 3943
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.6642687458000622e-06,
      "loss": 0.6728,
      "step": 3944
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.660381216490527e-06,
      "loss": 0.6914,
      "step": 3945
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.6564951845247335e-06,
      "loss": 0.7022,
      "step": 3946
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.6526106511049925e-06,
      "loss": 0.7162,
      "step": 3947
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.648727617433151e-06,
      "loss": 0.712,
      "step": 3948
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.6448460847105926e-06,
      "loss": 0.6857,
      "step": 3949
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.6409660541382344e-06,
      "loss": 0.6762,
      "step": 3950
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.6370875269165326e-06,
      "loss": 0.7078,
      "step": 3951
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.633210504245474e-06,
      "loss": 0.6953,
      "step": 3952
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.6293349873245827e-06,
      "loss": 0.7034,
      "step": 3953
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.6254609773529168e-06,
      "loss": 0.6961,
      "step": 3954
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.621588475529067e-06,
      "loss": 0.6817,
      "step": 3955
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.617717483051158e-06,
      "loss": 0.6885,
      "step": 3956
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.613848001116847e-06,
      "loss": 0.7193,
      "step": 3957
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.6099800309233247e-06,
      "loss": 0.6858,
      "step": 3958
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.606113573667313e-06,
      "loss": 0.7058,
      "step": 3959
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.602248630545068e-06,
      "loss": 0.7073,
      "step": 3960
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.5983852027523737e-06,
      "loss": 0.6836,
      "step": 3961
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.5945232914845485e-06,
      "loss": 0.6989,
      "step": 3962
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.59066289793644e-06,
      "loss": 0.6921,
      "step": 3963
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.586804023302427e-06,
      "loss": 0.7035,
      "step": 3964
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.582946668776421e-06,
      "loss": 0.7016,
      "step": 3965
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.579090835551853e-06,
      "loss": 0.6626,
      "step": 3966
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.5752365248216936e-06,
      "loss": 0.7102,
      "step": 3967
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.571383737778441e-06,
      "loss": 0.6596,
      "step": 3968
    },
    {
      "epoch": 2.79,
      "eval_loss": 0.7223830819129944,
      "eval_runtime": 30.8959,
      "eval_samples_per_second": 265.116,
      "eval_steps_per_second": 16.572,
      "step": 3968
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.5675324756141183e-06,
      "loss": 0.693,
      "step": 3969
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.56368273952028e-06,
      "loss": 0.6628,
      "step": 3970
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.5598345306880066e-06,
      "loss": 0.7376,
      "step": 3971
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.555987850307906e-06,
      "loss": 0.7116,
      "step": 3972
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.552142699570115e-06,
      "loss": 0.687,
      "step": 3973
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.548299079664292e-06,
      "loss": 0.7087,
      "step": 3974
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.544456991779633e-06,
      "loss": 0.6997,
      "step": 3975
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.5406164371048515e-06,
      "loss": 0.6988,
      "step": 3976
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.536777416828182e-06,
      "loss": 0.7103,
      "step": 3977
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.532939932137394e-06,
      "loss": 0.6977,
      "step": 3978
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.5291039842197765e-06,
      "loss": 0.6845,
      "step": 3979
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.525269574262147e-06,
      "loss": 0.7021,
      "step": 3980
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.5214367034508427e-06,
      "loss": 0.6985,
      "step": 3981
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.5176053729717276e-06,
      "loss": 0.7029,
      "step": 3982
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.5137755840101873e-06,
      "loss": 0.662,
      "step": 3983
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.5099473377511337e-06,
      "loss": 0.6906,
      "step": 3984
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.5061206353789958e-06,
      "loss": 0.6765,
      "step": 3985
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.5022954780777303e-06,
      "loss": 0.6988,
      "step": 3986
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.4984718670308156e-06,
      "loss": 0.713,
      "step": 3987
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.4946498034212446e-06,
      "loss": 0.7087,
      "step": 3988
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.4908292884315405e-06,
      "loss": 0.6757,
      "step": 3989
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.4870103232437415e-06,
      "loss": 0.6618,
      "step": 3990
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.4831929090394073e-06,
      "loss": 0.7214,
      "step": 3991
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.4793770469996226e-06,
      "loss": 0.7049,
      "step": 3992
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.4755627383049873e-06,
      "loss": 0.6955,
      "step": 3993
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.471749984135621e-06,
      "loss": 0.6972,
      "step": 3994
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.4679387856711612e-06,
      "loss": 0.6691,
      "step": 3995
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.464129144090768e-06,
      "loss": 0.6592,
      "step": 3996
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.4603210605731163e-06,
      "loss": 0.7126,
      "step": 3997
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.456514536296402e-06,
      "loss": 0.7197,
      "step": 3998
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.452709572438333e-06,
      "loss": 0.6991,
      "step": 3999
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.44890617017614e-06,
      "loss": 0.6838,
      "step": 4000
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.4451043306865677e-06,
      "loss": 0.6953,
      "step": 4001
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.4413040551458802e-06,
      "loss": 0.7076,
      "step": 4002
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.4375053447298553e-06,
      "loss": 0.7125,
      "step": 4003
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.433708200613786e-06,
      "loss": 0.6899,
      "step": 4004
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.429912623972483e-06,
      "loss": 0.6787,
      "step": 4005
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.4261186159802695e-06,
      "loss": 0.7114,
      "step": 4006
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.422326177810986e-06,
      "loss": 0.6825,
      "step": 4007
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.418535310637986e-06,
      "loss": 0.6831,
      "step": 4008
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.4147460156341357e-06,
      "loss": 0.7231,
      "step": 4009
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.410958293971817e-06,
      "loss": 0.7315,
      "step": 4010
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.4071721468229236e-06,
      "loss": 0.6725,
      "step": 4011
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.4033875753588635e-06,
      "loss": 0.6809,
      "step": 4012
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.3996045807505556e-06,
      "loss": 0.6871,
      "step": 4013
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.3958231641684316e-06,
      "loss": 0.718,
      "step": 4014
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.392043326782435e-06,
      "loss": 0.7016,
      "step": 4015
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.388265069762021e-06,
      "loss": 0.7257,
      "step": 4016
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.3844883942761548e-06,
      "loss": 0.7045,
      "step": 4017
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.3807133014933136e-06,
      "loss": 0.6748,
      "step": 4018
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.376939792581485e-06,
      "loss": 0.7232,
      "step": 4019
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.3731678687081675e-06,
      "loss": 0.6828,
      "step": 4020
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.3693975310403623e-06,
      "loss": 0.6841,
      "step": 4021
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.3656287807445884e-06,
      "loss": 0.6987,
      "step": 4022
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.3618616189868695e-06,
      "loss": 0.6642,
      "step": 4023
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.3580960469327408e-06,
      "loss": 0.6935,
      "step": 4024
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.3543320657472406e-06,
      "loss": 0.6692,
      "step": 4025
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.3505696765949202e-06,
      "loss": 0.7156,
      "step": 4026
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.346808880639836e-06,
      "loss": 0.6812,
      "step": 4027
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.3430496790455504e-06,
      "loss": 0.7288,
      "step": 4028
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.3392920729751352e-06,
      "loss": 0.7299,
      "step": 4029
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.335536063591166e-06,
      "loss": 0.6643,
      "step": 4030
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.331781652055726e-06,
      "loss": 0.6894,
      "step": 4031
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.328028839530403e-06,
      "loss": 0.6858,
      "step": 4032
    },
    {
      "epoch": 2.84,
      "eval_loss": 0.7214952707290649,
      "eval_runtime": 30.8841,
      "eval_samples_per_second": 265.217,
      "eval_steps_per_second": 16.578,
      "step": 4032
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.3242776271762916e-06,
      "loss": 0.7025,
      "step": 4033
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.3205280161539883e-06,
      "loss": 0.6895,
      "step": 4034
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.3167800076235985e-06,
      "loss": 0.6656,
      "step": 4035
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.3130336027447287e-06,
      "loss": 0.6936,
      "step": 4036
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.3092888026764893e-06,
      "loss": 0.711,
      "step": 4037
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.305545608577495e-06,
      "loss": 0.6877,
      "step": 4038
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.3018040216058636e-06,
      "loss": 0.7037,
      "step": 4039
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.298064042919217e-06,
      "loss": 0.693,
      "step": 4040
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.294325673674676e-06,
      "loss": 0.6799,
      "step": 4041
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.2905889150288703e-06,
      "loss": 0.6953,
      "step": 4042
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.2868537681379192e-06,
      "loss": 0.669,
      "step": 4043
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.283120234157454e-06,
      "loss": 0.6771,
      "step": 4044
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.2793883142426047e-06,
      "loss": 0.709,
      "step": 4045
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.275658009547999e-06,
      "loss": 0.7108,
      "step": 4046
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.271929321227768e-06,
      "loss": 0.7074,
      "step": 4047
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.268202250435541e-06,
      "loss": 0.7275,
      "step": 4048
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.264476798324448e-06,
      "loss": 0.6929,
      "step": 4049
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.2607529660471137e-06,
      "loss": 0.6915,
      "step": 4050
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.257030754755672e-06,
      "loss": 0.7015,
      "step": 4051
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.2533101656017447e-06,
      "loss": 0.6984,
      "step": 4052
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.2495911997364593e-06,
      "loss": 0.6934,
      "step": 4053
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.2458738583104324e-06,
      "loss": 0.6917,
      "step": 4054
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.2421581424737837e-06,
      "loss": 0.7197,
      "step": 4055
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.2384440533761316e-06,
      "loss": 0.6938,
      "step": 4056
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.2347315921665885e-06,
      "loss": 0.6949,
      "step": 4057
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.231020759993763e-06,
      "loss": 0.7384,
      "step": 4058
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.227311558005761e-06,
      "loss": 0.69,
      "step": 4059
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.2236039873501813e-06,
      "loss": 0.6883,
      "step": 4060
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.219898049174122e-06,
      "loss": 0.6892,
      "step": 4061
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.2161937446241747e-06,
      "loss": 0.6949,
      "step": 4062
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.2124910748464224e-06,
      "loss": 0.6911,
      "step": 4063
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.2087900409864503e-06,
      "loss": 0.6776,
      "step": 4064
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.2050906441893245e-06,
      "loss": 0.7124,
      "step": 4065
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.201392885599616e-06,
      "loss": 0.704,
      "step": 4066
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.1976967663613818e-06,
      "loss": 0.7092,
      "step": 4067
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.1940022876181805e-06,
      "loss": 0.737,
      "step": 4068
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.190309450513056e-06,
      "loss": 0.7274,
      "step": 4069
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.1866182561885443e-06,
      "loss": 0.6834,
      "step": 4070
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.182928705786675e-06,
      "loss": 0.6769,
      "step": 4071
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.1792408004489704e-06,
      "loss": 0.7152,
      "step": 4072
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.17555454131644e-06,
      "loss": 0.6934,
      "step": 4073
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.1718699295295884e-06,
      "loss": 0.7038,
      "step": 4074
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.168186966228411e-06,
      "loss": 0.7037,
      "step": 4075
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.1645056525523825e-06,
      "loss": 0.6975,
      "step": 4076
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.1608259896404806e-06,
      "loss": 0.6982,
      "step": 4077
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.1571479786311653e-06,
      "loss": 0.6977,
      "step": 4078
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.1534716206623877e-06,
      "loss": 0.7071,
      "step": 4079
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.149796916871588e-06,
      "loss": 0.6689,
      "step": 4080
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.146123868395692e-06,
      "loss": 0.6962,
      "step": 4081
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.142452476371115e-06,
      "loss": 0.7397,
      "step": 4082
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.1387827419337614e-06,
      "loss": 0.7039,
      "step": 4083
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.1351146662190176e-06,
      "loss": 0.7042,
      "step": 4084
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.131448250361764e-06,
      "loss": 0.6814,
      "step": 4085
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.127783495496361e-06,
      "loss": 0.6723,
      "step": 4086
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.1241204027566596e-06,
      "loss": 0.7191,
      "step": 4087
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.1204589732759927e-06,
      "loss": 0.7127,
      "step": 4088
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.116799208187183e-06,
      "loss": 0.6811,
      "step": 4089
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.1131411086225336e-06,
      "loss": 0.7078,
      "step": 4090
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.1094846757138366e-06,
      "loss": 0.7082,
      "step": 4091
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.1058299105923647e-06,
      "loss": 0.6536,
      "step": 4092
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.1021768143888776e-06,
      "loss": 0.6872,
      "step": 4093
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.0985253882336166e-06,
      "loss": 0.7047,
      "step": 4094
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.094875633256307e-06,
      "loss": 0.7087,
      "step": 4095
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.0912275505861577e-06,
      "loss": 0.7093,
      "step": 4096
    },
    {
      "epoch": 2.88,
      "eval_loss": 0.7217394709587097,
      "eval_runtime": 30.9256,
      "eval_samples_per_second": 264.861,
      "eval_steps_per_second": 16.556,
      "step": 4096
    }
  ],
  "logging_steps": 1,
  "max_steps": 5680,
  "num_train_epochs": 4,
  "save_steps": 256,
  "total_flos": 1.0339248560012263e+19,
  "trial_name": null,
  "trial_params": null
}