{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 15.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.009009009009009009,
      "grad_norm": 608.0,
      "learning_rate": 1.1976047904191619e-06,
      "loss": 58.5641,
      "step": 1
    },
    {
      "epoch": 0.04504504504504504,
      "grad_norm": 536.0,
      "learning_rate": 5.9880239520958085e-06,
      "loss": 54.74,
      "step": 5
    },
    {
      "epoch": 0.09009009009009009,
      "grad_norm": 488.0,
      "learning_rate": 1.1976047904191617e-05,
      "loss": 52.9637,
      "step": 10
    },
    {
      "epoch": 0.13513513513513514,
      "grad_norm": 392.0,
      "learning_rate": 1.7964071856287426e-05,
      "loss": 40.7219,
      "step": 15
    },
    {
      "epoch": 0.18018018018018017,
      "grad_norm": 46.5,
      "learning_rate": 2.3952095808383234e-05,
      "loss": 28.1416,
      "step": 20
    },
    {
      "epoch": 0.22522522522522523,
      "grad_norm": 40.25,
      "learning_rate": 2.994011976047904e-05,
      "loss": 24.2461,
      "step": 25
    },
    {
      "epoch": 0.2702702702702703,
      "grad_norm": 24.125,
      "learning_rate": 3.592814371257485e-05,
      "loss": 22.2451,
      "step": 30
    },
    {
      "epoch": 0.3153153153153153,
      "grad_norm": 13.5625,
      "learning_rate": 4.191616766467066e-05,
      "loss": 20.2199,
      "step": 35
    },
    {
      "epoch": 0.36036036036036034,
      "grad_norm": 7.5625,
      "learning_rate": 4.790419161676647e-05,
      "loss": 19.4029,
      "step": 40
    },
    {
      "epoch": 0.40540540540540543,
      "grad_norm": 9.1875,
      "learning_rate": 5.389221556886228e-05,
      "loss": 18.5345,
      "step": 45
    },
    {
      "epoch": 0.45045045045045046,
      "grad_norm": 15.75,
      "learning_rate": 5.988023952095808e-05,
      "loss": 17.2917,
      "step": 50
    },
    {
      "epoch": 0.4954954954954955,
      "grad_norm": 37.25,
      "learning_rate": 6.58682634730539e-05,
      "loss": 15.8479,
      "step": 55
    },
    {
      "epoch": 0.5405405405405406,
      "grad_norm": 59.75,
      "learning_rate": 7.18562874251497e-05,
      "loss": 10.5192,
      "step": 60
    },
    {
      "epoch": 0.5855855855855856,
      "grad_norm": 13.375,
      "learning_rate": 7.784431137724552e-05,
      "loss": 3.399,
      "step": 65
    },
    {
      "epoch": 0.6306306306306306,
      "grad_norm": 2.84375,
      "learning_rate": 8.383233532934131e-05,
      "loss": 2.1324,
      "step": 70
    },
    {
      "epoch": 0.6756756756756757,
      "grad_norm": 2.125,
      "learning_rate": 8.982035928143712e-05,
      "loss": 1.8224,
      "step": 75
    },
    {
      "epoch": 0.7207207207207207,
      "grad_norm": 3.703125,
      "learning_rate": 9.580838323353294e-05,
      "loss": 1.6034,
      "step": 80
    },
    {
      "epoch": 0.7657657657657657,
      "grad_norm": 2.75,
      "learning_rate": 0.00010179640718562875,
      "loss": 1.4811,
      "step": 85
    },
    {
      "epoch": 0.8108108108108109,
      "grad_norm": 7.53125,
      "learning_rate": 0.00010778443113772456,
      "loss": 1.4187,
      "step": 90
    },
    {
      "epoch": 0.8558558558558559,
      "grad_norm": 2.234375,
      "learning_rate": 0.00011377245508982037,
      "loss": 1.3517,
      "step": 95
    },
    {
      "epoch": 0.9009009009009009,
      "grad_norm": 8.375,
      "learning_rate": 0.00011976047904191617,
      "loss": 1.2567,
      "step": 100
    },
    {
      "epoch": 0.9459459459459459,
      "grad_norm": 5.90625,
      "learning_rate": 0.00012574850299401196,
      "loss": 1.2634,
      "step": 105
    },
    {
      "epoch": 0.990990990990991,
      "grad_norm": 4.1875,
      "learning_rate": 0.0001317365269461078,
      "loss": 1.2272,
      "step": 110
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.3900420665740967,
      "eval_runtime": 1.0174,
      "eval_samples_per_second": 4.914,
      "eval_steps_per_second": 1.966,
      "step": 111
    },
    {
      "epoch": 1.0360360360360361,
      "grad_norm": 3.234375,
      "learning_rate": 0.00013772455089820359,
      "loss": 1.1598,
      "step": 115
    },
    {
      "epoch": 1.0810810810810811,
      "grad_norm": 2.3125,
      "learning_rate": 0.0001437125748502994,
      "loss": 1.1474,
      "step": 120
    },
    {
      "epoch": 1.1261261261261262,
      "grad_norm": 6.375,
      "learning_rate": 0.0001497005988023952,
      "loss": 1.1051,
      "step": 125
    },
    {
      "epoch": 1.1711711711711712,
      "grad_norm": 6.625,
      "learning_rate": 0.00015568862275449103,
      "loss": 1.0869,
      "step": 130
    },
    {
      "epoch": 1.2162162162162162,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00016167664670658683,
      "loss": 1.0864,
      "step": 135
    },
    {
      "epoch": 1.2612612612612613,
      "grad_norm": 6.1875,
      "learning_rate": 0.00016766467065868263,
      "loss": 1.0419,
      "step": 140
    },
    {
      "epoch": 1.3063063063063063,
      "grad_norm": 2.84375,
      "learning_rate": 0.00017365269461077845,
      "loss": 1.0647,
      "step": 145
    },
    {
      "epoch": 1.3513513513513513,
      "grad_norm": 24.0,
      "learning_rate": 0.00017964071856287425,
      "loss": 1.0682,
      "step": 150
    },
    {
      "epoch": 1.3963963963963963,
      "grad_norm": 5.3125,
      "learning_rate": 0.00018562874251497007,
      "loss": 1.0375,
      "step": 155
    },
    {
      "epoch": 1.4414414414414414,
      "grad_norm": 2.5625,
      "learning_rate": 0.00019161676646706587,
      "loss": 1.0111,
      "step": 160
    },
    {
      "epoch": 1.4864864864864864,
      "grad_norm": 4.25,
      "learning_rate": 0.0001976047904191617,
      "loss": 0.9605,
      "step": 165
    },
    {
      "epoch": 1.5315315315315314,
      "grad_norm": 6.15625,
      "learning_rate": 0.00019999802081131313,
      "loss": 0.9843,
      "step": 170
    },
    {
      "epoch": 1.5765765765765765,
      "grad_norm": 2.484375,
      "learning_rate": 0.00019998592605305064,
      "loss": 1.0225,
      "step": 175
    },
    {
      "epoch": 1.6216216216216215,
      "grad_norm": 2.3125,
      "learning_rate": 0.00019996283741405458,
      "loss": 0.9908,
      "step": 180
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 28.375,
      "learning_rate": 0.00019992875743301967,
      "loss": 1.0038,
      "step": 185
    },
    {
      "epoch": 1.7117117117117115,
      "grad_norm": 2.390625,
      "learning_rate": 0.00019988368985718576,
      "loss": 0.9688,
      "step": 190
    },
    {
      "epoch": 1.7567567567567568,
      "grad_norm": 1.75,
      "learning_rate": 0.00019982763964192585,
      "loss": 0.9638,
      "step": 195
    },
    {
      "epoch": 1.8018018018018018,
      "grad_norm": 3.125,
      "learning_rate": 0.00019976061295020106,
      "loss": 0.9878,
      "step": 200
    },
    {
      "epoch": 1.8468468468468469,
      "grad_norm": 1.921875,
      "learning_rate": 0.00019968261715188324,
      "loss": 0.9394,
      "step": 205
    },
    {
      "epoch": 1.8918918918918919,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019959366082294438,
      "loss": 0.9285,
      "step": 210
    },
    {
      "epoch": 1.936936936936937,
      "grad_norm": 3.53125,
      "learning_rate": 0.00019949375374451377,
      "loss": 0.929,
      "step": 215
    },
    {
      "epoch": 1.981981981981982,
      "grad_norm": 1.625,
      "learning_rate": 0.00019938290690180247,
      "loss": 0.9374,
      "step": 220
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.192767381668091,
      "eval_runtime": 1.0169,
      "eval_samples_per_second": 4.917,
      "eval_steps_per_second": 1.967,
      "step": 222
    },
    {
      "epoch": 2.027027027027027,
      "grad_norm": 2.25,
      "learning_rate": 0.0001992611324828956,
      "loss": 0.8849,
      "step": 225
    },
    {
      "epoch": 2.0720720720720722,
      "grad_norm": 1.8125,
      "learning_rate": 0.00019912844387741195,
      "loss": 0.864,
      "step": 230
    },
    {
      "epoch": 2.1171171171171173,
      "grad_norm": 1.09375,
      "learning_rate": 0.00019898485567503187,
      "loss": 0.8497,
      "step": 235
    },
    {
      "epoch": 2.1621621621621623,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00019883038366389313,
      "loss": 0.8844,
      "step": 240
    },
    {
      "epoch": 2.2072072072072073,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0001986650448288548,
      "loss": 0.8509,
      "step": 245
    },
    {
      "epoch": 2.2522522522522523,
      "grad_norm": 3.078125,
      "learning_rate": 0.00019848885734962982,
      "loss": 0.8718,
      "step": 250
    },
    {
      "epoch": 2.2972972972972974,
      "grad_norm": 3.390625,
      "learning_rate": 0.00019830184059878606,
      "loss": 0.881,
      "step": 255
    },
    {
      "epoch": 2.3423423423423424,
      "grad_norm": 2.75,
      "learning_rate": 0.0001981040151396161,
      "loss": 0.854,
      "step": 260
    },
    {
      "epoch": 2.3873873873873874,
      "grad_norm": 1.25,
      "learning_rate": 0.0001978954027238763,
      "loss": 0.8856,
      "step": 265
    },
    {
      "epoch": 2.4324324324324325,
      "grad_norm": 1.640625,
      "learning_rate": 0.00019767602628939507,
      "loss": 0.8797,
      "step": 270
    },
    {
      "epoch": 2.4774774774774775,
      "grad_norm": 2.28125,
      "learning_rate": 0.00019744590995755088,
      "loss": 0.8829,
      "step": 275
    },
    {
      "epoch": 2.5225225225225225,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019720507903061974,
      "loss": 0.8319,
      "step": 280
    },
    {
      "epoch": 2.5675675675675675,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00019695355998899345,
      "loss": 0.8489,
      "step": 285
    },
    {
      "epoch": 2.6126126126126126,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019669138048826766,
      "loss": 0.8409,
      "step": 290
    },
    {
      "epoch": 2.6576576576576576,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00019641856935620117,
      "loss": 0.8454,
      "step": 295
    },
    {
      "epoch": 2.7027027027027026,
      "grad_norm": 0.625,
      "learning_rate": 0.00019613515658954624,
      "loss": 0.8501,
      "step": 300
    },
    {
      "epoch": 2.7477477477477477,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001958411733507502,
      "loss": 0.8223,
      "step": 305
    },
    {
      "epoch": 2.7927927927927927,
      "grad_norm": 0.875,
      "learning_rate": 0.00019553665196452903,
      "loss": 0.8396,
      "step": 310
    },
    {
      "epoch": 2.8378378378378377,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001952216259143132,
      "loss": 0.836,
      "step": 315
    },
    {
      "epoch": 2.8828828828828827,
      "grad_norm": 2.09375,
      "learning_rate": 0.0001948961298385659,
      "loss": 0.8453,
      "step": 320
    },
    {
      "epoch": 2.9279279279279278,
      "grad_norm": 0.62109375,
      "learning_rate": 0.00019456019952697462,
      "loss": 0.824,
      "step": 325
    },
    {
      "epoch": 2.972972972972973,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0001942138719165156,
      "loss": 0.8471,
      "step": 330
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.168168544769287,
      "eval_runtime": 1.0188,
      "eval_samples_per_second": 4.908,
      "eval_steps_per_second": 1.963,
      "step": 333
    },
    {
      "epoch": 3.018018018018018,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019385718508739262,
      "loss": 0.8242,
      "step": 335
    },
    {
      "epoch": 3.063063063063063,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019349017825885,
      "loss": 0.7439,
      "step": 340
    },
    {
      "epoch": 3.108108108108108,
      "grad_norm": 0.67578125,
      "learning_rate": 0.00019311289178486018,
      "loss": 0.7545,
      "step": 345
    },
    {
      "epoch": 3.153153153153153,
      "grad_norm": 1.71875,
      "learning_rate": 0.00019272536714968658,
      "loss": 0.7531,
      "step": 350
    },
    {
      "epoch": 3.1981981981981984,
      "grad_norm": 2.15625,
      "learning_rate": 0.0001923276469633223,
      "loss": 0.7731,
      "step": 355
    },
    {
      "epoch": 3.2432432432432434,
      "grad_norm": 1.453125,
      "learning_rate": 0.000191919774956805,
      "loss": 0.7558,
      "step": 360
    },
    {
      "epoch": 3.2882882882882885,
      "grad_norm": 1.71875,
      "learning_rate": 0.0001915017959774084,
      "loss": 0.782,
      "step": 365
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.7265625,
      "learning_rate": 0.00019107375598371112,
      "loss": 0.7795,
      "step": 370
    },
    {
      "epoch": 3.3783783783783785,
      "grad_norm": 1.71875,
      "learning_rate": 0.0001906357020405435,
      "loss": 0.7638,
      "step": 375
    },
    {
      "epoch": 3.4234234234234235,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00019018768231381238,
      "loss": 0.7511,
      "step": 380
    },
    {
      "epoch": 3.4684684684684686,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0001897297460652053,
      "loss": 0.7591,
      "step": 385
    },
    {
      "epoch": 3.5135135135135136,
      "grad_norm": 7.625,
      "learning_rate": 0.0001892619436467738,
      "loss": 0.7433,
      "step": 390
    },
    {
      "epoch": 3.5585585585585586,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018878432649539696,
      "loss": 0.7673,
      "step": 395
    },
    {
      "epoch": 3.6036036036036037,
      "grad_norm": 0.734375,
      "learning_rate": 0.0001882969471271259,
      "loss": 0.7633,
      "step": 400
    },
    {
      "epoch": 3.6486486486486487,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00018779985913140924,
      "loss": 0.7751,
      "step": 405
    },
    {
      "epoch": 3.6936936936936937,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00018729311716520074,
      "loss": 0.7628,
      "step": 410
    },
    {
      "epoch": 3.7387387387387387,
      "grad_norm": 1.09375,
      "learning_rate": 0.00018677677694694958,
      "loss": 0.7553,
      "step": 415
    },
    {
      "epoch": 3.7837837837837838,
      "grad_norm": 0.7578125,
      "learning_rate": 0.00018625089525047385,
      "loss": 0.7636,
      "step": 420
    },
    {
      "epoch": 3.828828828828829,
      "grad_norm": 0.76953125,
      "learning_rate": 0.00018571552989871806,
      "loss": 0.7764,
      "step": 425
    },
    {
      "epoch": 3.873873873873874,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00018517073975739514,
      "loss": 0.7715,
      "step": 430
    },
    {
      "epoch": 3.918918918918919,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0001846165847285141,
      "loss": 0.77,
      "step": 435
    },
    {
      "epoch": 3.963963963963964,
      "grad_norm": 3.0,
      "learning_rate": 0.0001840531257437934,
      "loss": 0.7873,
      "step": 440
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.2036337852478027,
      "eval_runtime": 1.0177,
      "eval_samples_per_second": 4.913,
      "eval_steps_per_second": 1.965,
      "step": 444
    },
    {
      "epoch": 4.009009009009009,
      "grad_norm": 2.0625,
      "learning_rate": 0.00018348042475796122,
      "loss": 0.7489,
      "step": 445
    },
    {
      "epoch": 4.054054054054054,
      "grad_norm": 0.6484375,
      "learning_rate": 0.00018289854474194347,
      "loss": 0.6615,
      "step": 450
    },
    {
      "epoch": 4.099099099099099,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018230754967593963,
      "loss": 0.6595,
      "step": 455
    },
    {
      "epoch": 4.1441441441441444,
      "grad_norm": 0.7578125,
      "learning_rate": 0.00018170750454238793,
      "loss": 0.6718,
      "step": 460
    },
    {
      "epoch": 4.1891891891891895,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00018109847531882047,
      "loss": 0.6528,
      "step": 465
    },
    {
      "epoch": 4.2342342342342345,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001804805289706083,
      "loss": 0.6505,
      "step": 470
    },
    {
      "epoch": 4.2792792792792795,
      "grad_norm": 1.25,
      "learning_rate": 0.0001798537334435986,
      "loss": 0.6708,
      "step": 475
    },
    {
      "epoch": 4.324324324324325,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0001792181576566437,
      "loss": 0.6702,
      "step": 480
    },
    {
      "epoch": 4.36936936936937,
      "grad_norm": 0.734375,
      "learning_rate": 0.0001785738714940231,
      "loss": 0.6755,
      "step": 485
    },
    {
      "epoch": 4.414414414414415,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017792094579775926,
      "loss": 0.6825,
      "step": 490
    },
    {
      "epoch": 4.45945945945946,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00017725945235982856,
      "loss": 0.6864,
      "step": 495
    },
    {
      "epoch": 4.504504504504505,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0001765894639142671,
      "loss": 0.6758,
      "step": 500
    },
    {
      "epoch": 4.54954954954955,
      "grad_norm": 0.71875,
      "learning_rate": 0.0001759110541291736,
      "loss": 0.6823,
      "step": 505
    },
    {
      "epoch": 4.594594594594595,
      "grad_norm": 0.6875,
      "learning_rate": 0.000175224297598609,
      "loss": 0.6882,
      "step": 510
    },
    {
      "epoch": 4.63963963963964,
      "grad_norm": 0.65234375,
      "learning_rate": 0.00017452926983439475,
      "loss": 0.6922,
      "step": 515
    },
    {
      "epoch": 4.684684684684685,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0001738260472578097,
      "loss": 0.6787,
      "step": 520
    },
    {
      "epoch": 4.72972972972973,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00017311470719118765,
      "loss": 0.7028,
      "step": 525
    },
    {
      "epoch": 4.774774774774775,
      "grad_norm": 0.80859375,
      "learning_rate": 0.00017239532784941495,
      "loss": 0.6851,
      "step": 530
    },
    {
      "epoch": 4.81981981981982,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001716679883313308,
      "loss": 0.6756,
      "step": 535
    },
    {
      "epoch": 4.864864864864865,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017093276861102988,
      "loss": 0.6697,
      "step": 540
    },
    {
      "epoch": 4.90990990990991,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017018974952906884,
      "loss": 0.689,
      "step": 545
    },
    {
      "epoch": 4.954954954954955,
      "grad_norm": 0.68359375,
      "learning_rate": 0.0001694390127835775,
      "loss": 0.6894,
      "step": 550
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016868064092127587,
      "loss": 0.685,
      "step": 555
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.297661304473877,
      "eval_runtime": 1.0166,
      "eval_samples_per_second": 4.918,
      "eval_steps_per_second": 1.967,
      "step": 555
    },
    {
      "epoch": 5.045045045045045,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016791471732839776,
      "loss": 0.5743,
      "step": 560
    },
    {
      "epoch": 5.09009009009009,
      "grad_norm": 0.7734375,
      "learning_rate": 0.00016714132622152206,
      "loss": 0.5415,
      "step": 565
    },
    {
      "epoch": 5.135135135135135,
      "grad_norm": 0.78515625,
      "learning_rate": 0.00016636055263831285,
      "loss": 0.5752,
      "step": 570
    },
    {
      "epoch": 5.18018018018018,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00016557248242816903,
      "loss": 0.5778,
      "step": 575
    },
    {
      "epoch": 5.225225225225225,
      "grad_norm": 1.015625,
      "learning_rate": 0.00016477720224278492,
      "loss": 0.5924,
      "step": 580
    },
    {
      "epoch": 5.27027027027027,
      "grad_norm": 0.8046875,
      "learning_rate": 0.00016397479952662248,
      "loss": 0.5683,
      "step": 585
    },
    {
      "epoch": 5.315315315315315,
      "grad_norm": 0.734375,
      "learning_rate": 0.0001631653625072965,
      "loss": 0.595,
      "step": 590
    },
    {
      "epoch": 5.36036036036036,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00016234898018587337,
      "loss": 0.5862,
      "step": 595
    },
    {
      "epoch": 5.405405405405405,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016152574232708534,
      "loss": 0.5806,
      "step": 600
    },
    {
      "epoch": 5.45045045045045,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016069573944946025,
      "loss": 0.6002,
      "step": 605
    },
    {
      "epoch": 5.495495495495495,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015985906281536875,
      "loss": 0.6061,
      "step": 610
    },
    {
      "epoch": 5.54054054054054,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00015901580442098968,
      "loss": 0.6029,
      "step": 615
    },
    {
      "epoch": 5.585585585585585,
      "grad_norm": 0.78515625,
      "learning_rate": 0.00015816605698619452,
      "loss": 0.5985,
      "step": 620
    },
    {
      "epoch": 5.63063063063063,
      "grad_norm": 0.73046875,
      "learning_rate": 0.0001573099139443525,
      "loss": 0.5863,
      "step": 625
    },
    {
      "epoch": 5.675675675675675,
      "grad_norm": 0.84375,
      "learning_rate": 0.00015644746943205734,
      "loss": 0.5927,
      "step": 630
    },
    {
      "epoch": 5.7207207207207205,
      "grad_norm": 0.796875,
      "learning_rate": 0.00015557881827877633,
      "loss": 0.6072,
      "step": 635
    },
    {
      "epoch": 5.7657657657657655,
      "grad_norm": 0.76171875,
      "learning_rate": 0.0001547040559964236,
      "loss": 0.6032,
      "step": 640
    },
    {
      "epoch": 5.8108108108108105,
      "grad_norm": 0.7109375,
      "learning_rate": 0.00015382327876885797,
      "loss": 0.6135,
      "step": 645
    },
    {
      "epoch": 5.8558558558558556,
      "grad_norm": 0.7265625,
      "learning_rate": 0.00015293658344130734,
      "loss": 0.6058,
      "step": 650
    },
    {
      "epoch": 5.900900900900901,
      "grad_norm": 0.69140625,
      "learning_rate": 0.00015204406750972,
      "loss": 0.5948,
      "step": 655
    },
    {
      "epoch": 5.945945945945946,
      "grad_norm": 0.65234375,
      "learning_rate": 0.00015114582911004466,
      "loss": 0.6108,
      "step": 660
    },
    {
      "epoch": 5.990990990990991,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0001502419670074398,
      "loss": 0.6223,
      "step": 665
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.4441299438476562,
      "eval_runtime": 1.0183,
      "eval_samples_per_second": 4.91,
      "eval_steps_per_second": 1.964,
      "step": 666
    },
    {
      "epoch": 6.036036036036036,
      "grad_norm": 1.109375,
      "learning_rate": 0.00014933258058541415,
      "loss": 0.5455,
      "step": 670
    },
    {
      "epoch": 6.081081081081081,
      "grad_norm": 0.703125,
      "learning_rate": 0.000148417769834899,
      "loss": 0.5079,
      "step": 675
    },
    {
      "epoch": 6.126126126126126,
      "grad_norm": 0.9375,
      "learning_rate": 0.0001474976353432539,
      "loss": 0.4906,
      "step": 680
    },
    {
      "epoch": 6.171171171171171,
      "grad_norm": 0.96875,
      "learning_rate": 0.00014657227828320635,
      "loss": 0.5065,
      "step": 685
    },
    {
      "epoch": 6.216216216216216,
      "grad_norm": 0.75,
      "learning_rate": 0.0001456418004017278,
      "loss": 0.4978,
      "step": 690
    },
    {
      "epoch": 6.261261261261261,
      "grad_norm": 0.75390625,
      "learning_rate": 0.00014470630400884575,
      "loss": 0.4966,
      "step": 695
    },
    {
      "epoch": 6.306306306306306,
      "grad_norm": 0.7578125,
      "learning_rate": 0.00014376589196639467,
      "loss": 0.4981,
      "step": 700
    },
    {
      "epoch": 6.351351351351352,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00014282066767670572,
      "loss": 0.5238,
      "step": 705
    },
    {
      "epoch": 6.396396396396397,
      "grad_norm": 0.66015625,
      "learning_rate": 0.0001418707350712372,
      "loss": 0.5052,
      "step": 710
    },
    {
      "epoch": 6.441441441441442,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00014091619859914694,
      "loss": 0.5045,
      "step": 715
    },
    {
      "epoch": 6.486486486486487,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0001399571632158076,
      "loss": 0.5169,
      "step": 720
    },
    {
      "epoch": 6.531531531531532,
      "grad_norm": 0.703125,
      "learning_rate": 0.00013899373437126637,
      "loss": 0.5208,
      "step": 725
    },
    {
      "epoch": 6.576576576576577,
      "grad_norm": 0.66015625,
      "learning_rate": 0.0001380260179986504,
      "loss": 0.5192,
      "step": 730
    },
    {
      "epoch": 6.621621621621622,
      "grad_norm": 0.77734375,
      "learning_rate": 0.00013705412050251883,
      "loss": 0.526,
      "step": 735
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 0.84375,
      "learning_rate": 0.00013607814874716325,
      "loss": 0.541,
      "step": 740
    },
    {
      "epoch": 6.711711711711712,
      "grad_norm": 0.875,
      "learning_rate": 0.0001350982100448575,
      "loss": 0.5119,
      "step": 745
    },
    {
      "epoch": 6.756756756756757,
      "grad_norm": 0.7578125,
      "learning_rate": 0.00013411441214405829,
      "loss": 0.5192,
      "step": 750
    },
    {
      "epoch": 6.801801801801802,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00013312686321755761,
      "loss": 0.5353,
      "step": 755
    },
    {
      "epoch": 6.846846846846847,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00013213567185058888,
      "loss": 0.5286,
      "step": 760
    },
    {
      "epoch": 6.891891891891892,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0001311409470288874,
      "loss": 0.5286,
      "step": 765
    },
    {
      "epoch": 6.936936936936937,
      "grad_norm": 0.796875,
      "learning_rate": 0.00013014279812670703,
      "loss": 0.5353,
      "step": 770
    },
    {
      "epoch": 6.981981981981982,
      "grad_norm": 0.6796875,
      "learning_rate": 0.00012914133489479385,
      "loss": 0.5378,
      "step": 775
    },
    {
      "epoch": 7.0,
      "eval_loss": 2.6714916229248047,
      "eval_runtime": 1.0156,
      "eval_samples_per_second": 4.923,
      "eval_steps_per_second": 1.969,
      "step": 777
    },
    {
      "epoch": 7.027027027027027,
      "grad_norm": 0.74609375,
      "learning_rate": 0.0001281366674483187,
      "loss": 0.4628,
      "step": 780
    },
    {
      "epoch": 7.072072072072072,
      "grad_norm": 0.75,
      "learning_rate": 0.00012712890625476953,
      "loss": 0.4205,
      "step": 785
    },
    {
      "epoch": 7.117117117117117,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0001261181621218051,
      "loss": 0.4286,
      "step": 790
    },
    {
      "epoch": 7.162162162162162,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00012510454618507106,
      "loss": 0.4393,
      "step": 795
    },
    {
      "epoch": 7.207207207207207,
      "grad_norm": 0.765625,
      "learning_rate": 0.0001240881698959802,
      "loss": 0.4276,
      "step": 800
    },
    {
      "epoch": 7.252252252252252,
      "grad_norm": 0.72265625,
      "learning_rate": 0.00012306914500945788,
      "loss": 0.4322,
      "step": 805
    },
    {
      "epoch": 7.297297297297297,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00012204758357165409,
      "loss": 0.442,
      "step": 810
    },
    {
      "epoch": 7.342342342342342,
      "grad_norm": 0.7578125,
      "learning_rate": 0.00012102359790762347,
      "loss": 0.4401,
      "step": 815
    },
    {
      "epoch": 7.387387387387387,
      "grad_norm": 0.73828125,
      "learning_rate": 0.00011999730060897476,
      "loss": 0.435,
      "step": 820
    },
    {
      "epoch": 7.4324324324324325,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00011896880452149077,
      "loss": 0.4448,
      "step": 825
    },
    {
      "epoch": 7.4774774774774775,
      "grad_norm": 0.8203125,
      "learning_rate": 0.00011793822273272066,
      "loss": 0.4448,
      "step": 830
    },
    {
      "epoch": 7.5225225225225225,
      "grad_norm": 0.78125,
      "learning_rate": 0.00011690566855954523,
      "loss": 0.4414,
      "step": 835
    },
    {
      "epoch": 7.5675675675675675,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0001158712555357175,
      "loss": 0.4461,
      "step": 840
    },
    {
      "epoch": 7.612612612612613,
      "grad_norm": 0.78515625,
      "learning_rate": 0.00011483509739937903,
      "loss": 0.4455,
      "step": 845
    },
    {
      "epoch": 7.657657657657658,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000113797308080554,
      "loss": 0.4521,
      "step": 850
    },
    {
      "epoch": 7.702702702702703,
      "grad_norm": 0.81640625,
      "learning_rate": 0.00011275800168862214,
      "loss": 0.4446,
      "step": 855
    },
    {
      "epoch": 7.747747747747748,
      "grad_norm": 0.76171875,
      "learning_rate": 0.00011171729249977169,
      "loss": 0.4333,
      "step": 860
    },
    {
      "epoch": 7.792792792792793,
      "grad_norm": 0.7734375,
      "learning_rate": 0.00011067529494443463,
      "loss": 0.4494,
      "step": 865
    },
    {
      "epoch": 7.837837837837838,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00010963212359470421,
      "loss": 0.4423,
      "step": 870
    },
    {
      "epoch": 7.882882882882883,
      "grad_norm": 1.15625,
      "learning_rate": 0.00010858789315173745,
      "loss": 0.4424,
      "step": 875
    },
    {
      "epoch": 7.927927927927928,
      "grad_norm": 0.76953125,
      "learning_rate": 0.00010754271843314325,
      "loss": 0.4491,
      "step": 880
    },
    {
      "epoch": 7.972972972972973,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00010649671436035753,
      "loss": 0.458,
      "step": 885
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.955517292022705,
      "eval_runtime": 1.015,
      "eval_samples_per_second": 4.926,
      "eval_steps_per_second": 1.97,
      "step": 888
    },
    {
      "epoch": 8.018018018018019,
      "grad_norm": 0.90625,
      "learning_rate": 0.0001054499959460074,
      "loss": 0.4181,
      "step": 890
    },
    {
      "epoch": 8.063063063063064,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00010440267828126478,
      "loss": 0.3664,
      "step": 895
    },
    {
      "epoch": 8.108108108108109,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00010335487652319182,
      "loss": 0.3723,
      "step": 900
    },
    {
      "epoch": 8.153153153153154,
      "grad_norm": 0.734375,
      "learning_rate": 0.00010230670588207873,
      "loss": 0.3588,
      "step": 905
    },
    {
      "epoch": 8.198198198198199,
      "grad_norm": 0.859375,
      "learning_rate": 0.00010125828160877604,
      "loss": 0.3619,
      "step": 910
    },
    {
      "epoch": 8.243243243243244,
      "grad_norm": 0.765625,
      "learning_rate": 0.00010020971898202218,
      "loss": 0.3651,
      "step": 915
    },
    {
      "epoch": 8.288288288288289,
      "grad_norm": 0.90625,
      "learning_rate": 9.916113329576818e-05,
      "loss": 0.3742,
      "step": 920
    },
    {
      "epoch": 8.333333333333334,
      "grad_norm": 0.8359375,
      "learning_rate": 9.811263984650056e-05,
      "loss": 0.3752,
      "step": 925
    },
    {
      "epoch": 8.378378378378379,
      "grad_norm": 1.015625,
      "learning_rate": 9.706435392056394e-05,
      "loss": 0.3746,
      "step": 930
    },
    {
      "epoch": 8.423423423423424,
      "grad_norm": 0.8046875,
      "learning_rate": 9.601639078148489e-05,
      "loss": 0.3744,
      "step": 935
    },
    {
      "epoch": 8.468468468468469,
      "grad_norm": 0.82421875,
      "learning_rate": 9.496886565729823e-05,
      "loss": 0.3765,
      "step": 940
    },
    {
      "epoch": 8.513513513513514,
      "grad_norm": 0.73828125,
      "learning_rate": 9.392189372787706e-05,
      "loss": 0.371,
      "step": 945
    },
    {
      "epoch": 8.558558558558559,
      "grad_norm": 0.796875,
      "learning_rate": 9.287559011226843e-05,
      "loss": 0.3747,
      "step": 950
    },
    {
      "epoch": 8.603603603603604,
      "grad_norm": 0.80859375,
      "learning_rate": 9.183006985603545e-05,
      "loss": 0.372,
      "step": 955
    },
    {
      "epoch": 8.64864864864865,
      "grad_norm": 0.7578125,
      "learning_rate": 9.078544791860739e-05,
      "loss": 0.3764,
      "step": 960
    },
    {
      "epoch": 8.693693693693694,
      "grad_norm": 0.81640625,
      "learning_rate": 8.974183916063968e-05,
      "loss": 0.375,
      "step": 965
    },
    {
      "epoch": 8.73873873873874,
      "grad_norm": 0.75,
      "learning_rate": 8.869935833138414e-05,
      "loss": 0.3836,
      "step": 970
    },
    {
      "epoch": 8.783783783783784,
      "grad_norm": 0.75390625,
      "learning_rate": 8.765812005607217e-05,
      "loss": 0.3804,
      "step": 975
    },
    {
      "epoch": 8.82882882882883,
      "grad_norm": 0.78515625,
      "learning_rate": 8.6618238823311e-05,
      "loss": 0.3857,
      "step": 980
    },
    {
      "epoch": 8.873873873873874,
      "grad_norm": 0.81640625,
      "learning_rate": 8.557982897249513e-05,
      "loss": 0.383,
      "step": 985
    },
    {
      "epoch": 8.91891891891892,
      "grad_norm": 0.8671875,
      "learning_rate": 8.45430046812344e-05,
      "loss": 0.3831,
      "step": 990
    },
    {
      "epoch": 8.963963963963964,
      "grad_norm": 0.83203125,
      "learning_rate": 8.35078799527996e-05,
      "loss": 0.3843,
      "step": 995
    },
    {
      "epoch": 9.0,
      "eval_loss": 3.43645977973938,
      "eval_runtime": 1.0159,
      "eval_samples_per_second": 4.922,
      "eval_steps_per_second": 1.969,
      "step": 999
    },
    {
      "epoch": 9.00900900900901,
      "grad_norm": 0.734375,
      "learning_rate": 8.247456860358725e-05,
      "loss": 0.3691,
      "step": 1000
    },
    {
      "epoch": 9.054054054054054,
      "grad_norm": 1.0546875,
      "learning_rate": 8.144318425060509e-05,
      "loss": 0.32,
      "step": 1005
    },
    {
      "epoch": 9.0990990990991,
      "grad_norm": 0.7890625,
      "learning_rate": 8.04138402989794e-05,
      "loss": 0.3192,
      "step": 1010
    },
    {
      "epoch": 9.144144144144144,
      "grad_norm": 0.88671875,
      "learning_rate": 7.938664992948549e-05,
      "loss": 0.3166,
      "step": 1015
    },
    {
      "epoch": 9.18918918918919,
      "grad_norm": 0.72265625,
      "learning_rate": 7.83617260861032e-05,
      "loss": 0.3123,
      "step": 1020
    },
    {
      "epoch": 9.234234234234235,
      "grad_norm": 0.87890625,
      "learning_rate": 7.733918146359815e-05,
      "loss": 0.3192,
      "step": 1025
    },
    {
      "epoch": 9.27927927927928,
      "grad_norm": 0.8046875,
      "learning_rate": 7.631912849513025e-05,
      "loss": 0.3127,
      "step": 1030
    },
    {
      "epoch": 9.324324324324325,
      "grad_norm": 0.765625,
      "learning_rate": 7.530167933989161e-05,
      "loss": 0.3149,
      "step": 1035
    },
    {
      "epoch": 9.36936936936937,
      "grad_norm": 0.7890625,
      "learning_rate": 7.428694587077371e-05,
      "loss": 0.318,
      "step": 1040
    },
    {
      "epoch": 9.414414414414415,
      "grad_norm": 0.8125,
      "learning_rate": 7.32750396620669e-05,
      "loss": 0.3136,
      "step": 1045
    },
    {
      "epoch": 9.45945945945946,
      "grad_norm": 0.859375,
      "learning_rate": 7.226607197719202e-05,
      "loss": 0.3193,
      "step": 1050
    },
    {
      "epoch": 9.504504504504505,
      "grad_norm": 0.75390625,
      "learning_rate": 7.126015375646666e-05,
      "loss": 0.319,
      "step": 1055
    },
    {
      "epoch": 9.54954954954955,
      "grad_norm": 0.75,
      "learning_rate": 7.025739560490675e-05,
      "loss": 0.3246,
      "step": 1060
    },
    {
      "epoch": 9.594594594594595,
      "grad_norm": 0.79296875,
      "learning_rate": 6.925790778006518e-05,
      "loss": 0.3223,
      "step": 1065
    },
    {
      "epoch": 9.63963963963964,
      "grad_norm": 0.77734375,
      "learning_rate": 6.826180017990828e-05,
      "loss": 0.3238,
      "step": 1070
    },
    {
      "epoch": 9.684684684684685,
      "grad_norm": 0.80078125,
      "learning_rate": 6.726918233073231e-05,
      "loss": 0.321,
      "step": 1075
    },
    {
      "epoch": 9.72972972972973,
      "grad_norm": 0.8203125,
      "learning_rate": 6.62801633751204e-05,
      "loss": 0.3161,
      "step": 1080
    },
    {
      "epoch": 9.774774774774775,
      "grad_norm": 0.83203125,
      "learning_rate": 6.52948520599419e-05,
      "loss": 0.3221,
      "step": 1085
    },
    {
      "epoch": 9.81981981981982,
      "grad_norm": 0.7734375,
      "learning_rate": 6.431335672439531e-05,
      "loss": 0.3251,
      "step": 1090
    },
    {
      "epoch": 9.864864864864865,
      "grad_norm": 0.7890625,
      "learning_rate": 6.333578528809574e-05,
      "loss": 0.3246,
      "step": 1095
    },
    {
      "epoch": 9.90990990990991,
      "grad_norm": 0.79296875,
      "learning_rate": 6.236224523920879e-05,
      "loss": 0.3238,
      "step": 1100
    },
    {
      "epoch": 9.954954954954955,
      "grad_norm": 0.765625,
      "learning_rate": 6.139284362263185e-05,
      "loss": 0.3194,
      "step": 1105
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.7109375,
      "learning_rate": 6.042768702822381e-05,
      "loss": 0.3241,
      "step": 1110
    },
    {
      "epoch": 10.0,
      "eval_loss": 3.8823204040527344,
      "eval_runtime": 1.0143,
      "eval_samples_per_second": 4.929,
      "eval_steps_per_second": 1.972,
      "step": 1110
    },
    {
      "epoch": 10.045045045045045,
      "grad_norm": 0.7890625,
      "learning_rate": 5.9466881579085275e-05,
      "loss": 0.2776,
      "step": 1115
    },
    {
      "epoch": 10.09009009009009,
      "grad_norm": 0.7890625,
      "learning_rate": 5.8510532919889804e-05,
      "loss": 0.2767,
      "step": 1120
    },
    {
      "epoch": 10.135135135135135,
      "grad_norm": 0.69140625,
      "learning_rate": 5.7558746205267756e-05,
      "loss": 0.2734,
      "step": 1125
    },
    {
      "epoch": 10.18018018018018,
      "grad_norm": 0.8203125,
      "learning_rate": 5.6611626088244194e-05,
      "loss": 0.2792,
      "step": 1130
    },
    {
      "epoch": 10.225225225225225,
      "grad_norm": 0.734375,
      "learning_rate": 5.56692767087318e-05,
      "loss": 0.2753,
      "step": 1135
    },
    {
      "epoch": 10.27027027027027,
      "grad_norm": 0.70703125,
      "learning_rate": 5.4731801682080206e-05,
      "loss": 0.2763,
      "step": 1140
    },
    {
      "epoch": 10.315315315315315,
      "grad_norm": 0.7265625,
      "learning_rate": 5.3799304087683146e-05,
      "loss": 0.2772,
      "step": 1145
    },
    {
      "epoch": 10.36036036036036,
      "grad_norm": 0.79296875,
      "learning_rate": 5.28718864576442e-05,
      "loss": 0.2763,
      "step": 1150
    },
    {
      "epoch": 10.405405405405405,
      "grad_norm": 0.75,
      "learning_rate": 5.194965076550323e-05,
      "loss": 0.2766,
      "step": 1155
    },
    {
      "epoch": 10.45045045045045,
      "grad_norm": 0.84375,
      "learning_rate": 5.1032698415023963e-05,
      "loss": 0.2802,
      "step": 1160
    },
    {
      "epoch": 10.495495495495495,
      "grad_norm": 0.8203125,
      "learning_rate": 5.0121130229043786e-05,
      "loss": 0.276,
      "step": 1165
    },
    {
      "epoch": 10.54054054054054,
      "grad_norm": 0.79296875,
      "learning_rate": 4.921504643838839e-05,
      "loss": 0.2801,
      "step": 1170
    },
    {
      "epoch": 10.585585585585585,
      "grad_norm": 0.7109375,
      "learning_rate": 4.8314546670850594e-05,
      "loss": 0.2775,
      "step": 1175
    },
    {
      "epoch": 10.63063063063063,
      "grad_norm": 0.7421875,
      "learning_rate": 4.741972994023601e-05,
      "loss": 0.2806,
      "step": 1180
    },
    {
      "epoch": 10.675675675675675,
      "grad_norm": 0.7421875,
      "learning_rate": 4.6530694635476e-05,
      "loss": 0.2775,
      "step": 1185
    },
    {
      "epoch": 10.72072072072072,
      "grad_norm": 0.7421875,
      "learning_rate": 4.5647538509809416e-05,
      "loss": 0.2802,
      "step": 1190
    },
    {
      "epoch": 10.765765765765765,
      "grad_norm": 0.73828125,
      "learning_rate": 4.477035867003405e-05,
      "loss": 0.2751,
      "step": 1195
    },
    {
      "epoch": 10.81081081081081,
      "grad_norm": 0.7734375,
      "learning_rate": 4.38992515658297e-05,
      "loss": 0.2802,
      "step": 1200
    },
    {
      "epoch": 10.855855855855856,
      "grad_norm": 0.79296875,
      "learning_rate": 4.303431297915277e-05,
      "loss": 0.2748,
      "step": 1205
    },
    {
      "epoch": 10.9009009009009,
      "grad_norm": 0.77734375,
      "learning_rate": 4.2175638013704655e-05,
      "loss": 0.2808,
      "step": 1210
    },
    {
      "epoch": 10.945945945945946,
      "grad_norm": 0.7421875,
      "learning_rate": 4.1323321084474896e-05,
      "loss": 0.2795,
      "step": 1215
    },
    {
      "epoch": 10.99099099099099,
      "grad_norm": 0.76953125,
      "learning_rate": 4.0477455907359715e-05,
      "loss": 0.2825,
      "step": 1220
    },
    {
      "epoch": 11.0,
      "eval_loss": 4.404350757598877,
      "eval_runtime": 1.0163,
      "eval_samples_per_second": 4.92,
      "eval_steps_per_second": 1.968,
      "step": 1221
    },
    {
      "epoch": 11.036036036036036,
      "grad_norm": 0.609375,
      "learning_rate": 3.963813548885751e-05,
      "loss": 0.2565,
      "step": 1225
    },
    {
      "epoch": 11.08108108108108,
      "grad_norm": 0.7109375,
      "learning_rate": 3.880545211584251e-05,
      "loss": 0.2458,
      "step": 1230
    },
    {
      "epoch": 11.126126126126126,
      "grad_norm": 0.6796875,
      "learning_rate": 3.7979497345417234e-05,
      "loss": 0.257,
      "step": 1235
    },
    {
      "epoch": 11.17117117117117,
      "grad_norm": 0.6640625,
      "learning_rate": 3.716036199484574e-05,
      "loss": 0.2449,
      "step": 1240
    },
    {
      "epoch": 11.216216216216216,
      "grad_norm": 0.83203125,
      "learning_rate": 3.634813613156753e-05,
      "loss": 0.2529,
      "step": 1245
    },
    {
      "epoch": 11.26126126126126,
      "grad_norm": 0.7421875,
      "learning_rate": 3.554290906329438e-05,
      "loss": 0.2519,
      "step": 1250
    },
    {
      "epoch": 11.306306306306306,
      "grad_norm": 0.734375,
      "learning_rate": 3.474476932819062e-05,
      "loss": 0.2551,
      "step": 1255
    },
    {
      "epoch": 11.35135135135135,
      "grad_norm": 0.69140625,
      "learning_rate": 3.395380468513805e-05,
      "loss": 0.2492,
      "step": 1260
    },
    {
      "epoch": 11.396396396396396,
      "grad_norm": 0.703125,
      "learning_rate": 3.317010210408617e-05,
      "loss": 0.2501,
      "step": 1265
    },
    {
      "epoch": 11.441441441441441,
      "grad_norm": 0.734375,
      "learning_rate": 3.2393747756489845e-05,
      "loss": 0.2493,
      "step": 1270
    },
    {
      "epoch": 11.486486486486486,
      "grad_norm": 0.70703125,
      "learning_rate": 3.162482700583402e-05,
      "loss": 0.2475,
      "step": 1275
    },
    {
      "epoch": 11.531531531531531,
      "grad_norm": 0.7578125,
      "learning_rate": 3.0863424398248064e-05,
      "loss": 0.2524,
      "step": 1280
    },
    {
      "epoch": 11.576576576576576,
      "grad_norm": 0.6953125,
      "learning_rate": 3.0109623653209163e-05,
      "loss": 0.2489,
      "step": 1285
    },
    {
      "epoch": 11.621621621621621,
      "grad_norm": 0.72265625,
      "learning_rate": 2.936350765433713e-05,
      "loss": 0.2507,
      "step": 1290
    },
    {
      "epoch": 11.666666666666666,
      "grad_norm": 0.68359375,
      "learning_rate": 2.862515844028111e-05,
      "loss": 0.2478,
      "step": 1295
    },
    {
      "epoch": 11.711711711711711,
      "grad_norm": 0.7109375,
      "learning_rate": 2.789465719569906e-05,
      "loss": 0.2521,
      "step": 1300
    },
    {
      "epoch": 11.756756756756756,
      "grad_norm": 0.6875,
      "learning_rate": 2.717208424233091e-05,
      "loss": 0.2525,
      "step": 1305
    },
    {
      "epoch": 11.801801801801801,
      "grad_norm": 0.7421875,
      "learning_rate": 2.6457519030167143e-05,
      "loss": 0.2566,
      "step": 1310
    },
    {
      "epoch": 11.846846846846846,
      "grad_norm": 0.7109375,
      "learning_rate": 2.57510401287128e-05,
      "loss": 0.2495,
      "step": 1315
    },
    {
      "epoch": 11.891891891891891,
      "grad_norm": 0.6796875,
      "learning_rate": 2.5052725218348394e-05,
      "loss": 0.2526,
      "step": 1320
    },
    {
      "epoch": 11.936936936936936,
      "grad_norm": 0.71484375,
      "learning_rate": 2.4362651081788667e-05,
      "loss": 0.249,
      "step": 1325
    },
    {
      "epoch": 11.981981981981981,
      "grad_norm": 0.77734375,
      "learning_rate": 2.3680893595640043e-05,
      "loss": 0.2549,
      "step": 1330
    },
    {
      "epoch": 12.0,
      "eval_loss": 4.838156700134277,
      "eval_runtime": 1.0155,
      "eval_samples_per_second": 4.923,
      "eval_steps_per_second": 1.969,
      "step": 1332
    },
    {
      "epoch": 12.027027027027026,
      "grad_norm": 0.625,
      "learning_rate": 2.3007527722057488e-05,
      "loss": 0.2452,
      "step": 1335
    },
    {
      "epoch": 12.072072072072071,
      "grad_norm": 0.6484375,
      "learning_rate": 2.234262750050241e-05,
      "loss": 0.2436,
      "step": 1340
    },
    {
      "epoch": 12.117117117117116,
      "grad_norm": 0.6171875,
      "learning_rate": 2.168626603960138e-05,
      "loss": 0.2396,
      "step": 1345
    },
    {
      "epoch": 12.162162162162161,
      "grad_norm": 0.671875,
      "learning_rate": 2.1038515509107736e-05,
      "loss": 0.2344,
      "step": 1350
    },
    {
      "epoch": 12.207207207207206,
      "grad_norm": 0.71875,
      "learning_rate": 2.0399447131966132e-05,
      "loss": 0.2359,
      "step": 1355
    },
    {
      "epoch": 12.252252252252251,
      "grad_norm": 0.65234375,
      "learning_rate": 1.976913117648128e-05,
      "loss": 0.2374,
      "step": 1360
    },
    {
      "epoch": 12.297297297297296,
      "grad_norm": 0.6015625,
      "learning_rate": 1.9147636948591653e-05,
      "loss": 0.2406,
      "step": 1365
    },
    {
      "epoch": 12.342342342342342,
      "grad_norm": 0.65234375,
      "learning_rate": 1.8535032784249028e-05,
      "loss": 0.2392,
      "step": 1370
    },
    {
      "epoch": 12.387387387387387,
      "grad_norm": 0.703125,
      "learning_rate": 1.7931386041904506e-05,
      "loss": 0.2347,
      "step": 1375
    },
    {
      "epoch": 12.432432432432432,
      "grad_norm": 0.66796875,
      "learning_rate": 1.733676309510245e-05,
      "loss": 0.241,
      "step": 1380
    },
    {
      "epoch": 12.477477477477478,
      "grad_norm": 0.71875,
      "learning_rate": 1.6751229325182195e-05,
      "loss": 0.2392,
      "step": 1385
    },
    {
      "epoch": 12.522522522522522,
      "grad_norm": 0.65234375,
      "learning_rate": 1.6174849114089064e-05,
      "loss": 0.2389,
      "step": 1390
    },
    {
      "epoch": 12.567567567567568,
      "grad_norm": 0.703125,
      "learning_rate": 1.5607685837295516e-05,
      "loss": 0.2397,
      "step": 1395
    },
    {
      "epoch": 12.612612612612612,
      "grad_norm": 0.6640625,
      "learning_rate": 1.5049801856832557e-05,
      "loss": 0.2343,
      "step": 1400
    },
    {
      "epoch": 12.657657657657658,
      "grad_norm": 0.67578125,
      "learning_rate": 1.4501258514432836e-05,
      "loss": 0.2364,
      "step": 1405
    },
    {
      "epoch": 12.702702702702704,
      "grad_norm": 0.69140625,
      "learning_rate": 1.3962116124785863e-05,
      "loss": 0.2416,
      "step": 1410
    },
    {
      "epoch": 12.747747747747749,
      "grad_norm": 0.6796875,
      "learning_rate": 1.3432433968906044e-05,
      "loss": 0.238,
      "step": 1415
    },
    {
      "epoch": 12.792792792792794,
      "grad_norm": 0.625,
      "learning_rate": 1.2912270287614736e-05,
      "loss": 0.2355,
      "step": 1420
    },
    {
      "epoch": 12.837837837837839,
      "grad_norm": 0.65625,
      "learning_rate": 1.240168227513614e-05,
      "loss": 0.2388,
      "step": 1425
    },
    {
      "epoch": 12.882882882882884,
      "grad_norm": 0.6640625,
      "learning_rate": 1.1900726072808666e-05,
      "loss": 0.2351,
      "step": 1430
    },
    {
      "epoch": 12.927927927927929,
      "grad_norm": 0.65234375,
      "learning_rate": 1.1409456762911985e-05,
      "loss": 0.2393,
      "step": 1435
    },
    {
      "epoch": 12.972972972972974,
      "grad_norm": 0.65234375,
      "learning_rate": 1.0927928362610462e-05,
      "loss": 0.2408,
      "step": 1440
    },
    {
      "epoch": 13.0,
      "eval_loss": 5.061110496520996,
      "eval_runtime": 1.0149,
      "eval_samples_per_second": 4.927,
      "eval_steps_per_second": 1.971,
      "step": 1443
    },
    {
      "epoch": 13.018018018018019,
      "grad_norm": 0.68359375,
      "learning_rate": 1.0456193818013726e-05,
      "loss": 0.2395,
      "step": 1445
    },
    {
      "epoch": 13.063063063063064,
      "grad_norm": 0.66015625,
      "learning_rate": 9.99430499835503e-06,
      "loss": 0.2342,
      "step": 1450
    },
    {
      "epoch": 13.108108108108109,
      "grad_norm": 0.75390625,
      "learning_rate": 9.542312690288036e-06,
      "loss": 0.2342,
      "step": 1455
    },
    {
      "epoch": 13.153153153153154,
      "grad_norm": 0.6015625,
      "learning_rate": 9.100266592302542e-06,
      "loss": 0.2332,
      "step": 1460
    },
    {
      "epoch": 13.198198198198199,
      "grad_norm": 0.671875,
      "learning_rate": 8.668215309259997e-06,
      "loss": 0.2327,
      "step": 1465
    },
    {
      "epoch": 13.243243243243244,
      "grad_norm": 0.62890625,
      "learning_rate": 8.246206347049079e-06,
      "loss": 0.2353,
      "step": 1470
    },
    {
      "epoch": 13.288288288288289,
      "grad_norm": 0.66796875,
      "learning_rate": 7.834286107362343e-06,
      "loss": 0.2331,
      "step": 1475
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 0.67578125,
      "learning_rate": 7.432499882594091e-06,
      "loss": 0.2323,
      "step": 1480
    },
    {
      "epoch": 13.378378378378379,
      "grad_norm": 0.68359375,
      "learning_rate": 7.040891850860287e-06,
      "loss": 0.2354,
      "step": 1485
    },
    {
      "epoch": 13.423423423423424,
      "grad_norm": 0.6328125,
      "learning_rate": 6.659505071140959e-06,
      "loss": 0.2345,
      "step": 1490
    },
    {
      "epoch": 13.468468468468469,
      "grad_norm": 0.62109375,
      "learning_rate": 6.2883814785457575e-06,
      "loss": 0.2376,
      "step": 1495
    },
    {
      "epoch": 13.513513513513514,
      "grad_norm": 0.64453125,
      "learning_rate": 5.927561879702903e-06,
      "loss": 0.2361,
      "step": 1500
    },
    {
      "epoch": 13.558558558558559,
      "grad_norm": 0.72265625,
      "learning_rate": 5.577085948272431e-06,
      "loss": 0.2331,
      "step": 1505
    },
    {
      "epoch": 13.603603603603604,
      "grad_norm": 0.609375,
      "learning_rate": 5.236992220583814e-06,
      "loss": 0.2312,
      "step": 1510
    },
    {
      "epoch": 13.64864864864865,
      "grad_norm": 0.6953125,
      "learning_rate": 4.90731809139876e-06,
      "loss": 0.2347,
      "step": 1515
    },
    {
      "epoch": 13.693693693693694,
      "grad_norm": 0.62890625,
      "learning_rate": 4.588099809799551e-06,
      "loss": 0.2359,
      "step": 1520
    },
    {
      "epoch": 13.73873873873874,
      "grad_norm": 0.62890625,
      "learning_rate": 4.279372475203181e-06,
      "loss": 0.2321,
      "step": 1525
    },
    {
      "epoch": 13.783783783783784,
      "grad_norm": 0.65625,
      "learning_rate": 3.981170033502158e-06,
      "loss": 0.2326,
      "step": 1530
    },
    {
      "epoch": 13.82882882882883,
      "grad_norm": 0.6328125,
      "learning_rate": 3.6935252733319213e-06,
      "loss": 0.2341,
      "step": 1535
    },
    {
      "epoch": 13.873873873873874,
      "grad_norm": 0.6171875,
      "learning_rate": 3.4164698224656066e-06,
      "loss": 0.2308,
      "step": 1540
    },
    {
      "epoch": 13.91891891891892,
      "grad_norm": 0.66796875,
      "learning_rate": 3.1500341443364556e-06,
      "loss": 0.2338,
      "step": 1545
    },
    {
      "epoch": 13.963963963963964,
      "grad_norm": 0.66796875,
      "learning_rate": 2.8942475346882435e-06,
      "loss": 0.2361,
      "step": 1550
    },
    {
      "epoch": 14.0,
      "eval_loss": 5.106085777282715,
      "eval_runtime": 1.0149,
      "eval_samples_per_second": 4.927,
      "eval_steps_per_second": 1.971,
      "step": 1554
    },
    {
      "epoch": 14.00900900900901,
      "grad_norm": 0.61328125,
      "learning_rate": 2.649138118354011e-06,
      "loss": 0.2352,
      "step": 1555
    },
    {
      "epoch": 14.054054054054054,
      "grad_norm": 0.6796875,
      "learning_rate": 2.4147328461637144e-06,
      "loss": 0.2335,
      "step": 1560
    },
    {
      "epoch": 14.0990990990991,
      "grad_norm": 0.6484375,
      "learning_rate": 2.1910574919808304e-06,
      "loss": 0.2308,
      "step": 1565
    },
    {
      "epoch": 14.144144144144144,
      "grad_norm": 0.62890625,
      "learning_rate": 1.978136649868345e-06,
      "loss": 0.2345,
      "step": 1570
    },
    {
      "epoch": 14.18918918918919,
      "grad_norm": 0.66015625,
      "learning_rate": 1.775993731384662e-06,
      "loss": 0.235,
      "step": 1575
    },
    {
      "epoch": 14.234234234234235,
      "grad_norm": 0.6796875,
      "learning_rate": 1.584650963009271e-06,
      "loss": 0.2337,
      "step": 1580
    },
    {
      "epoch": 14.27927927927928,
      "grad_norm": 0.68359375,
      "learning_rate": 1.4041293836989377e-06,
      "loss": 0.2319,
      "step": 1585
    },
    {
      "epoch": 14.324324324324325,
      "grad_norm": 0.64453125,
      "learning_rate": 1.234448842574365e-06,
      "loss": 0.2332,
      "step": 1590
    },
    {
      "epoch": 14.36936936936937,
      "grad_norm": 0.6328125,
      "learning_rate": 1.075627996737627e-06,
      "loss": 0.23,
      "step": 1595
    },
    {
      "epoch": 14.414414414414415,
      "grad_norm": 0.6875,
      "learning_rate": 9.276843092208553e-07,
      "loss": 0.2353,
      "step": 1600
    },
    {
      "epoch": 14.45945945945946,
      "grad_norm": 0.640625,
      "learning_rate": 7.90634047066019e-07,
      "loss": 0.2339,
      "step": 1605
    },
    {
      "epoch": 14.504504504504505,
      "grad_norm": 0.6796875,
      "learning_rate": 6.644922795363218e-07,
      "loss": 0.2381,
      "step": 1610
    },
    {
      "epoch": 14.54954954954955,
      "grad_norm": 0.66015625,
      "learning_rate": 5.492728764592725e-07,
      "loss": 0.2335,
      "step": 1615
    },
    {
      "epoch": 14.594594594594595,
      "grad_norm": 0.65234375,
      "learning_rate": 4.4498850670164906e-07,
      "loss": 0.2343,
      "step": 1620
    },
    {
      "epoch": 14.63963963963964,
      "grad_norm": 0.6015625,
      "learning_rate": 3.5165063677645714e-07,
      "loss": 0.2369,
      "step": 1625
    },
    {
      "epoch": 14.684684684684685,
      "grad_norm": 0.65234375,
      "learning_rate": 2.6926952958221674e-07,
      "loss": 0.2367,
      "step": 1630
    },
    {
      "epoch": 14.72972972972973,
      "grad_norm": 0.6328125,
      "learning_rate": 1.9785424327440906e-07,
      "loss": 0.2302,
      "step": 1635
    },
    {
      "epoch": 14.774774774774775,
      "grad_norm": 0.6640625,
      "learning_rate": 1.374126302696066e-07,
      "loss": 0.237,
      "step": 1640
    },
    {
      "epoch": 14.81981981981982,
      "grad_norm": 0.65625,
      "learning_rate": 8.795133638197506e-08,
      "loss": 0.2298,
      "step": 1645
    },
    {
      "epoch": 14.864864864864865,
      "grad_norm": 0.7109375,
      "learning_rate": 4.94758000925799e-08,
      "loss": 0.2372,
      "step": 1650
    },
    {
      "epoch": 14.90990990990991,
      "grad_norm": 0.62890625,
      "learning_rate": 2.1990251951398144e-08,
      "loss": 0.2312,
      "step": 1655
    },
    {
      "epoch": 14.954954954954955,
      "grad_norm": 0.62890625,
      "learning_rate": 5.497714112157048e-09,
      "loss": 0.2311,
      "step": 1660
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0,
      "loss": 0.2319,
      "step": 1665
    },
    {
      "epoch": 15.0,
      "eval_loss": 5.111132621765137,
      "eval_runtime": 1.0137,
      "eval_samples_per_second": 4.932,
      "eval_steps_per_second": 1.973,
      "step": 1665
    },
    {
      "epoch": 15.0,
      "step": 1665,
      "total_flos": 2.545573832974926e+18,
      "train_loss": 1.4823458194016694,
      "train_runtime": 13248.7618,
      "train_samples_per_second": 2.007,
      "train_steps_per_second": 0.126
    }
  ],
  "logging_steps": 5,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.545573832974926e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}