{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.6,
  "eval_steps": 5,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 2.5e-05,
      "loss": 3.6533,
      "step": 1
    },
    {
      "epoch": 0.01,
      "learning_rate": 2.5e-05,
      "loss": 3.9424,
      "step": 2
    },
    {
      "epoch": 0.01,
      "learning_rate": 2.5e-05,
      "loss": 4.2103,
      "step": 3
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 4.2343,
      "step": 4
    },
    {
      "epoch": 0.02,
      "learning_rate": 7.500000000000001e-05,
      "loss": 3.7824,
      "step": 5
    },
    {
      "epoch": 0.02,
      "eval_loss": 3.6731483936309814,
      "eval_runtime": 55.7597,
      "eval_samples_per_second": 4.484,
      "eval_steps_per_second": 0.574,
      "step": 5
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0001,
      "loss": 3.3199,
      "step": 6
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.000125,
      "loss": 3.096,
      "step": 7
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00015000000000000001,
      "loss": 2.4714,
      "step": 8
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.000175,
      "loss": 1.8879,
      "step": 9
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.568,
      "step": 10
    },
    {
      "epoch": 0.04,
      "eval_loss": 1.3280744552612305,
      "eval_runtime": 57.8978,
      "eval_samples_per_second": 4.318,
      "eval_steps_per_second": 0.553,
      "step": 10
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019917355371900828,
      "loss": 1.3964,
      "step": 11
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019834710743801655,
      "loss": 0.9918,
      "step": 12
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019752066115702482,
      "loss": 1.0949,
      "step": 13
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0001966942148760331,
      "loss": 0.9735,
      "step": 14
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019586776859504133,
      "loss": 1.0488,
      "step": 15
    },
    {
      "epoch": 0.06,
      "eval_loss": 0.9700968861579895,
      "eval_runtime": 58.1991,
      "eval_samples_per_second": 4.296,
      "eval_steps_per_second": 0.55,
      "step": 15
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0001950413223140496,
      "loss": 0.9574,
      "step": 16
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019421487603305787,
      "loss": 1.0327,
      "step": 17
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001933884297520661,
      "loss": 0.9662,
      "step": 18
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019256198347107438,
      "loss": 0.9054,
      "step": 19
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019173553719008265,
      "loss": 0.8758,
      "step": 20
    },
    {
      "epoch": 0.08,
      "eval_loss": 0.8345032930374146,
      "eval_runtime": 58.0971,
      "eval_samples_per_second": 4.303,
      "eval_steps_per_second": 0.551,
      "step": 20
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019090909090909092,
      "loss": 0.8895,
      "step": 21
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001900826446280992,
      "loss": 0.8132,
      "step": 22
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00018925619834710743,
      "loss": 0.8083,
      "step": 23
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001884297520661157,
      "loss": 0.74,
      "step": 24
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00018760330578512397,
      "loss": 0.8129,
      "step": 25
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.7498296499252319,
      "eval_runtime": 57.9009,
      "eval_samples_per_second": 4.318,
      "eval_steps_per_second": 0.553,
      "step": 25
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00018677685950413224,
      "loss": 0.7322,
      "step": 26
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001859504132231405,
      "loss": 0.7408,
      "step": 27
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00018512396694214878,
      "loss": 0.7624,
      "step": 28
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00018429752066115705,
      "loss": 0.737,
      "step": 29
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00018347107438016532,
      "loss": 0.6782,
      "step": 30
    },
    {
      "epoch": 0.12,
      "eval_loss": 0.716300904750824,
      "eval_runtime": 58.0078,
      "eval_samples_per_second": 4.31,
      "eval_steps_per_second": 0.552,
      "step": 30
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00018264462809917356,
      "loss": 0.7201,
      "step": 31
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00018181818181818183,
      "loss": 0.6426,
      "step": 32
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00018099173553719008,
      "loss": 0.695,
      "step": 33
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00018016528925619835,
      "loss": 0.6364,
      "step": 34
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00017933884297520662,
      "loss": 0.5778,
      "step": 35
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.6839221715927124,
      "eval_runtime": 58.2032,
      "eval_samples_per_second": 4.295,
      "eval_steps_per_second": 0.55,
      "step": 35
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00017851239669421489,
      "loss": 0.74,
      "step": 36
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00017768595041322316,
      "loss": 0.5662,
      "step": 37
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00017685950413223143,
      "loss": 0.6046,
      "step": 38
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00017603305785123967,
      "loss": 0.6705,
      "step": 39
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00017520661157024794,
      "loss": 0.6315,
      "step": 40
    },
    {
      "epoch": 0.16,
      "eval_loss": 0.6694893836975098,
      "eval_runtime": 57.899,
      "eval_samples_per_second": 4.318,
      "eval_steps_per_second": 0.553,
      "step": 40
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001743801652892562,
      "loss": 0.5753,
      "step": 41
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00017355371900826448,
      "loss": 0.4392,
      "step": 42
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00017272727272727275,
      "loss": 0.5003,
      "step": 43
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00017190082644628102,
      "loss": 0.4317,
      "step": 44
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00017107438016528926,
      "loss": 0.4277,
      "step": 45
    },
    {
      "epoch": 0.18,
      "eval_loss": 0.6936589479446411,
      "eval_runtime": 58.1678,
      "eval_samples_per_second": 4.298,
      "eval_steps_per_second": 0.55,
      "step": 45
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00017024793388429753,
      "loss": 0.5912,
      "step": 46
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00016942148760330577,
      "loss": 0.4614,
      "step": 47
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00016859504132231404,
      "loss": 0.3647,
      "step": 48
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001677685950413223,
      "loss": 0.381,
      "step": 49
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00016694214876033058,
      "loss": 0.6096,
      "step": 50
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.7051330804824829,
      "eval_runtime": 57.8438,
      "eval_samples_per_second": 4.322,
      "eval_steps_per_second": 0.553,
      "step": 50
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00016611570247933885,
      "loss": 1.1622,
      "step": 51
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00016528925619834712,
      "loss": 0.9989,
      "step": 52
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001644628099173554,
      "loss": 1.1117,
      "step": 53
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00016363636363636366,
      "loss": 0.9089,
      "step": 54
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001628099173553719,
      "loss": 0.7292,
      "step": 55
    },
    {
      "epoch": 0.22,
      "eval_loss": 0.6689890027046204,
      "eval_runtime": 57.8661,
      "eval_samples_per_second": 4.32,
      "eval_steps_per_second": 0.553,
      "step": 55
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00016198347107438017,
      "loss": 0.784,
      "step": 56
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00016115702479338844,
      "loss": 0.803,
      "step": 57
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001603305785123967,
      "loss": 0.7427,
      "step": 58
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00015950413223140498,
      "loss": 0.8347,
      "step": 59
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00015867768595041322,
      "loss": 0.8003,
      "step": 60
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.6277408003807068,
      "eval_runtime": 57.8174,
      "eval_samples_per_second": 4.324,
      "eval_steps_per_second": 0.553,
      "step": 60
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001578512396694215,
      "loss": 0.7497,
      "step": 61
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00015702479338842976,
      "loss": 0.7575,
      "step": 62
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.000156198347107438,
      "loss": 0.6312,
      "step": 63
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00015537190082644627,
      "loss": 0.7552,
      "step": 64
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00015454545454545454,
      "loss": 0.6101,
      "step": 65
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.6085646748542786,
      "eval_runtime": 58.0155,
      "eval_samples_per_second": 4.309,
      "eval_steps_per_second": 0.552,
      "step": 65
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00015371900826446281,
      "loss": 0.6897,
      "step": 66
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00015289256198347108,
      "loss": 0.5665,
      "step": 67
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00015206611570247935,
      "loss": 0.7022,
      "step": 68
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00015123966942148762,
      "loss": 0.7871,
      "step": 69
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001504132231404959,
      "loss": 0.6128,
      "step": 70
    },
    {
      "epoch": 0.28,
      "eval_loss": 0.6062941551208496,
      "eval_runtime": 57.9481,
      "eval_samples_per_second": 4.314,
      "eval_steps_per_second": 0.552,
      "step": 70
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00014958677685950414,
      "loss": 0.6889,
      "step": 71
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001487603305785124,
      "loss": 0.5652,
      "step": 72
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00014793388429752067,
      "loss": 0.6203,
      "step": 73
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00014710743801652894,
      "loss": 0.6553,
      "step": 74
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001462809917355372,
      "loss": 0.57,
      "step": 75
    },
    {
      "epoch": 0.3,
      "eval_loss": 0.61387038230896,
      "eval_runtime": 57.9724,
      "eval_samples_per_second": 4.312,
      "eval_steps_per_second": 0.552,
      "step": 75
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00014545454545454546,
      "loss": 0.5595,
      "step": 76
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00014462809917355373,
      "loss": 0.6794,
      "step": 77
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.000143801652892562,
      "loss": 0.5683,
      "step": 78
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00014297520661157024,
      "loss": 0.5436,
      "step": 79
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001421487603305785,
      "loss": 0.6996,
      "step": 80
    },
    {
      "epoch": 0.32,
      "eval_loss": 0.6077276468276978,
      "eval_runtime": 58.1658,
      "eval_samples_per_second": 4.298,
      "eval_steps_per_second": 0.55,
      "step": 80
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00014132231404958678,
      "loss": 0.4647,
      "step": 81
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00014049586776859505,
      "loss": 0.5812,
      "step": 82
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00013966942148760332,
      "loss": 0.625,
      "step": 83
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001388429752066116,
      "loss": 0.717,
      "step": 84
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00013801652892561986,
      "loss": 0.5178,
      "step": 85
    },
    {
      "epoch": 0.34,
      "eval_loss": 0.6118555665016174,
      "eval_runtime": 57.8518,
      "eval_samples_per_second": 4.321,
      "eval_steps_per_second": 0.553,
      "step": 85
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00013719008264462813,
      "loss": 0.4657,
      "step": 86
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00013636363636363637,
      "loss": 0.5238,
      "step": 87
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00013553719008264464,
      "loss": 0.4704,
      "step": 88
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00013471074380165288,
      "loss": 0.4564,
      "step": 89
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00013388429752066115,
      "loss": 0.3964,
      "step": 90
    },
    {
      "epoch": 0.36,
      "eval_loss": 0.6064205169677734,
      "eval_runtime": 57.9504,
      "eval_samples_per_second": 4.314,
      "eval_steps_per_second": 0.552,
      "step": 90
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00013305785123966942,
      "loss": 0.4646,
      "step": 91
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001322314049586777,
      "loss": 0.4212,
      "step": 92
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00013140495867768596,
      "loss": 0.3841,
      "step": 93
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00013057851239669423,
      "loss": 0.6348,
      "step": 94
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00012975206611570247,
      "loss": 0.563,
      "step": 95
    },
    {
      "epoch": 0.38,
      "eval_loss": 0.6179620623588562,
      "eval_runtime": 57.8595,
      "eval_samples_per_second": 4.321,
      "eval_steps_per_second": 0.553,
      "step": 95
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00012892561983471074,
      "loss": 0.6287,
      "step": 96
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.000128099173553719,
      "loss": 0.4541,
      "step": 97
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00012727272727272728,
      "loss": 0.5001,
      "step": 98
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00012727272727272728,
      "loss": 0.418,
      "step": 99
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00012644628099173555,
      "loss": 0.3845,
      "step": 100
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.6233624815940857,
      "eval_runtime": 58.1243,
      "eval_samples_per_second": 4.301,
      "eval_steps_per_second": 0.551,
      "step": 100
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00012561983471074382,
      "loss": 0.8978,
      "step": 101
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001247933884297521,
      "loss": 0.8395,
      "step": 102
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00012396694214876033,
      "loss": 0.9166,
      "step": 103
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001231404958677686,
      "loss": 0.719,
      "step": 104
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00012231404958677685,
      "loss": 0.6933,
      "step": 105
    },
    {
      "epoch": 0.42,
      "eval_loss": 0.6230288147926331,
      "eval_runtime": 57.9123,
      "eval_samples_per_second": 4.317,
      "eval_steps_per_second": 0.553,
      "step": 105
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00012148760330578513,
      "loss": 0.6997,
      "step": 106
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001206611570247934,
      "loss": 0.6827,
      "step": 107
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00011983471074380165,
      "loss": 0.9173,
      "step": 108
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00011900826446280992,
      "loss": 0.6493,
      "step": 109
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001181818181818182,
      "loss": 0.537,
      "step": 110
    },
    {
      "epoch": 0.44,
      "eval_loss": 0.6039621829986572,
      "eval_runtime": 58.0963,
      "eval_samples_per_second": 4.303,
      "eval_steps_per_second": 0.551,
      "step": 110
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00011735537190082646,
      "loss": 0.6706,
      "step": 111
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001165289256198347,
      "loss": 0.5669,
      "step": 112
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00011570247933884298,
      "loss": 0.6952,
      "step": 113
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00011487603305785125,
      "loss": 0.6874,
      "step": 114
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001140495867768595,
      "loss": 0.6718,
      "step": 115
    },
    {
      "epoch": 0.46,
      "eval_loss": 0.5879318118095398,
      "eval_runtime": 57.8905,
      "eval_samples_per_second": 4.319,
      "eval_steps_per_second": 0.553,
      "step": 115
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00011322314049586777,
      "loss": 0.7431,
      "step": 116
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00011239669421487604,
      "loss": 0.8516,
      "step": 117
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00011157024793388431,
      "loss": 0.5923,
      "step": 118
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00011074380165289258,
      "loss": 0.5841,
      "step": 119
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00010991735537190082,
      "loss": 0.5717,
      "step": 120
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.5779567956924438,
      "eval_runtime": 57.8881,
      "eval_samples_per_second": 4.319,
      "eval_steps_per_second": 0.553,
      "step": 120
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00010909090909090909,
      "loss": 0.4524,
      "step": 121
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00010826446280991735,
      "loss": 0.5259,
      "step": 122
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00010743801652892562,
      "loss": 0.6316,
      "step": 123
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00010661157024793389,
      "loss": 0.4857,
      "step": 124
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00010578512396694216,
      "loss": 0.5092,
      "step": 125
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.5742751955986023,
      "eval_runtime": 57.8734,
      "eval_samples_per_second": 4.32,
      "eval_steps_per_second": 0.553,
      "step": 125
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00010495867768595043,
      "loss": 0.5744,
      "step": 126
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001041322314049587,
      "loss": 0.7194,
      "step": 127
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00010330578512396694,
      "loss": 0.5746,
      "step": 128
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00010247933884297521,
      "loss": 0.5654,
      "step": 129
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00010165289256198347,
      "loss": 0.6142,
      "step": 130
    },
    {
      "epoch": 0.52,
      "eval_loss": 0.5754650831222534,
      "eval_runtime": 57.8524,
      "eval_samples_per_second": 4.321,
      "eval_steps_per_second": 0.553,
      "step": 130
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00010082644628099174,
      "loss": 0.4129,
      "step": 131
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001,
      "loss": 0.5766,
      "step": 132
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.917355371900827e-05,
      "loss": 0.5391,
      "step": 133
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.834710743801654e-05,
      "loss": 0.3944,
      "step": 134
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.75206611570248e-05,
      "loss": 0.5665,
      "step": 135
    },
    {
      "epoch": 0.54,
      "eval_loss": 0.5827150940895081,
      "eval_runtime": 57.9089,
      "eval_samples_per_second": 4.317,
      "eval_steps_per_second": 0.553,
      "step": 135
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.669421487603306e-05,
      "loss": 0.4864,
      "step": 136
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.586776859504133e-05,
      "loss": 0.5534,
      "step": 137
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.50413223140496e-05,
      "loss": 0.5339,
      "step": 138
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.421487603305785e-05,
      "loss": 0.4784,
      "step": 139
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.338842975206612e-05,
      "loss": 0.5378,
      "step": 140
    },
    {
      "epoch": 0.56,
      "eval_loss": 0.5869958996772766,
      "eval_runtime": 58.1247,
      "eval_samples_per_second": 4.301,
      "eval_steps_per_second": 0.551,
      "step": 140
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.256198347107439e-05,
      "loss": 0.5092,
      "step": 141
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.173553719008266e-05,
      "loss": 0.3735,
      "step": 142
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.090909090909092e-05,
      "loss": 0.4498,
      "step": 143
    },
    {
      "epoch": 0.58,
      "learning_rate": 9.008264462809917e-05,
      "loss": 0.3738,
      "step": 144
    },
    {
      "epoch": 0.58,
      "learning_rate": 8.925619834710744e-05,
      "loss": 0.4061,
      "step": 145
    },
    {
      "epoch": 0.58,
      "eval_loss": 0.5876795649528503,
      "eval_runtime": 57.9113,
      "eval_samples_per_second": 4.317,
      "eval_steps_per_second": 0.553,
      "step": 145
    },
    {
      "epoch": 0.58,
      "learning_rate": 8.842975206611571e-05,
      "loss": 0.3713,
      "step": 146
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.760330578512397e-05,
      "loss": 0.3532,
      "step": 147
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.677685950413224e-05,
      "loss": 0.3981,
      "step": 148
    },
    {
      "epoch": 0.6,
      "learning_rate": 8.595041322314051e-05,
      "loss": 0.4389,
      "step": 149
    },
    {
      "epoch": 0.6,
      "learning_rate": 8.512396694214876e-05,
      "loss": 0.3934,
      "step": 150
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.6049365401268005,
      "eval_runtime": 57.8527,
      "eval_samples_per_second": 4.321,
      "eval_steps_per_second": 0.553,
      "step": 150
    },
    {
      "epoch": 0.6,
      "learning_rate": 8.429752066115702e-05,
      "loss": 0.766,
      "step": 151
    },
    {
      "epoch": 0.61,
      "learning_rate": 8.347107438016529e-05,
      "loss": 0.6959,
      "step": 152
    },
    {
      "epoch": 0.61,
      "learning_rate": 8.264462809917356e-05,
      "loss": 0.7616,
      "step": 153
    },
    {
      "epoch": 0.62,
      "learning_rate": 8.181818181818183e-05,
      "loss": 0.8095,
      "step": 154
    },
    {
      "epoch": 0.62,
      "learning_rate": 8.099173553719009e-05,
      "loss": 0.6343,
      "step": 155
    },
    {
      "epoch": 0.62,
      "eval_loss": 0.5834559202194214,
      "eval_runtime": 58.0814,
      "eval_samples_per_second": 4.304,
      "eval_steps_per_second": 0.551,
      "step": 155
    },
    {
      "epoch": 0.62,
      "learning_rate": 8.016528925619836e-05,
      "loss": 0.7482,
      "step": 156
    },
    {
      "epoch": 0.63,
      "learning_rate": 7.933884297520661e-05,
      "loss": 0.6632,
      "step": 157
    },
    {
      "epoch": 0.63,
      "learning_rate": 7.851239669421488e-05,
      "loss": 0.5817,
      "step": 158
    },
    {
      "epoch": 0.64,
      "learning_rate": 7.768595041322314e-05,
      "loss": 0.6256,
      "step": 159
    },
    {
      "epoch": 0.64,
      "learning_rate": 7.685950413223141e-05,
      "loss": 0.6727,
      "step": 160
    },
    {
      "epoch": 0.64,
      "eval_loss": 0.5801185369491577,
      "eval_runtime": 57.9954,
      "eval_samples_per_second": 4.311,
      "eval_steps_per_second": 0.552,
      "step": 160
    },
    {
      "epoch": 0.64,
      "learning_rate": 7.603305785123968e-05,
      "loss": 0.7253,
      "step": 161
    },
    {
      "epoch": 0.65,
      "learning_rate": 7.520661157024795e-05,
      "loss": 0.6021,
      "step": 162
    },
    {
      "epoch": 0.65,
      "learning_rate": 7.43801652892562e-05,
      "loss": 0.5942,
      "step": 163
    },
    {
      "epoch": 0.66,
      "learning_rate": 7.355371900826447e-05,
      "loss": 0.5942,
      "step": 164
    },
    {
      "epoch": 0.66,
      "learning_rate": 7.272727272727273e-05,
      "loss": 0.602,
      "step": 165
    },
    {
      "epoch": 0.66,
      "eval_loss": 0.5663142800331116,
      "eval_runtime": 57.8541,
      "eval_samples_per_second": 4.321,
      "eval_steps_per_second": 0.553,
      "step": 165
    },
    {
      "epoch": 0.66,
      "learning_rate": 7.1900826446281e-05,
      "loss": 0.6503,
      "step": 166
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.107438016528925e-05,
      "loss": 0.6197,
      "step": 167
    },
    {
      "epoch": 0.67,
      "learning_rate": 7.024793388429752e-05,
      "loss": 0.4577,
      "step": 168
    },
    {
      "epoch": 0.68,
      "learning_rate": 6.94214876033058e-05,
      "loss": 0.6804,
      "step": 169
    },
    {
      "epoch": 0.68,
      "learning_rate": 6.859504132231406e-05,
      "loss": 0.7056,
      "step": 170
    },
    {
      "epoch": 0.68,
      "eval_loss": 0.5660744309425354,
      "eval_runtime": 57.981,
      "eval_samples_per_second": 4.312,
      "eval_steps_per_second": 0.552,
      "step": 170
    },
    {
      "epoch": 0.68,
      "learning_rate": 6.776859504132232e-05,
      "loss": 0.6003,
      "step": 171
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.694214876033058e-05,
      "loss": 0.6349,
      "step": 172
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.611570247933885e-05,
      "loss": 0.6308,
      "step": 173
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.528925619834711e-05,
      "loss": 0.5713,
      "step": 174
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.446280991735537e-05,
      "loss": 0.4743,
      "step": 175
    },
    {
      "epoch": 0.7,
      "eval_loss": 0.5626071691513062,
      "eval_runtime": 58.0516,
      "eval_samples_per_second": 4.307,
      "eval_steps_per_second": 0.551,
      "step": 175
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.363636363636364e-05,
      "loss": 0.4835,
      "step": 176
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.280991735537191e-05,
      "loss": 0.4553,
      "step": 177
    },
    {
      "epoch": 0.71,
      "learning_rate": 6.198347107438017e-05,
      "loss": 0.6067,
      "step": 178
    },
    {
      "epoch": 0.72,
      "learning_rate": 6.115702479338842e-05,
      "loss": 0.5319,
      "step": 179
    },
    {
      "epoch": 0.72,
      "learning_rate": 6.03305785123967e-05,
      "loss": 0.5656,
      "step": 180
    },
    {
      "epoch": 0.72,
      "eval_loss": 0.5597254633903503,
      "eval_runtime": 58.0733,
      "eval_samples_per_second": 4.305,
      "eval_steps_per_second": 0.551,
      "step": 180
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.950413223140496e-05,
      "loss": 0.3764,
      "step": 181
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.867768595041323e-05,
      "loss": 0.3883,
      "step": 182
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.785123966942149e-05,
      "loss": 0.4815,
      "step": 183
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.702479338842975e-05,
      "loss": 0.4923,
      "step": 184
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.619834710743802e-05,
      "loss": 0.5248,
      "step": 185
    },
    {
      "epoch": 0.74,
      "eval_loss": 0.5612524151802063,
      "eval_runtime": 58.039,
      "eval_samples_per_second": 4.307,
      "eval_steps_per_second": 0.551,
      "step": 185
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.537190082644629e-05,
      "loss": 0.4291,
      "step": 186
    },
    {
      "epoch": 0.75,
      "learning_rate": 5.4545454545454546e-05,
      "loss": 0.4774,
      "step": 187
    },
    {
      "epoch": 0.75,
      "learning_rate": 5.371900826446281e-05,
      "loss": 0.4568,
      "step": 188
    },
    {
      "epoch": 0.76,
      "learning_rate": 5.289256198347108e-05,
      "loss": 0.4394,
      "step": 189
    },
    {
      "epoch": 0.76,
      "learning_rate": 5.206611570247935e-05,
      "loss": 0.4516,
      "step": 190
    },
    {
      "epoch": 0.76,
      "eval_loss": 0.5654726624488831,
      "eval_runtime": 58.1592,
      "eval_samples_per_second": 4.299,
      "eval_steps_per_second": 0.55,
      "step": 190
    },
    {
      "epoch": 0.76,
      "learning_rate": 5.1239669421487605e-05,
      "loss": 0.5396,
      "step": 191
    },
    {
      "epoch": 0.77,
      "learning_rate": 5.041322314049587e-05,
      "loss": 0.4024,
      "step": 192
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.958677685950414e-05,
      "loss": 0.4171,
      "step": 193
    },
    {
      "epoch": 0.78,
      "learning_rate": 4.87603305785124e-05,
      "loss": 0.5578,
      "step": 194
    },
    {
      "epoch": 0.78,
      "learning_rate": 4.793388429752066e-05,
      "loss": 0.4754,
      "step": 195
    },
    {
      "epoch": 0.78,
      "eval_loss": 0.5674455761909485,
      "eval_runtime": 57.9022,
      "eval_samples_per_second": 4.318,
      "eval_steps_per_second": 0.553,
      "step": 195
    },
    {
      "epoch": 0.78,
      "learning_rate": 4.7107438016528926e-05,
      "loss": 0.4136,
      "step": 196
    },
    {
      "epoch": 0.79,
      "learning_rate": 4.6280991735537196e-05,
      "loss": 0.4266,
      "step": 197
    },
    {
      "epoch": 0.79,
      "learning_rate": 4.545454545454546e-05,
      "loss": 0.3173,
      "step": 198
    },
    {
      "epoch": 0.8,
      "learning_rate": 4.462809917355372e-05,
      "loss": 0.3905,
      "step": 199
    },
    {
      "epoch": 0.8,
      "learning_rate": 4.3801652892561984e-05,
      "loss": 0.6875,
      "step": 200
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.567625105381012,
      "eval_runtime": 57.8625,
      "eval_samples_per_second": 4.321,
      "eval_steps_per_second": 0.553,
      "step": 200
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00016536082474226803,
      "loss": 0.6259,
      "step": 201
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00016515463917525774,
      "loss": 0.8477,
      "step": 202
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00016494845360824742,
      "loss": 0.9355,
      "step": 203
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001647422680412371,
      "loss": 0.7632,
      "step": 204
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00016453608247422681,
      "loss": 0.5125,
      "step": 205
    },
    {
      "epoch": 0.82,
      "eval_loss": 0.5757731795310974,
      "eval_runtime": 58.6641,
      "eval_samples_per_second": 4.262,
      "eval_steps_per_second": 0.545,
      "step": 205
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001643298969072165,
      "loss": 0.7234,
      "step": 206
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00016412371134020618,
      "loss": 0.7542,
      "step": 207
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0001639175257731959,
      "loss": 0.6394,
      "step": 208
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00016371134020618558,
      "loss": 0.5565,
      "step": 209
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00016350515463917526,
      "loss": 0.6834,
      "step": 210
    },
    {
      "epoch": 0.84,
      "eval_loss": 0.5658813714981079,
      "eval_runtime": 58.24,
      "eval_samples_per_second": 4.293,
      "eval_steps_per_second": 0.549,
      "step": 210
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00016329896907216494,
      "loss": 0.662,
      "step": 211
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00016309278350515465,
      "loss": 0.5459,
      "step": 212
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00016288659793814434,
      "loss": 0.6164,
      "step": 213
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00016268041237113402,
      "loss": 0.6335,
      "step": 214
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00016247422680412373,
      "loss": 0.5906,
      "step": 215
    },
    {
      "epoch": 0.86,
      "eval_loss": 0.5624733567237854,
      "eval_runtime": 57.9447,
      "eval_samples_per_second": 4.314,
      "eval_steps_per_second": 0.552,
      "step": 215
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00016226804123711341,
      "loss": 0.5621,
      "step": 216
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001620618556701031,
      "loss": 0.5567,
      "step": 217
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00016185567010309278,
      "loss": 0.4252,
      "step": 218
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001616494845360825,
      "loss": 0.6901,
      "step": 219
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00016144329896907217,
      "loss": 0.6321,
      "step": 220
    },
    {
      "epoch": 0.88,
      "eval_loss": 0.5694482326507568,
      "eval_runtime": 58.2135,
      "eval_samples_per_second": 4.295,
      "eval_steps_per_second": 0.55,
      "step": 220
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00016123711340206186,
      "loss": 0.6163,
      "step": 221
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00016103092783505157,
      "loss": 0.4961,
      "step": 222
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00016082474226804125,
      "loss": 0.5632,
      "step": 223
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00016061855670103094,
      "loss": 0.4749,
      "step": 224
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00016041237113402065,
      "loss": 0.495,
      "step": 225
    },
    {
      "epoch": 0.9,
      "eval_loss": 0.5669766664505005,
      "eval_runtime": 57.9014,
      "eval_samples_per_second": 4.318,
      "eval_steps_per_second": 0.553,
      "step": 225
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00016020618556701033,
      "loss": 0.4442,
      "step": 226
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00016,
      "loss": 0.526,
      "step": 227
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001597938144329897,
      "loss": 0.5636,
      "step": 228
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0001595876288659794,
      "loss": 0.511,
      "step": 229
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0001593814432989691,
      "loss": 0.5395,
      "step": 230
    },
    {
      "epoch": 0.92,
      "eval_loss": 0.5704668164253235,
      "eval_runtime": 57.9046,
      "eval_samples_per_second": 4.317,
      "eval_steps_per_second": 0.553,
      "step": 230
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00015917525773195875,
      "loss": 0.5187,
      "step": 231
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00015896907216494846,
      "loss": 0.5611,
      "step": 232
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00015876288659793814,
      "loss": 0.4718,
      "step": 233
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00015855670103092782,
      "loss": 0.4049,
      "step": 234
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00015835051546391754,
      "loss": 0.5668,
      "step": 235
    },
    {
      "epoch": 0.94,
      "eval_loss": 0.5722946524620056,
      "eval_runtime": 58.2566,
      "eval_samples_per_second": 4.291,
      "eval_steps_per_second": 0.549,
      "step": 235
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00015814432989690722,
      "loss": 0.5245,
      "step": 236
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0001579381443298969,
      "loss": 0.4994,
      "step": 237
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0001577319587628866,
      "loss": 0.4223,
      "step": 238
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001575257731958763,
      "loss": 0.5222,
      "step": 239
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00015731958762886598,
      "loss": 0.4199,
      "step": 240
    },
    {
      "epoch": 0.96,
      "eval_loss": 0.5722372531890869,
      "eval_runtime": 57.804,
      "eval_samples_per_second": 4.325,
      "eval_steps_per_second": 0.554,
      "step": 240
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00015711340206185566,
      "loss": 0.4171,
      "step": 241
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00015690721649484537,
      "loss": 0.3978,
      "step": 242
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00015690721649484537,
      "loss": 0.5726,
      "step": 243
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00015670103092783506,
      "loss": 0.3324,
      "step": 244
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00015649484536082474,
      "loss": 0.4253,
      "step": 245
    },
    {
      "epoch": 0.98,
      "eval_loss": 0.5949168801307678,
      "eval_runtime": 58.1262,
      "eval_samples_per_second": 4.301,
      "eval_steps_per_second": 0.551,
      "step": 245
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00015628865979381445,
      "loss": 0.3992,
      "step": 246
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00015608247422680413,
      "loss": 0.4968,
      "step": 247
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00015587628865979382,
      "loss": 0.388,
      "step": 248
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001556701030927835,
      "loss": 0.3071,
      "step": 249
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001554639175257732,
      "loss": 0.4561,
      "step": 250
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6334843039512634,
      "eval_runtime": 58.0783,
      "eval_samples_per_second": 4.305,
      "eval_steps_per_second": 0.551,
      "step": 250
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001552577319587629,
      "loss": 0.8641,
      "step": 251
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00015505154639175258,
      "loss": 0.5815,
      "step": 252
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0001548453608247423,
      "loss": 0.6515,
      "step": 253
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00015463917525773197,
      "loss": 0.6192,
      "step": 254
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00015443298969072166,
      "loss": 0.4681,
      "step": 255
    },
    {
      "epoch": 1.02,
      "eval_loss": 0.5661035776138306,
      "eval_runtime": 58.219,
      "eval_samples_per_second": 4.294,
      "eval_steps_per_second": 0.55,
      "step": 255
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00015422680412371137,
      "loss": 0.5288,
      "step": 256
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00015402061855670105,
      "loss": 0.4924,
      "step": 257
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00015381443298969073,
      "loss": 0.5358,
      "step": 258
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00015360824742268042,
      "loss": 0.5082,
      "step": 259
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00015340206185567013,
      "loss": 0.4527,
      "step": 260
    },
    {
      "epoch": 1.04,
      "eval_loss": 0.5820726156234741,
      "eval_runtime": 57.7612,
      "eval_samples_per_second": 4.328,
      "eval_steps_per_second": 0.554,
      "step": 260
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0001531958762886598,
      "loss": 0.5516,
      "step": 261
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00015298969072164947,
      "loss": 0.5131,
      "step": 262
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00015278350515463918,
      "loss": 0.5988,
      "step": 263
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00015257731958762886,
      "loss": 0.4569,
      "step": 264
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00015237113402061855,
      "loss": 0.53,
      "step": 265
    },
    {
      "epoch": 1.06,
      "eval_loss": 0.5718860030174255,
      "eval_runtime": 58.2455,
      "eval_samples_per_second": 4.292,
      "eval_steps_per_second": 0.549,
      "step": 265
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00015216494845360826,
      "loss": 0.506,
      "step": 266
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00015195876288659794,
      "loss": 0.5774,
      "step": 267
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00015175257731958762,
      "loss": 0.3699,
      "step": 268
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00015154639175257733,
      "loss": 0.3874,
      "step": 269
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00015134020618556702,
      "loss": 0.552,
      "step": 270
    },
    {
      "epoch": 1.08,
      "eval_loss": 0.5731833577156067,
      "eval_runtime": 58.0655,
      "eval_samples_per_second": 4.305,
      "eval_steps_per_second": 0.551,
      "step": 270
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0001511340206185567,
      "loss": 0.4844,
      "step": 271
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00015092783505154638,
      "loss": 0.4164,
      "step": 272
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001507216494845361,
      "loss": 0.4694,
      "step": 273
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00015051546391752578,
      "loss": 0.4008,
      "step": 274
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00015030927835051546,
      "loss": 0.4439,
      "step": 275
    },
    {
      "epoch": 1.1,
      "eval_loss": 0.5790666341781616,
      "eval_runtime": 57.8801,
      "eval_samples_per_second": 4.319,
      "eval_steps_per_second": 0.553,
      "step": 275
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00015010309278350517,
      "loss": 0.4399,
      "step": 276
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00014989690721649486,
      "loss": 0.4316,
      "step": 277
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00014969072164948454,
      "loss": 0.4761,
      "step": 278
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00014948453608247422,
      "loss": 0.44,
      "step": 279
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00014927835051546393,
      "loss": 0.3519,
      "step": 280
    },
    {
      "epoch": 1.12,
      "eval_loss": 0.5812113881111145,
      "eval_runtime": 58.0042,
      "eval_samples_per_second": 4.31,
      "eval_steps_per_second": 0.552,
      "step": 280
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00014907216494845362,
      "loss": 0.3776,
      "step": 281
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001488659793814433,
      "loss": 0.3389,
      "step": 282
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.000148659793814433,
      "loss": 0.3219,
      "step": 283
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001484536082474227,
      "loss": 0.4546,
      "step": 284
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00014824742268041238,
      "loss": 0.4285,
      "step": 285
    },
    {
      "epoch": 1.14,
      "eval_loss": 0.5796990990638733,
      "eval_runtime": 57.8331,
      "eval_samples_per_second": 4.323,
      "eval_steps_per_second": 0.553,
      "step": 285
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001480412371134021,
      "loss": 0.3603,
      "step": 286
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00014783505154639177,
      "loss": 0.3516,
      "step": 287
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00014762886597938146,
      "loss": 0.3092,
      "step": 288
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00014742268041237114,
      "loss": 0.3408,
      "step": 289
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00014721649484536085,
      "loss": 0.2536,
      "step": 290
    },
    {
      "epoch": 1.16,
      "eval_loss": 0.5906583070755005,
      "eval_runtime": 58.0348,
      "eval_samples_per_second": 4.308,
      "eval_steps_per_second": 0.551,
      "step": 290
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00014701030927835053,
      "loss": 0.3943,
      "step": 291
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001468041237113402,
      "loss": 0.3739,
      "step": 292
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001465979381443299,
      "loss": 0.405,
      "step": 293
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00014639175257731958,
      "loss": 0.3548,
      "step": 294
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00014618556701030927,
      "loss": 0.462,
      "step": 295
    },
    {
      "epoch": 1.18,
      "eval_loss": 0.6061128973960876,
      "eval_runtime": 57.9993,
      "eval_samples_per_second": 4.31,
      "eval_steps_per_second": 0.552,
      "step": 295
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00014597938144329898,
      "loss": 0.3555,
      "step": 296
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00014577319587628866,
      "loss": 0.414,
      "step": 297
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00014556701030927834,
      "loss": 0.2916,
      "step": 298
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00014536082474226805,
      "loss": 0.2507,
      "step": 299
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00014515463917525774,
      "loss": 0.3295,
      "step": 300
    },
    {
      "epoch": 1.2,
      "eval_loss": 0.6123625636100769,
      "eval_runtime": 57.7714,
      "eval_samples_per_second": 4.327,
      "eval_steps_per_second": 0.554,
      "step": 300
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00014494845360824742,
      "loss": 0.5293,
      "step": 301
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0001447422680412371,
      "loss": 0.686,
      "step": 302
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00014453608247422682,
      "loss": 0.4869,
      "step": 303
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001443298969072165,
      "loss": 0.5161,
      "step": 304
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00014412371134020618,
      "loss": 0.4381,
      "step": 305
    },
    {
      "epoch": 1.22,
      "eval_loss": 0.576439619064331,
      "eval_runtime": 58.162,
      "eval_samples_per_second": 4.298,
      "eval_steps_per_second": 0.55,
      "step": 305
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001439175257731959,
      "loss": 0.5365,
      "step": 306
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00014371134020618558,
      "loss": 0.4769,
      "step": 307
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00014350515463917526,
      "loss": 0.555,
      "step": 308
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00014329896907216494,
      "loss": 0.5363,
      "step": 309
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00014309278350515465,
      "loss": 0.5696,
      "step": 310
    },
    {
      "epoch": 1.24,
      "eval_loss": 0.5902033448219299,
      "eval_runtime": 58.1218,
      "eval_samples_per_second": 4.301,
      "eval_steps_per_second": 0.551,
      "step": 310
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00014288659793814434,
      "loss": 0.5701,
      "step": 311
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00014268041237113402,
      "loss": 0.4742,
      "step": 312
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00014247422680412373,
      "loss": 0.5453,
      "step": 313
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00014226804123711342,
      "loss": 0.4795,
      "step": 314
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001420618556701031,
      "loss": 0.4853,
      "step": 315
    },
    {
      "epoch": 1.26,
      "eval_loss": 0.575286865234375,
      "eval_runtime": 57.7708,
      "eval_samples_per_second": 4.327,
      "eval_steps_per_second": 0.554,
      "step": 315
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001418556701030928,
      "loss": 0.4996,
      "step": 316
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001416494845360825,
      "loss": 0.3985,
      "step": 317
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00014144329896907218,
      "loss": 0.4474,
      "step": 318
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00014123711340206186,
      "loss": 0.5442,
      "step": 319
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00014103092783505157,
      "loss": 0.5395,
      "step": 320
    },
    {
      "epoch": 1.28,
      "eval_loss": 0.5742917656898499,
      "eval_runtime": 57.9054,
      "eval_samples_per_second": 4.317,
      "eval_steps_per_second": 0.553,
      "step": 320
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00014082474226804125,
      "loss": 0.4698,
      "step": 321
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0001406185567010309,
      "loss": 0.5232,
      "step": 322
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00014041237113402062,
      "loss": 0.41,
      "step": 323
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001402061855670103,
      "loss": 0.4899,
      "step": 324
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00014,
      "loss": 0.4355,
      "step": 325
    },
    {
      "epoch": 1.3,
      "eval_loss": 0.5757585763931274,
      "eval_runtime": 58.18,
      "eval_samples_per_second": 4.297,
      "eval_steps_per_second": 0.55,
      "step": 325
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001397938144329897,
      "loss": 0.4128,
      "step": 326
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00013958762886597938,
      "loss": 0.4048,
      "step": 327
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00013938144329896907,
      "loss": 0.4656,
      "step": 328
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00013917525773195878,
      "loss": 0.3682,
      "step": 329
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00013896907216494846,
      "loss": 0.4943,
      "step": 330
    },
    {
      "epoch": 1.32,
      "eval_loss": 0.5842174291610718,
      "eval_runtime": 57.8497,
      "eval_samples_per_second": 4.322,
      "eval_steps_per_second": 0.553,
      "step": 330
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00013876288659793814,
      "loss": 0.3817,
      "step": 331
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00013855670103092783,
      "loss": 0.4528,
      "step": 332
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00013835051546391754,
      "loss": 0.3355,
      "step": 333
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00013814432989690722,
      "loss": 0.4085,
      "step": 334
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001379381443298969,
      "loss": 0.3934,
      "step": 335
    },
    {
      "epoch": 1.34,
      "eval_loss": 0.5792785882949829,
      "eval_runtime": 58.0391,
      "eval_samples_per_second": 4.307,
      "eval_steps_per_second": 0.551,
      "step": 335
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00013773195876288661,
      "loss": 0.3289,
      "step": 336
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0001375257731958763,
      "loss": 0.3551,
      "step": 337
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00013731958762886598,
      "loss": 0.3062,
      "step": 338
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00013711340206185566,
      "loss": 0.3588,
      "step": 339
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00013690721649484538,
      "loss": 0.2947,
      "step": 340
    },
    {
      "epoch": 1.36,
      "eval_loss": 0.5880213975906372,
      "eval_runtime": 58.0627,
      "eval_samples_per_second": 4.306,
      "eval_steps_per_second": 0.551,
      "step": 340
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00013670103092783506,
      "loss": 0.4527,
      "step": 341
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00013649484536082474,
      "loss": 0.413,
      "step": 342
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00013628865979381445,
      "loss": 0.3439,
      "step": 343
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00013608247422680414,
      "loss": 0.3529,
      "step": 344
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00013587628865979382,
      "loss": 0.3427,
      "step": 345
    },
    {
      "epoch": 1.38,
      "eval_loss": 0.5902564525604248,
      "eval_runtime": 58.0661,
      "eval_samples_per_second": 4.305,
      "eval_steps_per_second": 0.551,
      "step": 345
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00013567010309278353,
      "loss": 0.3917,
      "step": 346
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00013546391752577321,
      "loss": 0.2853,
      "step": 347
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001352577319587629,
      "loss": 0.2518,
      "step": 348
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00013505154639175258,
      "loss": 0.3215,
      "step": 349
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001348453608247423,
      "loss": 0.2906,
      "step": 350
    },
    {
      "epoch": 1.4,
      "eval_loss": 0.6013336181640625,
      "eval_runtime": 58.0155,
      "eval_samples_per_second": 4.309,
      "eval_steps_per_second": 0.552,
      "step": 350
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00013463917525773197,
      "loss": 0.6485,
      "step": 351
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00013443298969072166,
      "loss": 0.6829,
      "step": 352
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00013422680412371134,
      "loss": 0.5409,
      "step": 353
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00013402061855670103,
      "loss": 0.7209,
      "step": 354
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0001338144329896907,
      "loss": 0.5774,
      "step": 355
    },
    {
      "epoch": 1.42,
      "eval_loss": 0.583857536315918,
      "eval_runtime": 57.8931,
      "eval_samples_per_second": 4.318,
      "eval_steps_per_second": 0.553,
      "step": 355
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00013360824742268042,
      "loss": 0.5123,
      "step": 356
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0001334020618556701,
      "loss": 0.6287,
      "step": 357
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00013319587628865979,
      "loss": 0.641,
      "step": 358
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001329896907216495,
      "loss": 0.5998,
      "step": 359
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00013278350515463918,
      "loss": 0.4529,
      "step": 360
    },
    {
      "epoch": 1.44,
      "eval_loss": 0.581903338432312,
      "eval_runtime": 57.8521,
      "eval_samples_per_second": 4.321,
      "eval_steps_per_second": 0.553,
      "step": 360
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00013257731958762886,
      "loss": 0.4551,
      "step": 361
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00013237113402061855,
      "loss": 0.4757,
      "step": 362
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00013216494845360826,
      "loss": 0.4402,
      "step": 363
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00013195876288659794,
      "loss": 0.4174,
      "step": 364
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00013175257731958762,
      "loss": 0.4431,
      "step": 365
    },
    {
      "epoch": 1.46,
      "eval_loss": 0.5788648128509521,
      "eval_runtime": 58.1681,
      "eval_samples_per_second": 4.298,
      "eval_steps_per_second": 0.55,
      "step": 365
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00013154639175257734,
      "loss": 0.4286,
      "step": 366
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00013134020618556702,
      "loss": 0.4597,
      "step": 367
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001311340206185567,
      "loss": 0.5148,
      "step": 368
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00013092783505154639,
      "loss": 0.5256,
      "step": 369
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001307216494845361,
      "loss": 0.3836,
      "step": 370
    },
    {
      "epoch": 1.48,
      "eval_loss": 0.5807417035102844,
      "eval_runtime": 57.7395,
      "eval_samples_per_second": 4.33,
      "eval_steps_per_second": 0.554,
      "step": 370
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00013051546391752578,
      "loss": 0.403,
      "step": 371
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00013030927835051546,
      "loss": 0.4942,
      "step": 372
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00013010309278350517,
      "loss": 0.411,
      "step": 373
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00012989690721649486,
      "loss": 0.4153,
      "step": 374
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00012969072164948454,
      "loss": 0.4054,
      "step": 375
    },
    {
      "epoch": 1.5,
      "eval_loss": 0.5868126153945923,
      "eval_runtime": 57.7457,
      "eval_samples_per_second": 4.329,
      "eval_steps_per_second": 0.554,
      "step": 375
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00012948453608247425,
      "loss": 0.4248,
      "step": 376
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00012927835051546393,
      "loss": 0.4231,
      "step": 377
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00012907216494845362,
      "loss": 0.422,
      "step": 378
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0001288659793814433,
      "loss": 0.3957,
      "step": 379
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.000128659793814433,
      "loss": 0.4463,
      "step": 380
    },
    {
      "epoch": 1.52,
      "eval_loss": 0.5862159132957458,
      "eval_runtime": 58.2762,
      "eval_samples_per_second": 4.29,
      "eval_steps_per_second": 0.549,
      "step": 380
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0001284536082474227,
      "loss": 0.4907,
      "step": 381
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00012824742268041238,
      "loss": 0.4444,
      "step": 382
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00012804123711340206,
      "loss": 0.4767,
      "step": 383
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00012783505154639175,
      "loss": 0.4122,
      "step": 384
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00012762886597938143,
      "loss": 0.4023,
      "step": 385
    },
    {
      "epoch": 1.54,
      "eval_loss": 0.582167387008667,
      "eval_runtime": 58.0372,
      "eval_samples_per_second": 4.308,
      "eval_steps_per_second": 0.551,
      "step": 385
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00012742268041237114,
      "loss": 0.4174,
      "step": 386
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00012721649484536082,
      "loss": 0.3888,
      "step": 387
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0001270103092783505,
      "loss": 0.2883,
      "step": 388
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00012680412371134022,
      "loss": 0.4039,
      "step": 389
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0001265979381443299,
      "loss": 0.3211,
      "step": 390
    },
    {
      "epoch": 1.56,
      "eval_loss": 0.5785840749740601,
      "eval_runtime": 58.0313,
      "eval_samples_per_second": 4.308,
      "eval_steps_per_second": 0.551,
      "step": 390
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00012639175257731958,
      "loss": 0.3414,
      "step": 391
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00012618556701030927,
      "loss": 0.3639,
      "step": 392
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00012597938144329898,
      "loss": 0.3687,
      "step": 393
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00012577319587628866,
      "loss": 0.2935,
      "step": 394
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00012556701030927835,
      "loss": 0.4914,
      "step": 395
    },
    {
      "epoch": 1.58,
      "eval_loss": 0.5804648399353027,
      "eval_runtime": 57.7946,
      "eval_samples_per_second": 4.326,
      "eval_steps_per_second": 0.554,
      "step": 395
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00012536082474226806,
      "loss": 0.302,
      "step": 396
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00012515463917525774,
      "loss": 0.2966,
      "step": 397
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00012494845360824742,
      "loss": 0.3726,
      "step": 398
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0001247422680412371,
      "loss": 0.3031,
      "step": 399
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00012453608247422682,
      "loss": 0.3497,
      "step": 400
    },
    {
      "epoch": 1.6,
      "eval_loss": 0.5969975590705872,
      "eval_runtime": 58.0419,
      "eval_samples_per_second": 4.307,
      "eval_steps_per_second": 0.551,
      "step": 400
    }
  ],
  "logging_steps": 1,
  "max_steps": 1000,
  "num_train_epochs": 4,
  "save_steps": 200,
  "total_flos": 9123743473139712.0,
  "trial_name": null,
  "trial_params": null
}