{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 15000,
  "global_step": 140625,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0035555555555555557,
      "eval_loss": 10.192023277282715,
      "eval_runtime": 3606.549,
      "eval_samples_per_second": 138.637,
      "eval_steps_per_second": 4.332,
      "step": 500
    },
    {
      "epoch": 0.0071111111111111115,
      "grad_norm": 2.387178421020508,
      "learning_rate": 7.103242320819113e-06,
      "loss": 10.1288,
      "step": 1000
    },
    {
      "epoch": 0.0071111111111111115,
      "eval_loss": 9.418336868286133,
      "eval_runtime": 3608.5139,
      "eval_samples_per_second": 138.561,
      "eval_steps_per_second": 4.33,
      "step": 1000
    },
    {
      "epoch": 0.010666666666666666,
      "eval_loss": 8.553510665893555,
      "eval_runtime": 3626.6638,
      "eval_samples_per_second": 137.868,
      "eval_steps_per_second": 4.308,
      "step": 1500
    },
    {
      "epoch": 0.014222222222222223,
      "grad_norm": 1.6770904064178467,
      "learning_rate": 1.4213594994311718e-05,
      "loss": 8.5935,
      "step": 2000
    },
    {
      "epoch": 0.014222222222222223,
      "eval_loss": 7.8756632804870605,
      "eval_runtime": 3629.9225,
      "eval_samples_per_second": 137.744,
      "eval_steps_per_second": 4.304,
      "step": 2000
    },
    {
      "epoch": 0.017777777777777778,
      "eval_loss": 7.625248432159424,
      "eval_runtime": 3637.2635,
      "eval_samples_per_second": 137.466,
      "eval_steps_per_second": 4.296,
      "step": 2500
    },
    {
      "epoch": 0.021333333333333333,
      "grad_norm": 1.6508774757385254,
      "learning_rate": 2.1323947667804326e-05,
      "loss": 7.6525,
      "step": 3000
    },
    {
      "epoch": 0.021333333333333333,
      "eval_loss": 7.534073829650879,
      "eval_runtime": 3654.9087,
      "eval_samples_per_second": 136.802,
      "eval_steps_per_second": 4.275,
      "step": 3000
    },
    {
      "epoch": 0.024888888888888887,
      "eval_loss": 7.45560884475708,
      "eval_runtime": 3671.4513,
      "eval_samples_per_second": 136.186,
      "eval_steps_per_second": 4.256,
      "step": 3500
    },
    {
      "epoch": 0.028444444444444446,
      "grad_norm": 2.0226364135742188,
      "learning_rate": 2.8434300341296933e-05,
      "loss": 7.4767,
      "step": 4000
    },
    {
      "epoch": 0.028444444444444446,
      "eval_loss": 7.391005992889404,
      "eval_runtime": 3680.5941,
      "eval_samples_per_second": 135.848,
      "eval_steps_per_second": 4.245,
      "step": 4000
    },
    {
      "epoch": 0.032,
      "eval_loss": 7.328425407409668,
      "eval_runtime": 3685.0206,
      "eval_samples_per_second": 135.684,
      "eval_steps_per_second": 4.24,
      "step": 4500
    },
    {
      "epoch": 0.035555555555555556,
      "grad_norm": 1.980565071105957,
      "learning_rate": 3.554465301478954e-05,
      "loss": 7.3497,
      "step": 5000
    },
    {
      "epoch": 0.035555555555555556,
      "eval_loss": 7.287872314453125,
      "eval_runtime": 3698.5108,
      "eval_samples_per_second": 135.19,
      "eval_steps_per_second": 4.225,
      "step": 5000
    },
    {
      "epoch": 0.03911111111111111,
      "eval_loss": 7.240482807159424,
      "eval_runtime": 3709.8586,
      "eval_samples_per_second": 134.776,
      "eval_steps_per_second": 4.212,
      "step": 5500
    },
    {
      "epoch": 0.042666666666666665,
      "grad_norm": 2.800004720687866,
      "learning_rate": 4.265500568828214e-05,
      "loss": 7.2465,
      "step": 6000
    },
    {
      "epoch": 0.042666666666666665,
      "eval_loss": 7.197748184204102,
      "eval_runtime": 3735.3694,
      "eval_samples_per_second": 133.856,
      "eval_steps_per_second": 4.183,
      "step": 6000
    },
    {
      "epoch": 0.04622222222222222,
      "eval_loss": 7.154662132263184,
      "eval_runtime": 3742.7005,
      "eval_samples_per_second": 133.593,
      "eval_steps_per_second": 4.175,
      "step": 6500
    },
    {
      "epoch": 0.049777777777777775,
      "grad_norm": 3.009158134460449,
      "learning_rate": 4.9765358361774746e-05,
      "loss": 7.1859,
      "step": 7000
    },
    {
      "epoch": 0.049777777777777775,
      "eval_loss": 7.120786190032959,
      "eval_runtime": 3747.8473,
      "eval_samples_per_second": 133.41,
      "eval_steps_per_second": 4.169,
      "step": 7000
    },
    {
      "epoch": 0.05333333333333334,
      "eval_loss": 7.0815510749816895,
      "eval_runtime": 3748.6541,
      "eval_samples_per_second": 133.381,
      "eval_steps_per_second": 4.168,
      "step": 7500
    },
    {
      "epoch": 0.05688888888888889,
      "grad_norm": 3.3026907444000244,
      "learning_rate": 4.963807983951255e-05,
      "loss": 7.1209,
      "step": 8000
    },
    {
      "epoch": 0.05688888888888889,
      "eval_loss": 7.053798675537109,
      "eval_runtime": 3748.7994,
      "eval_samples_per_second": 133.376,
      "eval_steps_per_second": 4.168,
      "step": 8000
    },
    {
      "epoch": 0.060444444444444446,
      "eval_loss": 7.023181438446045,
      "eval_runtime": 3634.6666,
      "eval_samples_per_second": 137.564,
      "eval_steps_per_second": 4.299,
      "step": 8500
    },
    {
      "epoch": 0.064,
      "grad_norm": 3.314805030822754,
      "learning_rate": 4.926380873249347e-05,
      "loss": 7.0348,
      "step": 9000
    },
    {
      "epoch": 0.07111111111111111,
      "grad_norm": 2.875911235809326,
      "learning_rate": 4.888953762547439e-05,
      "loss": 7.0072,
      "step": 10000
    },
    {
      "epoch": 0.07111111111111111,
      "eval_loss": 6.941441535949707,
      "eval_runtime": 3748.7055,
      "eval_samples_per_second": 133.379,
      "eval_steps_per_second": 4.168,
      "step": 10000
    },
    {
      "epoch": 0.07822222222222222,
      "grad_norm": 2.914552688598633,
      "learning_rate": 4.851526651845531e-05,
      "loss": 6.947,
      "step": 11000
    },
    {
      "epoch": 0.08533333333333333,
      "grad_norm": 2.964020252227783,
      "learning_rate": 4.814099541143623e-05,
      "loss": 6.8917,
      "step": 12000
    },
    {
      "epoch": 0.08533333333333333,
      "eval_loss": 6.830844402313232,
      "eval_runtime": 3577.7662,
      "eval_samples_per_second": 139.752,
      "eval_steps_per_second": 4.367,
      "step": 12000
    },
    {
      "epoch": 0.09244444444444444,
      "grad_norm": 3.0612735748291016,
      "learning_rate": 4.7766724304417146e-05,
      "loss": 6.8343,
      "step": 13000
    },
    {
      "epoch": 0.09955555555555555,
      "grad_norm": 2.7895731925964355,
      "learning_rate": 4.739245319739807e-05,
      "loss": 6.8006,
      "step": 14000
    },
    {
      "epoch": 0.09955555555555555,
      "eval_loss": 6.7523064613342285,
      "eval_runtime": 3593.4585,
      "eval_samples_per_second": 139.142,
      "eval_steps_per_second": 4.348,
      "step": 14000
    },
    {
      "epoch": 0.10666666666666667,
      "grad_norm": 4.3099260330200195,
      "learning_rate": 4.7018182090378985e-05,
      "loss": 6.7552,
      "step": 15000
    },
    {
      "epoch": 0.11377777777777778,
      "grad_norm": 4.26462459564209,
      "learning_rate": 4.664391098335991e-05,
      "loss": 6.716,
      "step": 16000
    },
    {
      "epoch": 0.11377777777777778,
      "eval_loss": 6.662391185760498,
      "eval_runtime": 3610.0752,
      "eval_samples_per_second": 138.501,
      "eval_steps_per_second": 4.328,
      "step": 16000
    },
    {
      "epoch": 0.12088888888888889,
      "grad_norm": 4.591431140899658,
      "learning_rate": 4.626963987634083e-05,
      "loss": 6.6886,
      "step": 17000
    },
    {
      "epoch": 0.128,
      "grad_norm": 3.8068220615386963,
      "learning_rate": 4.589536876932175e-05,
      "loss": 6.6432,
      "step": 18000
    },
    {
      "epoch": 0.128,
      "eval_loss": 6.585519313812256,
      "eval_runtime": 3618.075,
      "eval_samples_per_second": 138.195,
      "eval_steps_per_second": 4.319,
      "step": 18000
    },
    {
      "epoch": 0.1351111111111111,
      "grad_norm": 4.0343523025512695,
      "learning_rate": 4.552109766230267e-05,
      "loss": 6.6323,
      "step": 19000
    },
    {
      "epoch": 0.14222222222222222,
      "grad_norm": 4.53961181640625,
      "learning_rate": 4.5146826555283586e-05,
      "loss": 6.5772,
      "step": 20000
    },
    {
      "epoch": 0.14222222222222222,
      "eval_loss": 6.517657279968262,
      "eval_runtime": 3623.0508,
      "eval_samples_per_second": 138.005,
      "eval_steps_per_second": 4.313,
      "step": 20000
    },
    {
      "epoch": 0.14933333333333335,
      "grad_norm": 4.860204219818115,
      "learning_rate": 4.477255544826451e-05,
      "loss": 6.5456,
      "step": 21000
    },
    {
      "epoch": 0.15644444444444444,
      "grad_norm": 4.892744541168213,
      "learning_rate": 4.4398284341245425e-05,
      "loss": 6.4887,
      "step": 22000
    },
    {
      "epoch": 0.15644444444444444,
      "eval_loss": 6.45619010925293,
      "eval_runtime": 3632.2999,
      "eval_samples_per_second": 137.654,
      "eval_steps_per_second": 4.302,
      "step": 22000
    },
    {
      "epoch": 0.16355555555555557,
      "grad_norm": 4.8081135749816895,
      "learning_rate": 4.402401323422635e-05,
      "loss": 6.4871,
      "step": 23000
    },
    {
      "epoch": 0.17066666666666666,
      "grad_norm": 6.334560871124268,
      "learning_rate": 4.3649742127207264e-05,
      "loss": 6.4517,
      "step": 24000
    },
    {
      "epoch": 0.17066666666666666,
      "eval_loss": 6.385794639587402,
      "eval_runtime": 3643.4203,
      "eval_samples_per_second": 137.234,
      "eval_steps_per_second": 4.289,
      "step": 24000
    },
    {
      "epoch": 0.17777777777777778,
      "grad_norm": 6.487077713012695,
      "learning_rate": 4.3275471020188187e-05,
      "loss": 6.4336,
      "step": 25000
    },
    {
      "epoch": 0.18488888888888888,
      "grad_norm": 6.764550685882568,
      "learning_rate": 4.29011999131691e-05,
      "loss": 6.3861,
      "step": 26000
    },
    {
      "epoch": 0.18488888888888888,
      "eval_loss": 6.320189476013184,
      "eval_runtime": 3651.5073,
      "eval_samples_per_second": 136.93,
      "eval_steps_per_second": 4.279,
      "step": 26000
    },
    {
      "epoch": 0.192,
      "grad_norm": 6.320560932159424,
      "learning_rate": 4.2526928806150026e-05,
      "loss": 6.3545,
      "step": 27000
    },
    {
      "epoch": 0.1991111111111111,
      "grad_norm": 7.219266414642334,
      "learning_rate": 4.215265769913094e-05,
      "loss": 6.3368,
      "step": 28000
    },
    {
      "epoch": 0.1991111111111111,
      "eval_loss": 6.245439052581787,
      "eval_runtime": 3691.7383,
      "eval_samples_per_second": 135.438,
      "eval_steps_per_second": 4.232,
      "step": 28000
    },
    {
      "epoch": 0.20622222222222222,
      "grad_norm": 7.970364570617676,
      "learning_rate": 4.1778386592111864e-05,
      "loss": 6.2752,
      "step": 29000
    },
    {
      "epoch": 0.21333333333333335,
      "grad_norm": 4.8507232666015625,
      "learning_rate": 4.140411548509278e-05,
      "loss": 6.2344,
      "step": 30000
    },
    {
      "epoch": 0.21333333333333335,
      "eval_loss": 6.119101524353027,
      "eval_runtime": 3746.4316,
      "eval_samples_per_second": 133.46,
      "eval_steps_per_second": 4.171,
      "step": 30000
    },
    {
      "epoch": 0.22044444444444444,
      "grad_norm": 4.850826740264893,
      "learning_rate": 4.10298443780737e-05,
      "loss": 6.1825,
      "step": 31000
    },
    {
      "epoch": 0.22755555555555557,
      "grad_norm": 6.614843368530273,
      "learning_rate": 4.0655573271054626e-05,
      "loss": 6.1117,
      "step": 32000
    },
    {
      "epoch": 0.22755555555555557,
      "eval_loss": 5.977404594421387,
      "eval_runtime": 3748.1526,
      "eval_samples_per_second": 133.399,
      "eval_steps_per_second": 4.169,
      "step": 32000
    },
    {
      "epoch": 0.23466666666666666,
      "grad_norm": 7.195309162139893,
      "learning_rate": 4.028130216403554e-05,
      "loss": 6.0633,
      "step": 33000
    },
    {
      "epoch": 0.24177777777777779,
      "grad_norm": 6.1979522705078125,
      "learning_rate": 3.9907031057016465e-05,
      "loss": 5.9906,
      "step": 34000
    },
    {
      "epoch": 0.24177777777777779,
      "eval_loss": 5.844593524932861,
      "eval_runtime": 3748.7753,
      "eval_samples_per_second": 133.377,
      "eval_steps_per_second": 4.168,
      "step": 34000
    },
    {
      "epoch": 0.24888888888888888,
      "grad_norm": 5.907817840576172,
      "learning_rate": 3.953275994999738e-05,
      "loss": 5.9681,
      "step": 35000
    },
    {
      "epoch": 0.256,
      "grad_norm": 10.167774200439453,
      "learning_rate": 3.9158488842978304e-05,
      "loss": 5.8877,
      "step": 36000
    },
    {
      "epoch": 0.26311111111111113,
      "grad_norm": 6.226233005523682,
      "learning_rate": 3.878421773595922e-05,
      "loss": 5.8555,
      "step": 37000
    },
    {
      "epoch": 0.2702222222222222,
      "grad_norm": 8.957833290100098,
      "learning_rate": 3.840994662894014e-05,
      "loss": 5.7746,
      "step": 38000
    },
    {
      "epoch": 0.2773333333333333,
      "grad_norm": 5.283188819885254,
      "learning_rate": 3.803567552192106e-05,
      "loss": 5.7368,
      "step": 39000
    },
    {
      "epoch": 0.28444444444444444,
      "grad_norm": 8.641979217529297,
      "learning_rate": 3.766140441490198e-05,
      "loss": 5.684,
      "step": 40000
    },
    {
      "epoch": 0.28444444444444444,
      "eval_loss": 5.490408420562744,
      "eval_runtime": 3560.4593,
      "eval_samples_per_second": 140.431,
      "eval_steps_per_second": 4.388,
      "step": 40000
    },
    {
      "epoch": 0.29155555555555557,
      "grad_norm": 7.102558135986328,
      "learning_rate": 3.72871333078829e-05,
      "loss": 5.6058,
      "step": 41000
    },
    {
      "epoch": 0.2986666666666667,
      "grad_norm": 10.386979103088379,
      "learning_rate": 3.691286220086382e-05,
      "loss": 5.5434,
      "step": 42000
    },
    {
      "epoch": 0.30577777777777776,
      "grad_norm": 8.64822006225586,
      "learning_rate": 3.653859109384474e-05,
      "loss": 5.5001,
      "step": 43000
    },
    {
      "epoch": 0.3128888888888889,
      "grad_norm": 5.785928249359131,
      "learning_rate": 3.616431998682566e-05,
      "loss": 5.4771,
      "step": 44000
    },
    {
      "epoch": 0.32,
      "grad_norm": 5.5996294021606445,
      "learning_rate": 3.5790048879806576e-05,
      "loss": 5.4112,
      "step": 45000
    },
    {
      "epoch": 0.32,
      "eval_loss": 5.232267379760742,
      "eval_runtime": 3637.4962,
      "eval_samples_per_second": 137.457,
      "eval_steps_per_second": 4.296,
      "step": 45000
    },
    {
      "epoch": 0.32711111111111113,
      "grad_norm": 6.475677967071533,
      "learning_rate": 3.54157777727875e-05,
      "loss": 5.3516,
      "step": 46000
    },
    {
      "epoch": 0.3342222222222222,
      "grad_norm": 8.867932319641113,
      "learning_rate": 3.504150666576842e-05,
      "loss": 5.3162,
      "step": 47000
    },
    {
      "epoch": 0.3413333333333333,
      "grad_norm": 6.978850364685059,
      "learning_rate": 3.466723555874934e-05,
      "loss": 5.2946,
      "step": 48000
    },
    {
      "epoch": 0.34844444444444445,
      "grad_norm": 8.528263092041016,
      "learning_rate": 3.429296445173026e-05,
      "loss": 5.263,
      "step": 49000
    },
    {
      "epoch": 0.35555555555555557,
      "grad_norm": 8.980386734008789,
      "learning_rate": 3.3919067615818196e-05,
      "loss": 5.2126,
      "step": 50000
    },
    {
      "epoch": 0.3626666666666667,
      "grad_norm": 6.648774147033691,
      "learning_rate": 3.354479650879911e-05,
      "loss": 5.1586,
      "step": 51000
    },
    {
      "epoch": 0.36977777777777776,
      "grad_norm": 7.404205322265625,
      "learning_rate": 3.3170899672887054e-05,
      "loss": 5.1437,
      "step": 52000
    },
    {
      "epoch": 0.3768888888888889,
      "grad_norm": 8.521736145019531,
      "learning_rate": 3.279662856586798e-05,
      "loss": 5.1017,
      "step": 53000
    },
    {
      "epoch": 0.384,
      "grad_norm": 8.535130500793457,
      "learning_rate": 3.242273172995591e-05,
      "loss": 5.0448,
      "step": 54000
    },
    {
      "epoch": 0.39111111111111113,
      "grad_norm": 8.199433326721191,
      "learning_rate": 3.204846062293683e-05,
      "loss": 5.0525,
      "step": 55000
    },
    {
      "epoch": 0.3982222222222222,
      "grad_norm": 6.0434651374816895,
      "learning_rate": 3.167456378702477e-05,
      "loss": 5.028,
      "step": 56000
    },
    {
      "epoch": 0.4053333333333333,
      "grad_norm": 9.137014389038086,
      "learning_rate": 3.130029268000569e-05,
      "loss": 4.9989,
      "step": 57000
    },
    {
      "epoch": 0.41244444444444445,
      "grad_norm": 8.472779273986816,
      "learning_rate": 3.092639584409363e-05,
      "loss": 4.9428,
      "step": 58000
    },
    {
      "epoch": 0.41955555555555557,
      "grad_norm": 8.970288276672363,
      "learning_rate": 3.055212473707455e-05,
      "loss": 4.9541,
      "step": 59000
    },
    {
      "epoch": 0.4266666666666667,
      "grad_norm": 8.8677339553833,
      "learning_rate": 3.017785363005547e-05,
      "loss": 4.9165,
      "step": 60000
    },
    {
      "epoch": 0.4266666666666667,
      "eval_loss": 4.728287220001221,
      "eval_runtime": 1065.1935,
      "eval_samples_per_second": 469.398,
      "eval_steps_per_second": 14.669,
      "step": 60000
    },
    {
      "epoch": 0.43377777777777776,
      "grad_norm": 7.122616767883301,
      "learning_rate": 2.980358252303639e-05,
      "loss": 4.8869,
      "step": 61000
    },
    {
      "epoch": 0.4408888888888889,
      "grad_norm": 8.567378044128418,
      "learning_rate": 2.9429685687124324e-05,
      "loss": 4.87,
      "step": 62000
    },
    {
      "epoch": 0.448,
      "grad_norm": 6.809515953063965,
      "learning_rate": 2.9055788851212266e-05,
      "loss": 4.8423,
      "step": 63000
    },
    {
      "epoch": 0.45511111111111113,
      "grad_norm": 8.371034622192383,
      "learning_rate": 2.8681517744193186e-05,
      "loss": 4.8137,
      "step": 64000
    },
    {
      "epoch": 0.4622222222222222,
      "grad_norm": 11.026784896850586,
      "learning_rate": 2.8307246637174105e-05,
      "loss": 4.7887,
      "step": 65000
    },
    {
      "epoch": 0.4693333333333333,
      "grad_norm": 7.04138708114624,
      "learning_rate": 2.7933349801262044e-05,
      "loss": 4.7642,
      "step": 66000
    },
    {
      "epoch": 0.47644444444444445,
      "grad_norm": 6.654124736785889,
      "learning_rate": 2.7559078694242964e-05,
      "loss": 4.7682,
      "step": 67000
    },
    {
      "epoch": 0.48355555555555557,
      "grad_norm": 9.198745727539062,
      "learning_rate": 2.718480758722388e-05,
      "loss": 4.7505,
      "step": 68000
    },
    {
      "epoch": 0.49066666666666664,
      "grad_norm": 7.339107036590576,
      "learning_rate": 2.68105364802048e-05,
      "loss": 4.695,
      "step": 69000
    },
    {
      "epoch": 0.49777777777777776,
      "grad_norm": 7.021571636199951,
      "learning_rate": 2.6436639644292742e-05,
      "loss": 4.6952,
      "step": 70000
    },
    {
      "epoch": 0.5048888888888889,
      "grad_norm": 6.531971454620361,
      "learning_rate": 2.606236853727366e-05,
      "loss": 4.7041,
      "step": 71000
    },
    {
      "epoch": 0.512,
      "grad_norm": 8.730067253112793,
      "learning_rate": 2.56884717013616e-05,
      "loss": 4.6792,
      "step": 72000
    },
    {
      "epoch": 0.5191111111111111,
      "grad_norm": 6.701645374298096,
      "learning_rate": 2.531420059434252e-05,
      "loss": 4.6603,
      "step": 73000
    },
    {
      "epoch": 0.5262222222222223,
      "grad_norm": 8.028661727905273,
      "learning_rate": 2.493992948732344e-05,
      "loss": 4.6216,
      "step": 74000
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 8.593280792236328,
      "learning_rate": 2.456565838030436e-05,
      "loss": 4.5962,
      "step": 75000
    },
    {
      "epoch": 0.5333333333333333,
      "eval_loss": 4.437507152557373,
      "eval_runtime": 1115.0836,
      "eval_samples_per_second": 448.397,
      "eval_steps_per_second": 14.012,
      "step": 75000
    },
    {
      "epoch": 0.5404444444444444,
      "grad_norm": 7.844559192657471,
      "learning_rate": 2.4191761544392298e-05,
      "loss": 4.5922,
      "step": 76000
    },
    {
      "epoch": 0.5475555555555556,
      "grad_norm": 8.809325218200684,
      "learning_rate": 2.3817490437373217e-05,
      "loss": 4.5741,
      "step": 77000
    },
    {
      "epoch": 0.5546666666666666,
      "grad_norm": 8.746879577636719,
      "learning_rate": 2.3443593601461156e-05,
      "loss": 4.5682,
      "step": 78000
    },
    {
      "epoch": 0.5617777777777778,
      "grad_norm": 6.95306921005249,
      "learning_rate": 2.3069322494442076e-05,
      "loss": 4.5368,
      "step": 79000
    },
    {
      "epoch": 0.5688888888888889,
      "grad_norm": 6.743412494659424,
      "learning_rate": 2.2695425658530015e-05,
      "loss": 4.533,
      "step": 80000
    },
    {
      "epoch": 0.576,
      "grad_norm": 7.390259742736816,
      "learning_rate": 2.2321154551510934e-05,
      "loss": 4.4858,
      "step": 81000
    },
    {
      "epoch": 0.5831111111111111,
      "grad_norm": 6.666167736053467,
      "learning_rate": 2.1947257715598873e-05,
      "loss": 4.4989,
      "step": 82000
    },
    {
      "epoch": 0.5902222222222222,
      "grad_norm": 11.42460823059082,
      "learning_rate": 2.1572986608579793e-05,
      "loss": 4.4954,
      "step": 83000
    },
    {
      "epoch": 0.5973333333333334,
      "grad_norm": 8.539606094360352,
      "learning_rate": 2.1198715501560712e-05,
      "loss": 4.4719,
      "step": 84000
    },
    {
      "epoch": 0.6044444444444445,
      "grad_norm": 7.553934097290039,
      "learning_rate": 2.082481866564865e-05,
      "loss": 4.4676,
      "step": 85000
    },
    {
      "epoch": 0.6115555555555555,
      "grad_norm": 6.772820949554443,
      "learning_rate": 2.045054755862957e-05,
      "loss": 4.4472,
      "step": 86000
    },
    {
      "epoch": 0.6186666666666667,
      "grad_norm": 7.853897571563721,
      "learning_rate": 2.0076650722717507e-05,
      "loss": 4.4265,
      "step": 87000
    },
    {
      "epoch": 0.6257777777777778,
      "grad_norm": 9.26160717010498,
      "learning_rate": 1.9702379615698426e-05,
      "loss": 4.4468,
      "step": 88000
    },
    {
      "epoch": 0.6328888888888888,
      "grad_norm": 7.147493362426758,
      "learning_rate": 1.932810850867935e-05,
      "loss": 4.4117,
      "step": 89000
    },
    {
      "epoch": 0.64,
      "grad_norm": 7.0213446617126465,
      "learning_rate": 1.8954211672767285e-05,
      "loss": 4.4131,
      "step": 90000
    },
    {
      "epoch": 0.64,
      "eval_loss": 4.241979122161865,
      "eval_runtime": 1150.9235,
      "eval_samples_per_second": 434.434,
      "eval_steps_per_second": 13.576,
      "step": 90000
    },
    {
      "epoch": 0.6471111111111111,
      "grad_norm": 8.582975387573242,
      "learning_rate": 1.8580314836855227e-05,
      "loss": 4.4139,
      "step": 91000
    },
    {
      "epoch": 0.6542222222222223,
      "grad_norm": 8.507649421691895,
      "learning_rate": 1.8206043729836143e-05,
      "loss": 4.3542,
      "step": 92000
    },
    {
      "epoch": 0.6613333333333333,
      "grad_norm": 6.707493305206299,
      "learning_rate": 1.7832146893924085e-05,
      "loss": 4.3797,
      "step": 93000
    },
    {
      "epoch": 0.6684444444444444,
      "grad_norm": 7.590912818908691,
      "learning_rate": 1.7457875786905005e-05,
      "loss": 4.3638,
      "step": 94000
    },
    {
      "epoch": 0.6755555555555556,
      "grad_norm": 10.533121109008789,
      "learning_rate": 1.7083604679885924e-05,
      "loss": 4.3663,
      "step": 95000
    },
    {
      "epoch": 0.6826666666666666,
      "grad_norm": 9.008646965026855,
      "learning_rate": 1.670933357286684e-05,
      "loss": 4.3485,
      "step": 96000
    },
    {
      "epoch": 0.6897777777777778,
      "grad_norm": 5.894834518432617,
      "learning_rate": 1.6335436736954783e-05,
      "loss": 4.3309,
      "step": 97000
    },
    {
      "epoch": 0.6968888888888889,
      "grad_norm": 8.39154052734375,
      "learning_rate": 1.5961165629935702e-05,
      "loss": 4.3287,
      "step": 98000
    },
    {
      "epoch": 0.704,
      "grad_norm": 8.664682388305664,
      "learning_rate": 1.558726879402364e-05,
      "loss": 4.3453,
      "step": 99000
    },
    {
      "epoch": 0.7111111111111111,
      "grad_norm": 6.261595249176025,
      "learning_rate": 1.521299768700456e-05,
      "loss": 4.3248,
      "step": 100000
    },
    {
      "epoch": 0.7182222222222222,
      "grad_norm": 6.0963006019592285,
      "learning_rate": 1.4839100851092497e-05,
      "loss": 4.3157,
      "step": 101000
    },
    {
      "epoch": 0.7253333333333334,
      "grad_norm": 8.00066089630127,
      "learning_rate": 1.446482974407342e-05,
      "loss": 4.2899,
      "step": 102000
    },
    {
      "epoch": 0.7324444444444445,
      "grad_norm": 7.378486156463623,
      "learning_rate": 1.4090932908161355e-05,
      "loss": 4.2866,
      "step": 103000
    },
    {
      "epoch": 0.7395555555555555,
      "grad_norm": 7.146476745605469,
      "learning_rate": 1.3716661801142275e-05,
      "loss": 4.2811,
      "step": 104000
    },
    {
      "epoch": 0.7466666666666667,
      "grad_norm": 6.750655174255371,
      "learning_rate": 1.3342764965230214e-05,
      "loss": 4.286,
      "step": 105000
    },
    {
      "epoch": 0.7466666666666667,
      "eval_loss": 4.117337703704834,
      "eval_runtime": 1173.3243,
      "eval_samples_per_second": 426.14,
      "eval_steps_per_second": 13.317,
      "step": 105000
    },
    {
      "epoch": 0.7537777777777778,
      "grad_norm": 8.043438911437988,
      "learning_rate": 1.2968493858211133e-05,
      "loss": 4.2712,
      "step": 106000
    },
    {
      "epoch": 0.7608888888888888,
      "grad_norm": 6.856295585632324,
      "learning_rate": 1.2594597022299074e-05,
      "loss": 4.269,
      "step": 107000
    },
    {
      "epoch": 0.768,
      "grad_norm": 7.801486968994141,
      "learning_rate": 1.2220325915279993e-05,
      "loss": 4.2669,
      "step": 108000
    },
    {
      "epoch": 0.7751111111111111,
      "grad_norm": 6.378691673278809,
      "learning_rate": 1.184642907936793e-05,
      "loss": 4.2745,
      "step": 109000
    },
    {
      "epoch": 0.7822222222222223,
      "grad_norm": 6.500033855438232,
      "learning_rate": 1.1472157972348852e-05,
      "loss": 4.2327,
      "step": 110000
    },
    {
      "epoch": 0.7893333333333333,
      "grad_norm": 7.333186626434326,
      "learning_rate": 1.109826113643679e-05,
      "loss": 4.2546,
      "step": 111000
    },
    {
      "epoch": 0.7964444444444444,
      "grad_norm": 6.290594577789307,
      "learning_rate": 1.0723990029417709e-05,
      "loss": 4.211,
      "step": 112000
    },
    {
      "epoch": 0.8035555555555556,
      "grad_norm": 7.892796993255615,
      "learning_rate": 1.035009319350565e-05,
      "loss": 4.2283,
      "step": 113000
    },
    {
      "epoch": 0.8106666666666666,
      "grad_norm": 12.832432746887207,
      "learning_rate": 9.975822086486567e-06,
      "loss": 4.2151,
      "step": 114000
    },
    {
      "epoch": 0.8177777777777778,
      "grad_norm": 7.7401885986328125,
      "learning_rate": 9.601925250574506e-06,
      "loss": 4.2408,
      "step": 115000
    },
    {
      "epoch": 0.8248888888888889,
      "grad_norm": 7.005733489990234,
      "learning_rate": 9.227654143555427e-06,
      "loss": 4.2157,
      "step": 116000
    },
    {
      "epoch": 0.832,
      "grad_norm": 6.389645576477051,
      "learning_rate": 8.853383036536345e-06,
      "loss": 4.1953,
      "step": 117000
    },
    {
      "epoch": 0.8391111111111111,
      "grad_norm": 10.353365898132324,
      "learning_rate": 8.479860471731304e-06,
      "loss": 4.1834,
      "step": 118000
    },
    {
      "epoch": 0.8462222222222222,
      "grad_norm": 7.046708106994629,
      "learning_rate": 8.105589364712223e-06,
      "loss": 4.1948,
      "step": 119000
    },
    {
      "epoch": 0.8533333333333334,
      "grad_norm": 9.073925018310547,
      "learning_rate": 7.731318257693143e-06,
      "loss": 4.2043,
      "step": 120000
    },
    {
      "epoch": 0.8533333333333334,
      "eval_loss": 4.04020881652832,
      "eval_runtime": 1223.3756,
      "eval_samples_per_second": 408.705,
      "eval_steps_per_second": 12.772,
      "step": 120000
    },
    {
      "epoch": 0.8604444444444445,
      "grad_norm": 8.47717571258545,
      "learning_rate": 7.357421421781081e-06,
      "loss": 4.1805,
      "step": 121000
    },
    {
      "epoch": 0.8675555555555555,
      "grad_norm": 7.262242794036865,
      "learning_rate": 6.983150314762002e-06,
      "loss": 4.1904,
      "step": 122000
    },
    {
      "epoch": 0.8746666666666667,
      "grad_norm": 7.6930766105651855,
      "learning_rate": 6.608879207742922e-06,
      "loss": 4.1905,
      "step": 123000
    },
    {
      "epoch": 0.8817777777777778,
      "grad_norm": 8.956729888916016,
      "learning_rate": 6.23460810072384e-06,
      "loss": 4.1768,
      "step": 124000
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 8.849599838256836,
      "learning_rate": 5.860711264811779e-06,
      "loss": 4.1684,
      "step": 125000
    },
    {
      "epoch": 0.896,
      "grad_norm": 7.687771320343018,
      "learning_rate": 5.486440157792699e-06,
      "loss": 4.1535,
      "step": 126000
    },
    {
      "epoch": 0.9031111111111111,
      "grad_norm": 7.867640495300293,
      "learning_rate": 5.112543321880638e-06,
      "loss": 4.1821,
      "step": 127000
    },
    {
      "epoch": 0.9102222222222223,
      "grad_norm": 6.583714008331299,
      "learning_rate": 4.738272214861557e-06,
      "loss": 4.1653,
      "step": 128000
    },
    {
      "epoch": 0.9173333333333333,
      "grad_norm": 6.076399326324463,
      "learning_rate": 4.3643753789494955e-06,
      "loss": 4.1669,
      "step": 129000
    },
    {
      "epoch": 0.9244444444444444,
      "grad_norm": 7.333767890930176,
      "learning_rate": 3.990104271930416e-06,
      "loss": 4.1656,
      "step": 130000
    },
    {
      "epoch": 0.9315555555555556,
      "grad_norm": 7.766304969787598,
      "learning_rate": 3.616207436018355e-06,
      "loss": 4.1568,
      "step": 131000
    },
    {
      "epoch": 0.9386666666666666,
      "grad_norm": 7.178610324859619,
      "learning_rate": 3.241936328999274e-06,
      "loss": 4.1382,
      "step": 132000
    },
    {
      "epoch": 0.9457777777777778,
      "grad_norm": 7.358198165893555,
      "learning_rate": 2.868039493087213e-06,
      "loss": 4.1526,
      "step": 133000
    },
    {
      "epoch": 0.9528888888888889,
      "grad_norm": 6.876594066619873,
      "learning_rate": 2.4937683860681324e-06,
      "loss": 4.1653,
      "step": 134000
    },
    {
      "epoch": 0.96,
      "grad_norm": 8.42979621887207,
      "learning_rate": 2.119871550156071e-06,
      "loss": 4.1835,
      "step": 135000
    },
    {
      "epoch": 0.96,
      "eval_loss": 4.000138282775879,
      "eval_runtime": 1234.4744,
      "eval_samples_per_second": 405.031,
      "eval_steps_per_second": 12.657,
      "step": 135000
    },
    {
      "epoch": 0.9671111111111111,
      "grad_norm": 10.133835792541504,
      "learning_rate": 1.7456004431369907e-06,
      "loss": 4.1556,
      "step": 136000
    },
    {
      "epoch": 0.9742222222222222,
      "grad_norm": 8.190239906311035,
      "learning_rate": 1.3717036072249296e-06,
      "loss": 4.1504,
      "step": 137000
    },
    {
      "epoch": 0.9813333333333333,
      "grad_norm": 8.191669464111328,
      "learning_rate": 9.974325002058493e-07,
      "loss": 4.1432,
      "step": 138000
    },
    {
      "epoch": 0.9884444444444445,
      "grad_norm": 7.776695728302002,
      "learning_rate": 6.235356642937879e-07,
      "loss": 4.1321,
      "step": 139000
    },
    {
      "epoch": 0.9955555555555555,
      "grad_norm": 9.083742141723633,
      "learning_rate": 2.492645572747075e-07,
      "loss": 4.1393,
      "step": 140000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 140625,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 15000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.541019136e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}