{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.986078886310905,
  "eval_steps": 500,
  "global_step": 107,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "grad_norm": 28.83467674255371,
      "learning_rate": 2e-05,
      "loss": 2.7878,
      "step": 1
    },
    {
      "epoch": 0.04,
      "grad_norm": 28.955718994140625,
      "learning_rate": 4e-05,
      "loss": 2.7826,
      "step": 2
    },
    {
      "epoch": 0.06,
      "grad_norm": 17.856950759887695,
      "learning_rate": 6e-05,
      "loss": 2.2388,
      "step": 3
    },
    {
      "epoch": 0.07,
      "grad_norm": 16.615158081054688,
      "learning_rate": 8e-05,
      "loss": 1.7809,
      "step": 4
    },
    {
      "epoch": 0.09,
      "grad_norm": 34.79571533203125,
      "learning_rate": 0.0001,
      "loss": 1.9714,
      "step": 5
    },
    {
      "epoch": 0.11,
      "grad_norm": 6.3113694190979,
      "learning_rate": 9.935064935064936e-05,
      "loss": 1.4032,
      "step": 6
    },
    {
      "epoch": 0.13,
      "grad_norm": 8.230330467224121,
      "learning_rate": 9.870129870129871e-05,
      "loss": 1.241,
      "step": 7
    },
    {
      "epoch": 0.15,
      "grad_norm": 17.19231605529785,
      "learning_rate": 9.805194805194806e-05,
      "loss": 1.2388,
      "step": 8
    },
    {
      "epoch": 0.17,
      "grad_norm": 6.900284767150879,
      "learning_rate": 9.74025974025974e-05,
      "loss": 1.0881,
      "step": 9
    },
    {
      "epoch": 0.19,
      "grad_norm": 2.7498185634613037,
      "learning_rate": 9.675324675324677e-05,
      "loss": 1.0071,
      "step": 10
    },
    {
      "epoch": 0.2,
      "grad_norm": 4.016067028045654,
      "learning_rate": 9.610389610389611e-05,
      "loss": 0.9962,
      "step": 11
    },
    {
      "epoch": 0.22,
      "grad_norm": 2.365187883377075,
      "learning_rate": 9.545454545454546e-05,
      "loss": 0.9928,
      "step": 12
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.4348371028900146,
      "learning_rate": 9.480519480519481e-05,
      "loss": 0.9107,
      "step": 13
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.8148932456970215,
      "learning_rate": 9.415584415584417e-05,
      "loss": 0.853,
      "step": 14
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.9855575561523438,
      "learning_rate": 9.35064935064935e-05,
      "loss": 0.8598,
      "step": 15
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.9960970878601074,
      "learning_rate": 9.285714285714286e-05,
      "loss": 0.8384,
      "step": 16
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.4565762281417847,
      "learning_rate": 9.220779220779221e-05,
      "loss": 0.7895,
      "step": 17
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.419858694076538,
      "learning_rate": 9.155844155844156e-05,
      "loss": 0.8162,
      "step": 18
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.3651608228683472,
      "learning_rate": 9.090909090909092e-05,
      "loss": 0.7924,
      "step": 19
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.1524626016616821,
      "learning_rate": 9.025974025974027e-05,
      "loss": 0.7484,
      "step": 20
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.9174069166183472,
      "learning_rate": 8.961038961038961e-05,
      "loss": 0.7276,
      "step": 21
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.1832919120788574,
      "learning_rate": 8.896103896103896e-05,
      "loss": 0.6658,
      "step": 22
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.3024309873580933,
      "learning_rate": 8.831168831168831e-05,
      "loss": 0.7379,
      "step": 23
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.1416062116622925,
      "learning_rate": 8.766233766233767e-05,
      "loss": 0.7202,
      "step": 24
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.2058277130126953,
      "learning_rate": 8.701298701298701e-05,
      "loss": 0.6777,
      "step": 25
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.0915583372116089,
      "learning_rate": 8.636363636363637e-05,
      "loss": 0.7063,
      "step": 26
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.0384303331375122,
      "learning_rate": 8.571428571428571e-05,
      "loss": 0.6447,
      "step": 27
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.131259560585022,
      "learning_rate": 8.506493506493507e-05,
      "loss": 0.6911,
      "step": 28
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.1505099534988403,
      "learning_rate": 8.441558441558442e-05,
      "loss": 0.6142,
      "step": 29
    },
    {
      "epoch": 0.56,
      "grad_norm": 2.6675026416778564,
      "learning_rate": 8.376623376623377e-05,
      "loss": 0.6181,
      "step": 30
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.2021816968917847,
      "learning_rate": 8.311688311688312e-05,
      "loss": 0.6542,
      "step": 31
    },
    {
      "epoch": 0.59,
      "grad_norm": 3.363269090652466,
      "learning_rate": 8.246753246753248e-05,
      "loss": 0.5968,
      "step": 32
    },
    {
      "epoch": 0.61,
      "grad_norm": 3.8007972240448,
      "learning_rate": 8.181818181818183e-05,
      "loss": 0.5983,
      "step": 33
    },
    {
      "epoch": 0.63,
      "grad_norm": 3.570556163787842,
      "learning_rate": 8.116883116883117e-05,
      "loss": 0.6375,
      "step": 34
    },
    {
      "epoch": 0.65,
      "grad_norm": 4.428549766540527,
      "learning_rate": 8.051948051948052e-05,
      "loss": 0.6519,
      "step": 35
    },
    {
      "epoch": 0.67,
      "grad_norm": 3.7240254878997803,
      "learning_rate": 7.987012987012987e-05,
      "loss": 0.5935,
      "step": 36
    },
    {
      "epoch": 0.69,
      "grad_norm": 2.453350782394409,
      "learning_rate": 7.922077922077923e-05,
      "loss": 0.604,
      "step": 37
    },
    {
      "epoch": 0.71,
      "grad_norm": 3.341071844100952,
      "learning_rate": 7.857142857142858e-05,
      "loss": 0.5994,
      "step": 38
    },
    {
      "epoch": 0.72,
      "grad_norm": 3.4222161769866943,
      "learning_rate": 7.792207792207793e-05,
      "loss": 0.5731,
      "step": 39
    },
    {
      "epoch": 0.74,
      "grad_norm": 2.9063804149627686,
      "learning_rate": 7.727272727272727e-05,
      "loss": 0.5566,
      "step": 40
    },
    {
      "epoch": 0.76,
      "grad_norm": 3.170330762863159,
      "learning_rate": 7.662337662337662e-05,
      "loss": 0.5684,
      "step": 41
    },
    {
      "epoch": 0.78,
      "grad_norm": 2.8363170623779297,
      "learning_rate": 7.597402597402598e-05,
      "loss": 0.5936,
      "step": 42
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.3944592475891113,
      "learning_rate": 7.532467532467533e-05,
      "loss": 0.5571,
      "step": 43
    },
    {
      "epoch": 0.82,
      "grad_norm": 2.816237688064575,
      "learning_rate": 7.467532467532467e-05,
      "loss": 0.5598,
      "step": 44
    },
    {
      "epoch": 0.84,
      "grad_norm": 2.57438325881958,
      "learning_rate": 7.402597402597404e-05,
      "loss": 0.5615,
      "step": 45
    },
    {
      "epoch": 0.85,
      "grad_norm": 2.3164243698120117,
      "learning_rate": 7.337662337662338e-05,
      "loss": 0.5381,
      "step": 46
    },
    {
      "epoch": 0.87,
      "grad_norm": 2.4226622581481934,
      "learning_rate": 7.272727272727273e-05,
      "loss": 0.5592,
      "step": 47
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.5568475723266602,
      "learning_rate": 7.207792207792208e-05,
      "loss": 0.5729,
      "step": 48
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.1320440769195557,
      "learning_rate": 7.142857142857143e-05,
      "loss": 0.5546,
      "step": 49
    },
    {
      "epoch": 0.93,
      "grad_norm": 2.220273733139038,
      "learning_rate": 7.077922077922077e-05,
      "loss": 0.5587,
      "step": 50
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.8099664449691772,
      "learning_rate": 7.012987012987014e-05,
      "loss": 0.5403,
      "step": 51
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.7992054224014282,
      "learning_rate": 6.948051948051948e-05,
      "loss": 0.5374,
      "step": 52
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.5891600847244263,
      "learning_rate": 6.883116883116883e-05,
      "loss": 0.5188,
      "step": 53
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.713174819946289,
      "learning_rate": 6.818181818181818e-05,
      "loss": 0.5632,
      "step": 54
    },
    {
      "epoch": 1.02,
      "grad_norm": 1.529587984085083,
      "learning_rate": 6.753246753246754e-05,
      "loss": 0.4607,
      "step": 55
    },
    {
      "epoch": 1.04,
      "grad_norm": 1.243905782699585,
      "learning_rate": 6.688311688311688e-05,
      "loss": 0.4807,
      "step": 56
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.92779141664505,
      "learning_rate": 6.623376623376624e-05,
      "loss": 0.4515,
      "step": 57
    },
    {
      "epoch": 1.08,
      "grad_norm": 1.8945612907409668,
      "learning_rate": 6.55844155844156e-05,
      "loss": 0.4788,
      "step": 58
    },
    {
      "epoch": 1.1,
      "grad_norm": 1.6997075080871582,
      "learning_rate": 6.493506493506494e-05,
      "loss": 0.4411,
      "step": 59
    },
    {
      "epoch": 1.11,
      "grad_norm": 1.4320181608200073,
      "learning_rate": 6.428571428571429e-05,
      "loss": 0.4474,
      "step": 60
    },
    {
      "epoch": 1.13,
      "grad_norm": 1.5225821733474731,
      "learning_rate": 6.363636363636364e-05,
      "loss": 0.4578,
      "step": 61
    },
    {
      "epoch": 1.15,
      "grad_norm": 1.4867883920669556,
      "learning_rate": 6.2987012987013e-05,
      "loss": 0.4838,
      "step": 62
    },
    {
      "epoch": 1.17,
      "grad_norm": 1.3935158252716064,
      "learning_rate": 6.233766233766233e-05,
      "loss": 0.4383,
      "step": 63
    },
    {
      "epoch": 1.19,
      "grad_norm": 1.3193836212158203,
      "learning_rate": 6.16883116883117e-05,
      "loss": 0.4369,
      "step": 64
    },
    {
      "epoch": 1.21,
      "grad_norm": 1.0565460920333862,
      "learning_rate": 6.103896103896104e-05,
      "loss": 0.4618,
      "step": 65
    },
    {
      "epoch": 1.23,
      "grad_norm": 1.9069595336914062,
      "learning_rate": 6.03896103896104e-05,
      "loss": 0.4636,
      "step": 66
    },
    {
      "epoch": 1.24,
      "grad_norm": 1.592588186264038,
      "learning_rate": 5.9740259740259744e-05,
      "loss": 0.4363,
      "step": 67
    },
    {
      "epoch": 1.26,
      "grad_norm": 1.392088770866394,
      "learning_rate": 5.90909090909091e-05,
      "loss": 0.4573,
      "step": 68
    },
    {
      "epoch": 1.28,
      "grad_norm": 1.4312165975570679,
      "learning_rate": 5.844155844155844e-05,
      "loss": 0.4404,
      "step": 69
    },
    {
      "epoch": 1.3,
      "grad_norm": 1.4288945198059082,
      "learning_rate": 5.7792207792207796e-05,
      "loss": 0.4517,
      "step": 70
    },
    {
      "epoch": 1.32,
      "grad_norm": 1.247237205505371,
      "learning_rate": 5.714285714285714e-05,
      "loss": 0.4731,
      "step": 71
    },
    {
      "epoch": 1.34,
      "grad_norm": 1.4292420148849487,
      "learning_rate": 5.64935064935065e-05,
      "loss": 0.4561,
      "step": 72
    },
    {
      "epoch": 1.35,
      "grad_norm": 1.1647123098373413,
      "learning_rate": 5.584415584415584e-05,
      "loss": 0.4702,
      "step": 73
    },
    {
      "epoch": 1.37,
      "grad_norm": 1.6123448610305786,
      "learning_rate": 5.51948051948052e-05,
      "loss": 0.4681,
      "step": 74
    },
    {
      "epoch": 1.39,
      "grad_norm": 1.3807860612869263,
      "learning_rate": 5.4545454545454546e-05,
      "loss": 0.4217,
      "step": 75
    },
    {
      "epoch": 1.41,
      "grad_norm": 1.4928926229476929,
      "learning_rate": 5.38961038961039e-05,
      "loss": 0.4344,
      "step": 76
    },
    {
      "epoch": 1.43,
      "grad_norm": 1.3118643760681152,
      "learning_rate": 5.3246753246753245e-05,
      "loss": 0.4509,
      "step": 77
    },
    {
      "epoch": 1.45,
      "grad_norm": 1.1260207891464233,
      "learning_rate": 5.25974025974026e-05,
      "loss": 0.4423,
      "step": 78
    },
    {
      "epoch": 1.47,
      "grad_norm": 1.1223907470703125,
      "learning_rate": 5.1948051948051944e-05,
      "loss": 0.4238,
      "step": 79
    },
    {
      "epoch": 1.48,
      "grad_norm": 1.123289704322815,
      "learning_rate": 5.1298701298701304e-05,
      "loss": 0.4415,
      "step": 80
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.8493527770042419,
      "learning_rate": 5.064935064935065e-05,
      "loss": 0.4278,
      "step": 81
    },
    {
      "epoch": 1.52,
      "grad_norm": 1.3558621406555176,
      "learning_rate": 5e-05,
      "loss": 0.4326,
      "step": 82
    },
    {
      "epoch": 1.54,
      "grad_norm": 1.198364496231079,
      "learning_rate": 4.9350649350649355e-05,
      "loss": 0.4504,
      "step": 83
    },
    {
      "epoch": 1.56,
      "grad_norm": 1.318117380142212,
      "learning_rate": 4.87012987012987e-05,
      "loss": 0.4389,
      "step": 84
    },
    {
      "epoch": 1.58,
      "grad_norm": 1.323292851448059,
      "learning_rate": 4.8051948051948054e-05,
      "loss": 0.4559,
      "step": 85
    },
    {
      "epoch": 1.6,
      "grad_norm": 1.0558339357376099,
      "learning_rate": 4.740259740259741e-05,
      "loss": 0.4613,
      "step": 86
    },
    {
      "epoch": 1.61,
      "grad_norm": 1.0301963090896606,
      "learning_rate": 4.675324675324675e-05,
      "loss": 0.4194,
      "step": 87
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.9772946834564209,
      "learning_rate": 4.6103896103896106e-05,
      "loss": 0.4278,
      "step": 88
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.93486487865448,
      "learning_rate": 4.545454545454546e-05,
      "loss": 0.454,
      "step": 89
    },
    {
      "epoch": 1.67,
      "grad_norm": 1.1458832025527954,
      "learning_rate": 4.4805194805194805e-05,
      "loss": 0.4363,
      "step": 90
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.9339586496353149,
      "learning_rate": 4.415584415584416e-05,
      "loss": 0.434,
      "step": 91
    },
    {
      "epoch": 1.71,
      "grad_norm": 1.3206239938735962,
      "learning_rate": 4.3506493506493503e-05,
      "loss": 0.4431,
      "step": 92
    },
    {
      "epoch": 1.73,
      "grad_norm": 1.1199604272842407,
      "learning_rate": 4.2857142857142856e-05,
      "loss": 0.4472,
      "step": 93
    },
    {
      "epoch": 1.74,
      "grad_norm": 1.0672180652618408,
      "learning_rate": 4.220779220779221e-05,
      "loss": 0.4091,
      "step": 94
    },
    {
      "epoch": 1.76,
      "grad_norm": 1.0920594930648804,
      "learning_rate": 4.155844155844156e-05,
      "loss": 0.417,
      "step": 95
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.9002637267112732,
      "learning_rate": 4.0909090909090915e-05,
      "loss": 0.4103,
      "step": 96
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.8468071222305298,
      "learning_rate": 4.025974025974026e-05,
      "loss": 0.4167,
      "step": 97
    },
    {
      "epoch": 1.82,
      "grad_norm": 0.9398940205574036,
      "learning_rate": 3.9610389610389614e-05,
      "loss": 0.4201,
      "step": 98
    },
    {
      "epoch": 1.84,
      "grad_norm": 0.7842636108398438,
      "learning_rate": 3.8961038961038966e-05,
      "loss": 0.4064,
      "step": 99
    },
    {
      "epoch": 1.86,
      "grad_norm": 1.0805550813674927,
      "learning_rate": 3.831168831168831e-05,
      "loss": 0.4271,
      "step": 100
    },
    {
      "epoch": 1.87,
      "grad_norm": 0.8424351215362549,
      "learning_rate": 3.7662337662337665e-05,
      "loss": 0.4087,
      "step": 101
    },
    {
      "epoch": 1.89,
      "grad_norm": 1.0850481986999512,
      "learning_rate": 3.701298701298702e-05,
      "loss": 0.409,
      "step": 102
    },
    {
      "epoch": 1.91,
      "grad_norm": 0.9894604086875916,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 0.4139,
      "step": 103
    },
    {
      "epoch": 1.93,
      "grad_norm": 0.97883540391922,
      "learning_rate": 3.571428571428572e-05,
      "loss": 0.4281,
      "step": 104
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.9291779398918152,
      "learning_rate": 3.506493506493507e-05,
      "loss": 0.4376,
      "step": 105
    },
    {
      "epoch": 1.97,
      "grad_norm": 0.8272209167480469,
      "learning_rate": 3.4415584415584416e-05,
      "loss": 0.42,
      "step": 106
    },
    {
      "epoch": 1.99,
      "grad_norm": 0.8428053855895996,
      "learning_rate": 3.376623376623377e-05,
      "loss": 0.4131,
      "step": 107
    }
  ],
  "logging_steps": 1,
  "max_steps": 159,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 5,
  "total_flos": 1.6171847189151744e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}