{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.07272727272727272,
  "eval_steps": 1000,
  "global_step": 8000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0009090909090909091,
      "grad_norm": 0.5637483596801758,
      "learning_rate": 0.00066,
      "loss": 7.38281005859375,
      "step": 100
    },
    {
      "epoch": 0.0018181818181818182,
      "grad_norm": 0.6073591709136963,
      "learning_rate": 0.000999999509056267,
      "loss": 5.783165283203125,
      "step": 200
    },
    {
      "epoch": 0.0027272727272727275,
      "grad_norm": 0.480266273021698,
      "learning_rate": 0.0009999954604635122,
      "loss": 5.232196044921875,
      "step": 300
    },
    {
      "epoch": 0.0036363636363636364,
      "grad_norm": 0.45949065685272217,
      "learning_rate": 0.0009999873224161853,
      "loss": 4.946490478515625,
      "step": 400
    },
    {
      "epoch": 0.004545454545454545,
      "grad_norm": 0.41542109847068787,
      "learning_rate": 0.000999975094980847,
      "loss": 4.719376831054688,
      "step": 500
    },
    {
      "epoch": 0.005454545454545455,
      "grad_norm": 0.40697741508483887,
      "learning_rate": 0.0009999587782575051,
      "loss": 4.591694030761719,
      "step": 600
    },
    {
      "epoch": 0.006363636363636364,
      "grad_norm": 0.3962944447994232,
      "learning_rate": 0.0009999383723796143,
      "loss": 4.466532592773437,
      "step": 700
    },
    {
      "epoch": 0.007272727272727273,
      "grad_norm": 0.6857792139053345,
      "learning_rate": 0.0009999138775140734,
      "loss": 4.349520874023438,
      "step": 800
    },
    {
      "epoch": 0.008181818181818182,
      "grad_norm": 0.4181463420391083,
      "learning_rate": 0.000999885293861226,
      "loss": 4.234326782226563,
      "step": 900
    },
    {
      "epoch": 0.00909090909090909,
      "grad_norm": 0.3620698153972626,
      "learning_rate": 0.0009998526216548572,
      "loss": 4.120015869140625,
      "step": 1000
    },
    {
      "epoch": 0.00909090909090909,
      "eval_loss": 4.375123977661133,
      "eval_runtime": 6.9666,
      "eval_samples_per_second": 82.681,
      "eval_steps_per_second": 20.67,
      "step": 1000
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.3509594798088074,
      "learning_rate": 0.0009998158611621922,
      "loss": 3.993647155761719,
      "step": 1100
    },
    {
      "epoch": 0.01090909090909091,
      "grad_norm": 0.3581515848636627,
      "learning_rate": 0.0009997750126838946,
      "loss": 3.9498220825195314,
      "step": 1200
    },
    {
      "epoch": 0.011818181818181818,
      "grad_norm": 0.36733096837997437,
      "learning_rate": 0.0009997300765540635,
      "loss": 3.831287841796875,
      "step": 1300
    },
    {
      "epoch": 0.012727272727272728,
      "grad_norm": 0.29147249460220337,
      "learning_rate": 0.0009996810531402306,
      "loss": 3.7590908813476562,
      "step": 1400
    },
    {
      "epoch": 0.013636363636363636,
      "grad_norm": 0.29391637444496155,
      "learning_rate": 0.0009996279428433578,
      "loss": 3.722886047363281,
      "step": 1500
    },
    {
      "epoch": 0.014545454545454545,
      "grad_norm": 0.3317511975765228,
      "learning_rate": 0.000999570746097833,
      "loss": 3.7012667846679688,
      "step": 1600
    },
    {
      "epoch": 0.015454545454545455,
      "grad_norm": 0.32525742053985596,
      "learning_rate": 0.0009995094633714677,
      "loss": 3.6526870727539062,
      "step": 1700
    },
    {
      "epoch": 0.016363636363636365,
      "grad_norm": 0.3586934208869934,
      "learning_rate": 0.0009994440951654924,
      "loss": 3.6262481689453123,
      "step": 1800
    },
    {
      "epoch": 0.017272727272727273,
      "grad_norm": 0.30995482206344604,
      "learning_rate": 0.0009993746420145523,
      "loss": 3.594487609863281,
      "step": 1900
    },
    {
      "epoch": 0.01818181818181818,
      "grad_norm": 0.3067973852157593,
      "learning_rate": 0.0009993011044867038,
      "loss": 3.5707028198242186,
      "step": 2000
    },
    {
      "epoch": 0.01818181818181818,
      "eval_loss": 3.896928548812866,
      "eval_runtime": 6.7872,
      "eval_samples_per_second": 84.865,
      "eval_steps_per_second": 21.216,
      "step": 2000
    },
    {
      "epoch": 0.019090909090909092,
      "grad_norm": 0.2716964781284332,
      "learning_rate": 0.000999223483183409,
      "loss": 3.5685223388671874,
      "step": 2100
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.2595560550689697,
      "learning_rate": 0.000999141778739531,
      "loss": 3.533115234375,
      "step": 2200
    },
    {
      "epoch": 0.02090909090909091,
      "grad_norm": 0.2871851325035095,
      "learning_rate": 0.0009990559918233294,
      "loss": 3.51209716796875,
      "step": 2300
    },
    {
      "epoch": 0.02181818181818182,
      "grad_norm": 0.2916632890701294,
      "learning_rate": 0.0009989661231364542,
      "loss": 3.4977175903320314,
      "step": 2400
    },
    {
      "epoch": 0.022727272727272728,
      "grad_norm": 0.2674938440322876,
      "learning_rate": 0.0009988721734139395,
      "loss": 3.4819268798828125,
      "step": 2500
    },
    {
      "epoch": 0.023636363636363636,
      "grad_norm": 0.28354012966156006,
      "learning_rate": 0.0009987741434241983,
      "loss": 3.4985772705078126,
      "step": 2600
    },
    {
      "epoch": 0.024545454545454544,
      "grad_norm": 0.250453919172287,
      "learning_rate": 0.000998672033969017,
      "loss": 3.4824765014648436,
      "step": 2700
    },
    {
      "epoch": 0.025454545454545455,
      "grad_norm": 0.25479111075401306,
      "learning_rate": 0.0009985658458835465,
      "loss": 3.43513916015625,
      "step": 2800
    },
    {
      "epoch": 0.026363636363636363,
      "grad_norm": 0.5312749743461609,
      "learning_rate": 0.0009984555800362977,
      "loss": 3.4338519287109377,
      "step": 2900
    },
    {
      "epoch": 0.02727272727272727,
      "grad_norm": 0.24919964373111725,
      "learning_rate": 0.0009983412373291332,
      "loss": 3.394432373046875,
      "step": 3000
    },
    {
      "epoch": 0.02727272727272727,
      "eval_loss": 3.7320191860198975,
      "eval_runtime": 6.8112,
      "eval_samples_per_second": 84.566,
      "eval_steps_per_second": 21.142,
      "step": 3000
    },
    {
      "epoch": 0.028181818181818183,
      "grad_norm": 0.2345028668642044,
      "learning_rate": 0.0009982228186972597,
      "loss": 3.3772811889648438,
      "step": 3100
    },
    {
      "epoch": 0.02909090909090909,
      "grad_norm": 0.2587430775165558,
      "learning_rate": 0.0009981003251092215,
      "loss": 3.368675537109375,
      "step": 3200
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.25332900881767273,
      "learning_rate": 0.0009979737575668917,
      "loss": 3.3781060791015625,
      "step": 3300
    },
    {
      "epoch": 0.03090909090909091,
      "grad_norm": 0.2273174375295639,
      "learning_rate": 0.000997843117105464,
      "loss": 3.328343505859375,
      "step": 3400
    },
    {
      "epoch": 0.031818181818181815,
      "grad_norm": 0.2621140778064728,
      "learning_rate": 0.0009977084047934448,
      "loss": 3.3511767578125,
      "step": 3500
    },
    {
      "epoch": 0.03272727272727273,
      "grad_norm": 0.22301150858402252,
      "learning_rate": 0.0009975696217326435,
      "loss": 3.3483154296875,
      "step": 3600
    },
    {
      "epoch": 0.03363636363636364,
      "grad_norm": 0.22785483300685883,
      "learning_rate": 0.0009974267690581644,
      "loss": 3.351199951171875,
      "step": 3700
    },
    {
      "epoch": 0.034545454545454546,
      "grad_norm": 0.24507705867290497,
      "learning_rate": 0.0009972798479383977,
      "loss": 3.311331787109375,
      "step": 3800
    },
    {
      "epoch": 0.035454545454545454,
      "grad_norm": 0.22445634007453918,
      "learning_rate": 0.0009971288595750083,
      "loss": 3.2849237060546876,
      "step": 3900
    },
    {
      "epoch": 0.03636363636363636,
      "grad_norm": 0.2134762406349182,
      "learning_rate": 0.0009969738052029277,
      "loss": 3.2534124755859377,
      "step": 4000
    },
    {
      "epoch": 0.03636363636363636,
      "eval_loss": 3.6267645359039307,
      "eval_runtime": 6.8001,
      "eval_samples_per_second": 84.704,
      "eval_steps_per_second": 21.176,
      "step": 4000
    },
    {
      "epoch": 0.03727272727272727,
      "grad_norm": 0.2137739211320877,
      "learning_rate": 0.000996814686090343,
      "loss": 3.288787841796875,
      "step": 4100
    },
    {
      "epoch": 0.038181818181818185,
      "grad_norm": 0.20396745204925537,
      "learning_rate": 0.0009966515035386867,
      "loss": 3.277693786621094,
      "step": 4200
    },
    {
      "epoch": 0.03909090909090909,
      "grad_norm": 0.20048071444034576,
      "learning_rate": 0.0009964842588826267,
      "loss": 3.2614068603515625,
      "step": 4300
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.2545357346534729,
      "learning_rate": 0.0009963129534900535,
      "loss": 3.241942138671875,
      "step": 4400
    },
    {
      "epoch": 0.04090909090909091,
      "grad_norm": 0.21150407195091248,
      "learning_rate": 0.0009961375887620715,
      "loss": 3.254181213378906,
      "step": 4500
    },
    {
      "epoch": 0.04181818181818182,
      "grad_norm": 0.25913697481155396,
      "learning_rate": 0.0009959581661329855,
      "loss": 3.256631164550781,
      "step": 4600
    },
    {
      "epoch": 0.042727272727272725,
      "grad_norm": 0.21662315726280212,
      "learning_rate": 0.000995774687070291,
      "loss": 3.246457214355469,
      "step": 4700
    },
    {
      "epoch": 0.04363636363636364,
      "grad_norm": 0.21578437089920044,
      "learning_rate": 0.0009955871530746593,
      "loss": 3.2591009521484375,
      "step": 4800
    },
    {
      "epoch": 0.04454545454545455,
      "grad_norm": 0.2277086228132248,
      "learning_rate": 0.0009953955656799285,
      "loss": 3.2139767456054686,
      "step": 4900
    },
    {
      "epoch": 0.045454545454545456,
      "grad_norm": 0.21145710349082947,
      "learning_rate": 0.0009951999264530885,
      "loss": 3.217757263183594,
      "step": 5000
    },
    {
      "epoch": 0.045454545454545456,
      "eval_loss": 3.559657573699951,
      "eval_runtime": 6.8094,
      "eval_samples_per_second": 84.589,
      "eval_steps_per_second": 21.147,
      "step": 5000
    },
    {
      "epoch": 0.046363636363636364,
      "grad_norm": 0.2059008628129959,
      "learning_rate": 0.0009950002369942693,
      "loss": 3.2310595703125,
      "step": 5100
    },
    {
      "epoch": 0.04727272727272727,
      "grad_norm": 0.31273841857910156,
      "learning_rate": 0.0009947964989367275,
      "loss": 3.236092529296875,
      "step": 5200
    },
    {
      "epoch": 0.04818181818181818,
      "grad_norm": 0.20427817106246948,
      "learning_rate": 0.0009945887139468335,
      "loss": 3.20990966796875,
      "step": 5300
    },
    {
      "epoch": 0.04909090909090909,
      "grad_norm": 0.21813584864139557,
      "learning_rate": 0.0009943768837240572,
      "loss": 3.23782470703125,
      "step": 5400
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.21357640624046326,
      "learning_rate": 0.0009941610100009546,
      "loss": 3.2014178466796874,
      "step": 5500
    },
    {
      "epoch": 0.05090909090909091,
      "grad_norm": 0.19788503646850586,
      "learning_rate": 0.0009939410945431532,
      "loss": 3.201743469238281,
      "step": 5600
    },
    {
      "epoch": 0.05181818181818182,
      "grad_norm": 0.2109798789024353,
      "learning_rate": 0.0009937171391493375,
      "loss": 3.1780130004882814,
      "step": 5700
    },
    {
      "epoch": 0.05272727272727273,
      "grad_norm": 0.1952320784330368,
      "learning_rate": 0.0009934891456512356,
      "loss": 3.1811758422851564,
      "step": 5800
    },
    {
      "epoch": 0.053636363636363635,
      "grad_norm": 0.19197046756744385,
      "learning_rate": 0.0009932571159136023,
      "loss": 3.1767172241210937,
      "step": 5900
    },
    {
      "epoch": 0.05454545454545454,
      "grad_norm": 0.20798423886299133,
      "learning_rate": 0.0009930210518342049,
      "loss": 3.1700228881835937,
      "step": 6000
    },
    {
      "epoch": 0.05454545454545454,
      "eval_loss": 3.494601249694824,
      "eval_runtime": 6.8416,
      "eval_samples_per_second": 84.191,
      "eval_steps_per_second": 21.048,
      "step": 6000
    },
    {
      "epoch": 0.05545454545454546,
      "grad_norm": 0.18669773638248444,
      "learning_rate": 0.0009927809553438077,
      "loss": 3.1545779418945314,
      "step": 6100
    },
    {
      "epoch": 0.056363636363636366,
      "grad_norm": 0.2077275514602661,
      "learning_rate": 0.0009925368284061555,
      "loss": 3.160372619628906,
      "step": 6200
    },
    {
      "epoch": 0.057272727272727274,
      "grad_norm": 0.18006086349487305,
      "learning_rate": 0.000992288673017959,
      "loss": 3.1383782958984376,
      "step": 6300
    },
    {
      "epoch": 0.05818181818181818,
      "grad_norm": 0.19835108518600464,
      "learning_rate": 0.000992036491208877,
      "loss": 3.15177001953125,
      "step": 6400
    },
    {
      "epoch": 0.05909090909090909,
      "grad_norm": 0.19680027663707733,
      "learning_rate": 0.0009917802850414998,
      "loss": 3.16474365234375,
      "step": 6500
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.28204840421676636,
      "learning_rate": 0.0009915200566113338,
      "loss": 3.1483929443359373,
      "step": 6600
    },
    {
      "epoch": 0.060909090909090906,
      "grad_norm": 0.18320120871067047,
      "learning_rate": 0.000991255808046783,
      "loss": 3.150434875488281,
      "step": 6700
    },
    {
      "epoch": 0.06181818181818182,
      "grad_norm": 0.19188480079174042,
      "learning_rate": 0.000990987541509132,
      "loss": 3.118103332519531,
      "step": 6800
    },
    {
      "epoch": 0.06272727272727273,
      "grad_norm": 0.1749107539653778,
      "learning_rate": 0.0009907152591925282,
      "loss": 3.1161154174804686,
      "step": 6900
    },
    {
      "epoch": 0.06363636363636363,
      "grad_norm": 0.17627771198749542,
      "learning_rate": 0.0009904389633239644,
      "loss": 3.140033874511719,
      "step": 7000
    },
    {
      "epoch": 0.06363636363636363,
      "eval_loss": 3.460543632507324,
      "eval_runtime": 6.8262,
      "eval_samples_per_second": 84.381,
      "eval_steps_per_second": 21.095,
      "step": 7000
    },
    {
      "epoch": 0.06454545454545454,
      "grad_norm": 0.19298093020915985,
      "learning_rate": 0.0009901586561632597,
      "loss": 3.1127783203125,
      "step": 7100
    },
    {
      "epoch": 0.06545454545454546,
      "grad_norm": 0.2005312442779541,
      "learning_rate": 0.0009898743400030417,
      "loss": 3.131772155761719,
      "step": 7200
    },
    {
      "epoch": 0.06636363636363636,
      "grad_norm": 0.20932316780090332,
      "learning_rate": 0.0009895860171687276,
      "loss": 3.1065127563476564,
      "step": 7300
    },
    {
      "epoch": 0.06727272727272728,
      "grad_norm": 0.1870882213115692,
      "learning_rate": 0.0009892936900185053,
      "loss": 3.1155380249023437,
      "step": 7400
    },
    {
      "epoch": 0.06818181818181818,
      "grad_norm": 0.18347717821598053,
      "learning_rate": 0.0009889973609433136,
      "loss": 3.094842529296875,
      "step": 7500
    },
    {
      "epoch": 0.06909090909090909,
      "grad_norm": 0.20144793391227722,
      "learning_rate": 0.000988697032366823,
      "loss": 3.1140432739257813,
      "step": 7600
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.17255835235118866,
      "learning_rate": 0.0009883927067454157,
      "loss": 3.0936923217773438,
      "step": 7700
    },
    {
      "epoch": 0.07090909090909091,
      "grad_norm": 0.19149735569953918,
      "learning_rate": 0.0009880843865681666,
      "loss": 3.0889437866210936,
      "step": 7800
    },
    {
      "epoch": 0.07181818181818182,
      "grad_norm": 0.18358978629112244,
      "learning_rate": 0.0009877720743568204,
      "loss": 3.0625601196289063,
      "step": 7900
    },
    {
      "epoch": 0.07272727272727272,
      "grad_norm": 0.16700264811515808,
      "learning_rate": 0.0009874557726657744,
      "loss": 3.0754776000976562,
      "step": 8000
    },
    {
      "epoch": 0.07272727272727272,
      "eval_loss": 3.418283700942993,
      "eval_runtime": 6.8139,
      "eval_samples_per_second": 84.533,
      "eval_steps_per_second": 21.133,
      "step": 8000
    }
  ],
  "logging_steps": 100,
  "max_steps": 110000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 4000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.99316647968768e+17,
  "train_batch_size": 22,
  "trial_name": null,
  "trial_params": null
}