{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008,
      "grad_norm": 34.44804382324219,
      "learning_rate": 1e-05,
      "loss": 13.0101,
      "mean_token_accuracy": 0.4696590006351471,
      "step": 1
    },
    {
      "epoch": 0.016,
      "grad_norm": 30.779788970947266,
      "learning_rate": 2e-05,
      "loss": 12.3851,
      "mean_token_accuracy": 0.47303473204374313,
      "step": 2
    },
    {
      "epoch": 0.024,
      "grad_norm": 29.67559242248535,
      "learning_rate": 3e-05,
      "loss": 12.3488,
      "mean_token_accuracy": 0.49709559231996536,
      "step": 3
    },
    {
      "epoch": 0.032,
      "grad_norm": 26.862010955810547,
      "learning_rate": 4e-05,
      "loss": 11.6596,
      "mean_token_accuracy": 0.5584611147642136,
      "step": 4
    },
    {
      "epoch": 0.04,
      "grad_norm": 22.10072135925293,
      "learning_rate": 5e-05,
      "loss": 10.1384,
      "mean_token_accuracy": 0.5924926251173019,
      "step": 5
    },
    {
      "epoch": 0.048,
      "grad_norm": 20.171361923217773,
      "learning_rate": 4.9473684210526315e-05,
      "loss": 9.5421,
      "mean_token_accuracy": 0.5888276249170303,
      "step": 6
    },
    {
      "epoch": 0.056,
      "grad_norm": 16.452842712402344,
      "learning_rate": 4.8947368421052635e-05,
      "loss": 8.4344,
      "mean_token_accuracy": 0.632336363196373,
      "step": 7
    },
    {
      "epoch": 0.064,
      "grad_norm": 12.61925983428955,
      "learning_rate": 4.842105263157895e-05,
      "loss": 7.7786,
      "mean_token_accuracy": 0.6633019298315048,
      "step": 8
    },
    {
      "epoch": 0.072,
      "grad_norm": 11.540078163146973,
      "learning_rate": 4.789473684210526e-05,
      "loss": 7.5607,
      "mean_token_accuracy": 0.6721473336219788,
      "step": 9
    },
    {
      "epoch": 0.08,
      "grad_norm": 9.264492988586426,
      "learning_rate": 4.736842105263158e-05,
      "loss": 7.5589,
      "mean_token_accuracy": 0.6924590468406677,
      "step": 10
    },
    {
      "epoch": 0.088,
      "grad_norm": 9.118257522583008,
      "learning_rate": 4.68421052631579e-05,
      "loss": 7.2038,
      "mean_token_accuracy": 0.6936477273702621,
      "step": 11
    },
    {
      "epoch": 0.096,
      "grad_norm": 9.140292167663574,
      "learning_rate": 4.6315789473684214e-05,
      "loss": 6.7882,
      "mean_token_accuracy": 0.7149988412857056,
      "step": 12
    },
    {
      "epoch": 0.104,
      "grad_norm": 8.944159507751465,
      "learning_rate": 4.5789473684210527e-05,
      "loss": 6.7822,
      "mean_token_accuracy": 0.7038442492485046,
      "step": 13
    },
    {
      "epoch": 0.112,
      "grad_norm": 8.406500816345215,
      "learning_rate": 4.5263157894736846e-05,
      "loss": 6.6675,
      "mean_token_accuracy": 0.6912005394697189,
      "step": 14
    },
    {
      "epoch": 0.12,
      "grad_norm": 7.81226921081543,
      "learning_rate": 4.473684210526316e-05,
      "loss": 6.156,
      "mean_token_accuracy": 0.7322177290916443,
      "step": 15
    },
    {
      "epoch": 0.128,
      "grad_norm": 10.26094913482666,
      "learning_rate": 4.421052631578947e-05,
      "loss": 6.3945,
      "mean_token_accuracy": 0.7067171931266785,
      "step": 16
    },
    {
      "epoch": 0.136,
      "grad_norm": 9.275789260864258,
      "learning_rate": 4.368421052631579e-05,
      "loss": 6.0611,
      "mean_token_accuracy": 0.7207391560077667,
      "step": 17
    },
    {
      "epoch": 0.144,
      "grad_norm": 9.354033470153809,
      "learning_rate": 4.3157894736842105e-05,
      "loss": 6.1713,
      "mean_token_accuracy": 0.7210505157709122,
      "step": 18
    },
    {
      "epoch": 0.152,
      "grad_norm": 7.362099647521973,
      "learning_rate": 4.2631578947368425e-05,
      "loss": 6.4072,
      "mean_token_accuracy": 0.7182824611663818,
      "step": 19
    },
    {
      "epoch": 0.16,
      "grad_norm": 7.947159290313721,
      "learning_rate": 4.210526315789474e-05,
      "loss": 6.2321,
      "mean_token_accuracy": 0.7235151827335358,
      "step": 20
    },
    {
      "epoch": 0.168,
      "grad_norm": 9.023740768432617,
      "learning_rate": 4.157894736842106e-05,
      "loss": 6.0462,
      "mean_token_accuracy": 0.7328377515077591,
      "step": 21
    },
    {
      "epoch": 0.176,
      "grad_norm": 7.097681999206543,
      "learning_rate": 4.105263157894737e-05,
      "loss": 5.1996,
      "mean_token_accuracy": 0.7640947103500366,
      "step": 22
    },
    {
      "epoch": 0.184,
      "grad_norm": 7.417032718658447,
      "learning_rate": 4.0526315789473684e-05,
      "loss": 6.2611,
      "mean_token_accuracy": 0.7178521305322647,
      "step": 23
    },
    {
      "epoch": 0.192,
      "grad_norm": 6.950761318206787,
      "learning_rate": 4e-05,
      "loss": 5.5961,
      "mean_token_accuracy": 0.7444904744625092,
      "step": 24
    },
    {
      "epoch": 0.2,
      "grad_norm": 11.129012107849121,
      "learning_rate": 3.9473684210526316e-05,
      "loss": 5.3222,
      "mean_token_accuracy": 0.75727978348732,
      "step": 25
    },
    {
      "epoch": 0.208,
      "grad_norm": 7.0950469970703125,
      "learning_rate": 3.894736842105263e-05,
      "loss": 5.4028,
      "mean_token_accuracy": 0.7472832798957825,
      "step": 26
    },
    {
      "epoch": 0.216,
      "grad_norm": 9.279951095581055,
      "learning_rate": 3.842105263157895e-05,
      "loss": 5.6822,
      "mean_token_accuracy": 0.7470583468675613,
      "step": 27
    },
    {
      "epoch": 0.224,
      "grad_norm": 7.986962795257568,
      "learning_rate": 3.789473684210527e-05,
      "loss": 5.5314,
      "mean_token_accuracy": 0.7577493786811829,
      "step": 28
    },
    {
      "epoch": 0.232,
      "grad_norm": 7.804550647735596,
      "learning_rate": 3.736842105263158e-05,
      "loss": 5.6165,
      "mean_token_accuracy": 0.7538002282381058,
      "step": 29
    },
    {
      "epoch": 0.24,
      "grad_norm": 7.608943939208984,
      "learning_rate": 3.6842105263157895e-05,
      "loss": 5.8104,
      "mean_token_accuracy": 0.7309564054012299,
      "step": 30
    },
    {
      "epoch": 0.248,
      "grad_norm": 7.622748374938965,
      "learning_rate": 3.6315789473684214e-05,
      "loss": 6.0722,
      "mean_token_accuracy": 0.740648090839386,
      "step": 31
    },
    {
      "epoch": 0.256,
      "grad_norm": 7.36374044418335,
      "learning_rate": 3.578947368421053e-05,
      "loss": 5.1848,
      "mean_token_accuracy": 0.7709122896194458,
      "step": 32
    },
    {
      "epoch": 0.264,
      "grad_norm": 6.820411682128906,
      "learning_rate": 3.526315789473684e-05,
      "loss": 5.0621,
      "mean_token_accuracy": 0.7692567408084869,
      "step": 33
    },
    {
      "epoch": 0.272,
      "grad_norm": 6.473299980163574,
      "learning_rate": 3.473684210526316e-05,
      "loss": 5.3879,
      "mean_token_accuracy": 0.7631748914718628,
      "step": 34
    },
    {
      "epoch": 0.28,
      "grad_norm": 7.249595642089844,
      "learning_rate": 3.421052631578947e-05,
      "loss": 5.7251,
      "mean_token_accuracy": 0.7535725235939026,
      "step": 35
    },
    {
      "epoch": 0.288,
      "grad_norm": 6.396822929382324,
      "learning_rate": 3.368421052631579e-05,
      "loss": 5.5203,
      "mean_token_accuracy": 0.7570009678602219,
      "step": 36
    },
    {
      "epoch": 0.296,
      "grad_norm": 5.999199390411377,
      "learning_rate": 3.3157894736842106e-05,
      "loss": 4.8803,
      "mean_token_accuracy": 0.7936893254518509,
      "step": 37
    },
    {
      "epoch": 0.304,
      "grad_norm": 6.992559909820557,
      "learning_rate": 3.2631578947368426e-05,
      "loss": 5.3993,
      "mean_token_accuracy": 0.7533106952905655,
      "step": 38
    },
    {
      "epoch": 0.312,
      "grad_norm": 7.724341869354248,
      "learning_rate": 3.210526315789474e-05,
      "loss": 5.6051,
      "mean_token_accuracy": 0.7366128116846085,
      "step": 39
    },
    {
      "epoch": 0.32,
      "grad_norm": 6.947137832641602,
      "learning_rate": 3.157894736842105e-05,
      "loss": 5.5221,
      "mean_token_accuracy": 0.7558678537607193,
      "step": 40
    },
    {
      "epoch": 0.328,
      "grad_norm": 7.61790657043457,
      "learning_rate": 3.105263157894737e-05,
      "loss": 5.0442,
      "mean_token_accuracy": 0.7664971798658371,
      "step": 41
    },
    {
      "epoch": 0.336,
      "grad_norm": 6.234894275665283,
      "learning_rate": 3.0526315789473684e-05,
      "loss": 5.1083,
      "mean_token_accuracy": 0.7650484591722488,
      "step": 42
    },
    {
      "epoch": 0.344,
      "grad_norm": 6.606740951538086,
      "learning_rate": 3e-05,
      "loss": 4.7578,
      "mean_token_accuracy": 0.7714228928089142,
      "step": 43
    },
    {
      "epoch": 0.352,
      "grad_norm": 7.6674957275390625,
      "learning_rate": 2.9473684210526314e-05,
      "loss": 4.8684,
      "mean_token_accuracy": 0.7777758836746216,
      "step": 44
    },
    {
      "epoch": 0.36,
      "grad_norm": 6.467043399810791,
      "learning_rate": 2.8947368421052634e-05,
      "loss": 4.8455,
      "mean_token_accuracy": 0.7818402796983719,
      "step": 45
    },
    {
      "epoch": 0.368,
      "grad_norm": 8.154311180114746,
      "learning_rate": 2.842105263157895e-05,
      "loss": 4.2716,
      "mean_token_accuracy": 0.7975698858499527,
      "step": 46
    },
    {
      "epoch": 0.376,
      "grad_norm": 6.493333339691162,
      "learning_rate": 2.7894736842105263e-05,
      "loss": 4.8411,
      "mean_token_accuracy": 0.7828859686851501,
      "step": 47
    },
    {
      "epoch": 0.384,
      "grad_norm": 6.82868766784668,
      "learning_rate": 2.7368421052631583e-05,
      "loss": 5.1772,
      "mean_token_accuracy": 0.7642232775688171,
      "step": 48
    },
    {
      "epoch": 0.392,
      "grad_norm": 7.6515793800354,
      "learning_rate": 2.6842105263157896e-05,
      "loss": 5.3152,
      "mean_token_accuracy": 0.7572789639234543,
      "step": 49
    },
    {
      "epoch": 0.4,
      "grad_norm": 7.436869144439697,
      "learning_rate": 2.6315789473684212e-05,
      "loss": 5.6628,
      "mean_token_accuracy": 0.7437437325716019,
      "step": 50
    },
    {
      "epoch": 0.408,
      "grad_norm": 7.494150161743164,
      "learning_rate": 2.578947368421053e-05,
      "loss": 4.7967,
      "mean_token_accuracy": 0.7860404402017593,
      "step": 51
    },
    {
      "epoch": 0.416,
      "grad_norm": 7.129732131958008,
      "learning_rate": 2.5263157894736845e-05,
      "loss": 4.1962,
      "mean_token_accuracy": 0.8091482371091843,
      "step": 52
    },
    {
      "epoch": 0.424,
      "grad_norm": 6.54312801361084,
      "learning_rate": 2.4736842105263158e-05,
      "loss": 5.1673,
      "mean_token_accuracy": 0.7621930986642838,
      "step": 53
    },
    {
      "epoch": 0.432,
      "grad_norm": 7.091205596923828,
      "learning_rate": 2.4210526315789474e-05,
      "loss": 4.9484,
      "mean_token_accuracy": 0.7637773156166077,
      "step": 54
    },
    {
      "epoch": 0.44,
      "grad_norm": 6.452225685119629,
      "learning_rate": 2.368421052631579e-05,
      "loss": 5.2121,
      "mean_token_accuracy": 0.7751423120498657,
      "step": 55
    },
    {
      "epoch": 0.448,
      "grad_norm": 7.079778671264648,
      "learning_rate": 2.3157894736842107e-05,
      "loss": 4.4913,
      "mean_token_accuracy": 0.7925348579883575,
      "step": 56
    },
    {
      "epoch": 0.456,
      "grad_norm": 6.021428108215332,
      "learning_rate": 2.2631578947368423e-05,
      "loss": 5.1743,
      "mean_token_accuracy": 0.7735058069229126,
      "step": 57
    },
    {
      "epoch": 0.464,
      "grad_norm": 6.639064788818359,
      "learning_rate": 2.2105263157894736e-05,
      "loss": 4.2683,
      "mean_token_accuracy": 0.7983661592006683,
      "step": 58
    },
    {
      "epoch": 0.472,
      "grad_norm": 7.0270256996154785,
      "learning_rate": 2.1578947368421053e-05,
      "loss": 4.7973,
      "mean_token_accuracy": 0.7833641171455383,
      "step": 59
    },
    {
      "epoch": 0.48,
      "grad_norm": 6.48342752456665,
      "learning_rate": 2.105263157894737e-05,
      "loss": 4.7573,
      "mean_token_accuracy": 0.7896635830402374,
      "step": 60
    },
    {
      "epoch": 0.488,
      "grad_norm": 7.281738758087158,
      "learning_rate": 2.0526315789473685e-05,
      "loss": 5.3465,
      "mean_token_accuracy": 0.7754404097795486,
      "step": 61
    },
    {
      "epoch": 0.496,
      "grad_norm": 7.372372627258301,
      "learning_rate": 2e-05,
      "loss": 4.3753,
      "mean_token_accuracy": 0.7940351366996765,
      "step": 62
    },
    {
      "epoch": 0.504,
      "grad_norm": 7.551279544830322,
      "learning_rate": 1.9473684210526315e-05,
      "loss": 4.7479,
      "mean_token_accuracy": 0.7812730073928833,
      "step": 63
    },
    {
      "epoch": 0.512,
      "grad_norm": 6.71967077255249,
      "learning_rate": 1.8947368421052634e-05,
      "loss": 4.7599,
      "mean_token_accuracy": 0.7793499678373337,
      "step": 64
    },
    {
      "epoch": 0.52,
      "grad_norm": 7.2533063888549805,
      "learning_rate": 1.8421052631578947e-05,
      "loss": 4.8565,
      "mean_token_accuracy": 0.7772107869386673,
      "step": 65
    },
    {
      "epoch": 0.528,
      "grad_norm": 6.089155197143555,
      "learning_rate": 1.7894736842105264e-05,
      "loss": 4.4817,
      "mean_token_accuracy": 0.7843845635652542,
      "step": 66
    },
    {
      "epoch": 0.536,
      "grad_norm": 7.141996383666992,
      "learning_rate": 1.736842105263158e-05,
      "loss": 4.8889,
      "mean_token_accuracy": 0.7912164181470871,
      "step": 67
    },
    {
      "epoch": 0.544,
      "grad_norm": 6.36722993850708,
      "learning_rate": 1.6842105263157896e-05,
      "loss": 5.663,
      "mean_token_accuracy": 0.7410993576049805,
      "step": 68
    },
    {
      "epoch": 0.552,
      "grad_norm": 6.695736885070801,
      "learning_rate": 1.6315789473684213e-05,
      "loss": 5.4292,
      "mean_token_accuracy": 0.7530855089426041,
      "step": 69
    },
    {
      "epoch": 0.56,
      "grad_norm": 6.8679518699646,
      "learning_rate": 1.5789473684210526e-05,
      "loss": 4.6897,
      "mean_token_accuracy": 0.7747458964586258,
      "step": 70
    },
    {
      "epoch": 0.568,
      "grad_norm": 7.445884704589844,
      "learning_rate": 1.5263157894736842e-05,
      "loss": 4.9695,
      "mean_token_accuracy": 0.782129317522049,
      "step": 71
    },
    {
      "epoch": 0.576,
      "grad_norm": 6.467252254486084,
      "learning_rate": 1.4736842105263157e-05,
      "loss": 4.2316,
      "mean_token_accuracy": 0.8010141849517822,
      "step": 72
    },
    {
      "epoch": 0.584,
      "grad_norm": 6.673328876495361,
      "learning_rate": 1.4210526315789475e-05,
      "loss": 5.5348,
      "mean_token_accuracy": 0.7492043673992157,
      "step": 73
    },
    {
      "epoch": 0.592,
      "grad_norm": 6.37332820892334,
      "learning_rate": 1.3684210526315791e-05,
      "loss": 4.1118,
      "mean_token_accuracy": 0.8063640743494034,
      "step": 74
    },
    {
      "epoch": 0.6,
      "grad_norm": 6.5554327964782715,
      "learning_rate": 1.3157894736842106e-05,
      "loss": 4.3587,
      "mean_token_accuracy": 0.7785357385873795,
      "step": 75
    },
    {
      "epoch": 0.608,
      "grad_norm": 7.242586135864258,
      "learning_rate": 1.2631578947368422e-05,
      "loss": 4.7259,
      "mean_token_accuracy": 0.7722227722406387,
      "step": 76
    },
    {
      "epoch": 0.616,
      "grad_norm": 6.499831676483154,
      "learning_rate": 1.2105263157894737e-05,
      "loss": 4.582,
      "mean_token_accuracy": 0.8019344508647919,
      "step": 77
    },
    {
      "epoch": 0.624,
      "grad_norm": 6.212286949157715,
      "learning_rate": 1.1578947368421053e-05,
      "loss": 4.3784,
      "mean_token_accuracy": 0.7915610671043396,
      "step": 78
    },
    {
      "epoch": 0.632,
      "grad_norm": 6.525815963745117,
      "learning_rate": 1.1052631578947368e-05,
      "loss": 4.6062,
      "mean_token_accuracy": 0.7760037034749985,
      "step": 79
    },
    {
      "epoch": 0.64,
      "grad_norm": 6.437106609344482,
      "learning_rate": 1.0526315789473684e-05,
      "loss": 4.3811,
      "mean_token_accuracy": 0.7914712876081467,
      "step": 80
    },
    {
      "epoch": 0.648,
      "grad_norm": 6.544358730316162,
      "learning_rate": 1e-05,
      "loss": 4.1778,
      "mean_token_accuracy": 0.813565582036972,
      "step": 81
    },
    {
      "epoch": 0.656,
      "grad_norm": 5.799524307250977,
      "learning_rate": 9.473684210526317e-06,
      "loss": 4.5642,
      "mean_token_accuracy": 0.7896921932697296,
      "step": 82
    },
    {
      "epoch": 0.664,
      "grad_norm": 6.26992130279541,
      "learning_rate": 8.947368421052632e-06,
      "loss": 4.4728,
      "mean_token_accuracy": 0.7932559102773666,
      "step": 83
    },
    {
      "epoch": 0.672,
      "grad_norm": 7.898167133331299,
      "learning_rate": 8.421052631578948e-06,
      "loss": 4.7496,
      "mean_token_accuracy": 0.7824009358882904,
      "step": 84
    },
    {
      "epoch": 0.68,
      "grad_norm": 5.840052127838135,
      "learning_rate": 7.894736842105263e-06,
      "loss": 4.4144,
      "mean_token_accuracy": 0.8065742999315262,
      "step": 85
    },
    {
      "epoch": 0.688,
      "grad_norm": 5.884693145751953,
      "learning_rate": 7.3684210526315784e-06,
      "loss": 4.0572,
      "mean_token_accuracy": 0.8163146525621414,
      "step": 86
    },
    {
      "epoch": 0.696,
      "grad_norm": 6.5094380378723145,
      "learning_rate": 6.842105263157896e-06,
      "loss": 4.692,
      "mean_token_accuracy": 0.7821700870990753,
      "step": 87
    },
    {
      "epoch": 0.704,
      "grad_norm": 6.171222686767578,
      "learning_rate": 6.315789473684211e-06,
      "loss": 4.7373,
      "mean_token_accuracy": 0.7865629494190216,
      "step": 88
    },
    {
      "epoch": 0.712,
      "grad_norm": 6.430420875549316,
      "learning_rate": 5.789473684210527e-06,
      "loss": 5.1443,
      "mean_token_accuracy": 0.7733548134565353,
      "step": 89
    },
    {
      "epoch": 0.72,
      "grad_norm": 6.6331987380981445,
      "learning_rate": 5.263157894736842e-06,
      "loss": 4.8172,
      "mean_token_accuracy": 0.7886734008789062,
      "step": 90
    },
    {
      "epoch": 0.728,
      "grad_norm": 6.18582820892334,
      "learning_rate": 4.736842105263159e-06,
      "loss": 4.0192,
      "mean_token_accuracy": 0.8108067065477371,
      "step": 91
    },
    {
      "epoch": 0.736,
      "grad_norm": 6.787937641143799,
      "learning_rate": 4.210526315789474e-06,
      "loss": 4.3634,
      "mean_token_accuracy": 0.7937927544116974,
      "step": 92
    },
    {
      "epoch": 0.744,
      "grad_norm": 6.101965427398682,
      "learning_rate": 3.6842105263157892e-06,
      "loss": 4.7228,
      "mean_token_accuracy": 0.7906162887811661,
      "step": 93
    },
    {
      "epoch": 0.752,
      "grad_norm": 6.77475118637085,
      "learning_rate": 3.1578947368421056e-06,
      "loss": 4.4398,
      "mean_token_accuracy": 0.7847026586532593,
      "step": 94
    },
    {
      "epoch": 0.76,
      "grad_norm": 6.1204023361206055,
      "learning_rate": 2.631578947368421e-06,
      "loss": 4.3872,
      "mean_token_accuracy": 0.7921232283115387,
      "step": 95
    },
    {
      "epoch": 0.768,
      "grad_norm": 6.262160301208496,
      "learning_rate": 2.105263157894737e-06,
      "loss": 4.2201,
      "mean_token_accuracy": 0.7995105534791946,
      "step": 96
    },
    {
      "epoch": 0.776,
      "grad_norm": 6.433416843414307,
      "learning_rate": 1.5789473684210528e-06,
      "loss": 4.2633,
      "mean_token_accuracy": 0.782690092921257,
      "step": 97
    },
    {
      "epoch": 0.784,
      "grad_norm": 6.012047290802002,
      "learning_rate": 1.0526315789473685e-06,
      "loss": 4.2714,
      "mean_token_accuracy": 0.791715532541275,
      "step": 98
    },
    {
      "epoch": 0.792,
      "grad_norm": 6.2551445960998535,
      "learning_rate": 5.263157894736843e-07,
      "loss": 4.4735,
      "mean_token_accuracy": 0.7887705862522125,
      "step": 99
    },
    {
      "epoch": 0.8,
      "grad_norm": 6.641271591186523,
      "learning_rate": 0.0,
      "loss": 4.7771,
      "mean_token_accuracy": 0.7853755056858063,
      "step": 100
    }
  ],
  "logging_steps": 1,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1546494935040000.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}