{
  "best_global_step": 20000,
  "best_metric": 6.920494556427002,
  "best_model_checkpoint": "distilbert_km_50_v1/checkpoint-20000",
  "epoch": 25.0,
  "eval_steps": 10000,
  "global_step": 59550,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.20990764063811923,
      "grad_norm": 2.6400372982025146,
      "learning_rate": 5e-06,
      "loss": 13.2025,
      "step": 500
    },
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 2.3887388706207275,
      "learning_rate": 1e-05,
      "loss": 11.425,
      "step": 1000
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 5.628344535827637,
      "learning_rate": 1.5e-05,
      "loss": 10.0752,
      "step": 1500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 9.419818878173828,
      "learning_rate": 2e-05,
      "loss": 8.9098,
      "step": 2000
    },
    {
      "epoch": 1.0495382031905962,
      "grad_norm": 7.596254348754883,
      "learning_rate": 2.5e-05,
      "loss": 8.3163,
      "step": 2500
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 11.880325317382812,
      "learning_rate": 3e-05,
      "loss": 7.9723,
      "step": 3000
    },
    {
      "epoch": 1.4693534844668346,
      "grad_norm": 10.95488166809082,
      "learning_rate": 3.5e-05,
      "loss": 7.7977,
      "step": 3500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 11.735587120056152,
      "learning_rate": 4e-05,
      "loss": 7.6844,
      "step": 4000
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 10.144121170043945,
      "learning_rate": 4.5e-05,
      "loss": 7.5924,
      "step": 4500
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 11.868781089782715,
      "learning_rate": 5e-05,
      "loss": 7.4575,
      "step": 5000
    },
    {
      "epoch": 2.3089840470193117,
      "grad_norm": 10.09228515625,
      "learning_rate": 5.500000000000001e-05,
      "loss": 7.3757,
      "step": 5500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 11.239814758300781,
      "learning_rate": 6e-05,
      "loss": 7.3384,
      "step": 6000
    },
    {
      "epoch": 2.72879932829555,
      "grad_norm": 5.699562072753906,
      "learning_rate": 6.500000000000001e-05,
      "loss": 7.3122,
      "step": 6500
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 6.938751220703125,
      "learning_rate": 7e-05,
      "loss": 7.2587,
      "step": 7000
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 6.043473243713379,
      "learning_rate": 7.500000000000001e-05,
      "loss": 7.1687,
      "step": 7500
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 7.517977714538574,
      "learning_rate": 8e-05,
      "loss": 7.1558,
      "step": 8000
    },
    {
      "epoch": 3.568429890848027,
      "grad_norm": 3.7469329833984375,
      "learning_rate": 8.5e-05,
      "loss": 7.1506,
      "step": 8500
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 6.092977046966553,
      "learning_rate": 9e-05,
      "loss": 7.1332,
      "step": 9000
    },
    {
      "epoch": 3.988245172124265,
      "grad_norm": 5.3122758865356445,
      "learning_rate": 9.5e-05,
      "loss": 7.1163,
      "step": 9500
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 4.237173557281494,
      "learning_rate": 0.0001,
      "loss": 7.0054,
      "step": 10000
    },
    {
      "epoch": 4.198152812762385,
      "eval_accuracy": 0.15031024792008682,
      "eval_loss": 7.000421047210693,
      "eval_runtime": 0.8685,
      "eval_samples_per_second": 551.517,
      "eval_steps_per_second": 5.757,
      "step": 10000
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 4.337963104248047,
      "learning_rate": 9.899091826437942e-05,
      "loss": 7.014,
      "step": 10500
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 2.9401566982269287,
      "learning_rate": 9.798183652875883e-05,
      "loss": 7.0018,
      "step": 11000
    },
    {
      "epoch": 4.827875734676742,
      "grad_norm": 4.914291858673096,
      "learning_rate": 9.697275479313825e-05,
      "loss": 6.9856,
      "step": 11500
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 3.450915813446045,
      "learning_rate": 9.596367305751767e-05,
      "loss": 6.9456,
      "step": 12000
    },
    {
      "epoch": 5.247691015952981,
      "grad_norm": 3.833019733428955,
      "learning_rate": 9.495459132189707e-05,
      "loss": 6.8469,
      "step": 12500
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 4.264673233032227,
      "learning_rate": 9.394550958627649e-05,
      "loss": 6.8571,
      "step": 13000
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 5.008898735046387,
      "learning_rate": 9.293642785065591e-05,
      "loss": 6.8572,
      "step": 13500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 3.5240471363067627,
      "learning_rate": 9.192734611503532e-05,
      "loss": 6.8428,
      "step": 14000
    },
    {
      "epoch": 6.087321578505458,
      "grad_norm": 3.9131734371185303,
      "learning_rate": 9.091826437941474e-05,
      "loss": 6.7869,
      "step": 14500
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 4.489323616027832,
      "learning_rate": 8.990918264379415e-05,
      "loss": 6.7214,
      "step": 15000
    },
    {
      "epoch": 6.507136859781696,
      "grad_norm": 3.5834033489227295,
      "learning_rate": 8.890010090817356e-05,
      "loss": 6.7227,
      "step": 15500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 4.640388488769531,
      "learning_rate": 8.789101917255298e-05,
      "loss": 6.7273,
      "step": 16000
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 4.809169292449951,
      "learning_rate": 8.68819374369324e-05,
      "loss": 6.7251,
      "step": 16500
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 3.2672581672668457,
      "learning_rate": 8.58728557013118e-05,
      "loss": 6.6367,
      "step": 17000
    },
    {
      "epoch": 7.346767422334173,
      "grad_norm": 3.8383593559265137,
      "learning_rate": 8.486377396569122e-05,
      "loss": 6.6069,
      "step": 17500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 3.420926094055176,
      "learning_rate": 8.385469223007064e-05,
      "loss": 6.6184,
      "step": 18000
    },
    {
      "epoch": 7.766582703610411,
      "grad_norm": 4.380417346954346,
      "learning_rate": 8.284561049445006e-05,
      "loss": 6.6307,
      "step": 18500
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 3.3602373600006104,
      "learning_rate": 8.183652875882947e-05,
      "loss": 6.6352,
      "step": 19000
    },
    {
      "epoch": 8.18639798488665,
      "grad_norm": 3.5435893535614014,
      "learning_rate": 8.082744702320889e-05,
      "loss": 6.5175,
      "step": 19500
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 4.299872398376465,
      "learning_rate": 7.98183652875883e-05,
      "loss": 6.5192,
      "step": 20000
    },
    {
      "epoch": 8.39630562552477,
      "eval_accuracy": 0.15236117736187885,
      "eval_loss": 6.920494556427002,
      "eval_runtime": 0.8358,
      "eval_samples_per_second": 573.138,
      "eval_steps_per_second": 5.983,
      "step": 20000
    },
    {
      "epoch": 8.606213266162888,
      "grad_norm": 6.067690372467041,
      "learning_rate": 7.880928355196771e-05,
      "loss": 6.5272,
      "step": 20500
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 5.165555000305176,
      "learning_rate": 7.780020181634713e-05,
      "loss": 6.5409,
      "step": 21000
    },
    {
      "epoch": 9.026028547439127,
      "grad_norm": 3.1958303451538086,
      "learning_rate": 7.679112008072655e-05,
      "loss": 6.5146,
      "step": 21500
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 4.4170637130737305,
      "learning_rate": 7.578203834510595e-05,
      "loss": 6.4165,
      "step": 22000
    },
    {
      "epoch": 9.445843828715365,
      "grad_norm": 4.27263879776001,
      "learning_rate": 7.477295660948537e-05,
      "loss": 6.4342,
      "step": 22500
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 3.1532535552978516,
      "learning_rate": 7.376387487386479e-05,
      "loss": 6.4436,
      "step": 23000
    },
    {
      "epoch": 9.865659109991604,
      "grad_norm": 4.744621753692627,
      "learning_rate": 7.27547931382442e-05,
      "loss": 6.443,
      "step": 23500
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 6.787881374359131,
      "learning_rate": 7.174571140262362e-05,
      "loss": 6.4044,
      "step": 24000
    },
    {
      "epoch": 10.285474391267842,
      "grad_norm": 4.035264492034912,
      "learning_rate": 7.073662966700304e-05,
      "loss": 6.3572,
      "step": 24500
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 3.052518844604492,
      "learning_rate": 6.972754793138244e-05,
      "loss": 6.3639,
      "step": 25000
    },
    {
      "epoch": 10.705289672544081,
      "grad_norm": 4.323153495788574,
      "learning_rate": 6.871846619576186e-05,
      "loss": 6.3701,
      "step": 25500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 4.366807460784912,
      "learning_rate": 6.770938446014128e-05,
      "loss": 6.3771,
      "step": 26000
    },
    {
      "epoch": 11.125104953820319,
      "grad_norm": 2.823150157928467,
      "learning_rate": 6.670030272452068e-05,
      "loss": 6.3179,
      "step": 26500
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 3.2791450023651123,
      "learning_rate": 6.56912209889001e-05,
      "loss": 6.2935,
      "step": 27000
    },
    {
      "epoch": 11.544920235096557,
      "grad_norm": 3.697699785232544,
      "learning_rate": 6.468213925327952e-05,
      "loss": 6.3014,
      "step": 27500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 4.799094200134277,
      "learning_rate": 6.367305751765893e-05,
      "loss": 6.3067,
      "step": 28000
    },
    {
      "epoch": 11.964735516372796,
      "grad_norm": 4.567785263061523,
      "learning_rate": 6.266397578203835e-05,
      "loss": 6.307,
      "step": 28500
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 4.78318977355957,
      "learning_rate": 6.165489404641777e-05,
      "loss": 6.2463,
      "step": 29000
    },
    {
      "epoch": 12.384550797649034,
      "grad_norm": 3.9838807582855225,
      "learning_rate": 6.064581231079718e-05,
      "loss": 6.2438,
      "step": 29500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 3.5604918003082275,
      "learning_rate": 5.963673057517659e-05,
      "loss": 6.2558,
      "step": 30000
    },
    {
      "epoch": 12.594458438287154,
      "eval_accuracy": 0.15306265215323053,
      "eval_loss": 6.938628673553467,
      "eval_runtime": 0.8407,
      "eval_samples_per_second": 569.738,
      "eval_steps_per_second": 5.947,
      "step": 30000
    },
    {
      "epoch": 12.804366078925273,
      "grad_norm": 3.1477909088134766,
      "learning_rate": 5.8627648839556e-05,
      "loss": 6.2488,
      "step": 30500
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 4.524667263031006,
      "learning_rate": 5.7618567103935416e-05,
      "loss": 6.2476,
      "step": 31000
    },
    {
      "epoch": 13.224181360201511,
      "grad_norm": 2.575509786605835,
      "learning_rate": 5.660948536831484e-05,
      "loss": 6.1936,
      "step": 31500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 3.360851764678955,
      "learning_rate": 5.5600403632694254e-05,
      "loss": 6.1965,
      "step": 32000
    },
    {
      "epoch": 13.64399664147775,
      "grad_norm": 4.844574928283691,
      "learning_rate": 5.4591321897073666e-05,
      "loss": 6.2077,
      "step": 32500
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 5.347206115722656,
      "learning_rate": 5.358224016145308e-05,
      "loss": 6.2056,
      "step": 33000
    },
    {
      "epoch": 14.063811922753988,
      "grad_norm": 3.8560969829559326,
      "learning_rate": 5.257315842583249e-05,
      "loss": 6.1882,
      "step": 33500
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 3.1811022758483887,
      "learning_rate": 5.15640766902119e-05,
      "loss": 6.1599,
      "step": 34000
    },
    {
      "epoch": 14.483627204030226,
      "grad_norm": 3.7264819145202637,
      "learning_rate": 5.055499495459133e-05,
      "loss": 6.1608,
      "step": 34500
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 4.093605995178223,
      "learning_rate": 4.954591321897074e-05,
      "loss": 6.1611,
      "step": 35000
    },
    {
      "epoch": 14.903442485306465,
      "grad_norm": 2.888972043991089,
      "learning_rate": 4.853683148335015e-05,
      "loss": 6.1677,
      "step": 35500
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 2.6536176204681396,
      "learning_rate": 4.7527749747729566e-05,
      "loss": 6.1436,
      "step": 36000
    },
    {
      "epoch": 15.323257766582703,
      "grad_norm": 3.367471218109131,
      "learning_rate": 4.6518668012108985e-05,
      "loss": 6.1296,
      "step": 36500
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 4.522226810455322,
      "learning_rate": 4.55095862764884e-05,
      "loss": 6.1226,
      "step": 37000
    },
    {
      "epoch": 15.743073047858942,
      "grad_norm": 5.049285411834717,
      "learning_rate": 4.450050454086781e-05,
      "loss": 6.1251,
      "step": 37500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 3.515643358230591,
      "learning_rate": 4.349142280524723e-05,
      "loss": 6.1255,
      "step": 38000
    },
    {
      "epoch": 16.162888329135182,
      "grad_norm": 3.279512882232666,
      "learning_rate": 4.248234106962664e-05,
      "loss": 6.1036,
      "step": 38500
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 6.875465393066406,
      "learning_rate": 4.147325933400605e-05,
      "loss": 6.0978,
      "step": 39000
    },
    {
      "epoch": 16.582703610411418,
      "grad_norm": 3.5350372791290283,
      "learning_rate": 4.046417759838547e-05,
      "loss": 6.1027,
      "step": 39500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 4.244364261627197,
      "learning_rate": 3.9455095862764884e-05,
      "loss": 6.0932,
      "step": 40000
    },
    {
      "epoch": 16.79261125104954,
      "eval_accuracy": 0.1522653358752944,
      "eval_loss": 7.17228364944458,
      "eval_runtime": 0.826,
      "eval_samples_per_second": 579.887,
      "eval_steps_per_second": 6.053,
      "step": 40000
    },
    {
      "epoch": 17.002518891687657,
      "grad_norm": 4.679691791534424,
      "learning_rate": 3.84460141271443e-05,
      "loss": 6.0948,
      "step": 40500
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 3.755824565887451,
      "learning_rate": 3.7436932391523715e-05,
      "loss": 6.0711,
      "step": 41000
    },
    {
      "epoch": 17.422334172963897,
      "grad_norm": 2.5312466621398926,
      "learning_rate": 3.642785065590313e-05,
      "loss": 6.068,
      "step": 41500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 3.2410781383514404,
      "learning_rate": 3.541876892028255e-05,
      "loss": 6.0712,
      "step": 42000
    },
    {
      "epoch": 17.842149454240133,
      "grad_norm": 1.638897180557251,
      "learning_rate": 3.440968718466196e-05,
      "loss": 6.0697,
      "step": 42500
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 4.406405448913574,
      "learning_rate": 3.340060544904137e-05,
      "loss": 6.0613,
      "step": 43000
    },
    {
      "epoch": 18.261964735516372,
      "grad_norm": 6.098072528839111,
      "learning_rate": 3.239152371342079e-05,
      "loss": 6.046,
      "step": 43500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 3.8030357360839844,
      "learning_rate": 3.13824419778002e-05,
      "loss": 6.0477,
      "step": 44000
    },
    {
      "epoch": 18.68178001679261,
      "grad_norm": 3.326488733291626,
      "learning_rate": 3.0373360242179615e-05,
      "loss": 6.0474,
      "step": 44500
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 2.1230294704437256,
      "learning_rate": 2.9364278506559034e-05,
      "loss": 6.0495,
      "step": 45000
    },
    {
      "epoch": 19.10159529806885,
      "grad_norm": 4.3825201988220215,
      "learning_rate": 2.8355196770938446e-05,
      "loss": 6.0394,
      "step": 45500
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 4.986055850982666,
      "learning_rate": 2.734611503531786e-05,
      "loss": 6.0234,
      "step": 46000
    },
    {
      "epoch": 19.521410579345087,
      "grad_norm": 5.249682903289795,
      "learning_rate": 2.6337033299697278e-05,
      "loss": 6.0248,
      "step": 46500
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 1.4368300437927246,
      "learning_rate": 2.532795156407669e-05,
      "loss": 6.0251,
      "step": 47000
    },
    {
      "epoch": 19.941225860621326,
      "grad_norm": 3.742292642593384,
      "learning_rate": 2.4318869828456106e-05,
      "loss": 6.0207,
      "step": 47500
    },
    {
      "epoch": 20.151133501259444,
      "grad_norm": 3.802196502685547,
      "learning_rate": 2.330978809283552e-05,
      "loss": 6.0125,
      "step": 48000
    },
    {
      "epoch": 20.361041141897566,
      "grad_norm": 5.184817790985107,
      "learning_rate": 2.2300706357214934e-05,
      "loss": 6.0073,
      "step": 48500
    },
    {
      "epoch": 20.570948782535684,
      "grad_norm": 3.9014132022857666,
      "learning_rate": 2.129162462159435e-05,
      "loss": 6.0045,
      "step": 49000
    },
    {
      "epoch": 20.7808564231738,
      "grad_norm": 4.178951740264893,
      "learning_rate": 2.0282542885973765e-05,
      "loss": 6.0043,
      "step": 49500
    },
    {
      "epoch": 20.990764063811923,
      "grad_norm": 3.3247692584991455,
      "learning_rate": 1.927346115035318e-05,
      "loss": 6.0035,
      "step": 50000
    },
    {
      "epoch": 20.990764063811923,
      "eval_accuracy": 0.1518210110185886,
      "eval_loss": 7.3042073249816895,
      "eval_runtime": 0.8158,
      "eval_samples_per_second": 587.126,
      "eval_steps_per_second": 6.129,
      "step": 50000
    },
    {
      "epoch": 21.20067170445004,
      "grad_norm": 2.6522934436798096,
      "learning_rate": 1.8264379414732593e-05,
      "loss": 5.9908,
      "step": 50500
    },
    {
      "epoch": 21.410579345088163,
      "grad_norm": 4.521886348724365,
      "learning_rate": 1.725529767911201e-05,
      "loss": 5.9951,
      "step": 51000
    },
    {
      "epoch": 21.62048698572628,
      "grad_norm": 3.8180289268493652,
      "learning_rate": 1.6246215943491424e-05,
      "loss": 5.9937,
      "step": 51500
    },
    {
      "epoch": 21.8303946263644,
      "grad_norm": 1.0543891191482544,
      "learning_rate": 1.5237134207870837e-05,
      "loss": 5.9912,
      "step": 52000
    },
    {
      "epoch": 22.04030226700252,
      "grad_norm": 3.1725728511810303,
      "learning_rate": 1.4228052472250252e-05,
      "loss": 5.9876,
      "step": 52500
    },
    {
      "epoch": 22.250209907640638,
      "grad_norm": 4.868059158325195,
      "learning_rate": 1.3218970736629668e-05,
      "loss": 5.9748,
      "step": 53000
    },
    {
      "epoch": 22.460117548278756,
      "grad_norm": 1.7380059957504272,
      "learning_rate": 1.2209889001009082e-05,
      "loss": 5.9751,
      "step": 53500
    },
    {
      "epoch": 22.670025188916878,
      "grad_norm": 1.1676714420318604,
      "learning_rate": 1.1200807265388496e-05,
      "loss": 5.9792,
      "step": 54000
    },
    {
      "epoch": 22.879932829554996,
      "grad_norm": 0.9109513759613037,
      "learning_rate": 1.0191725529767912e-05,
      "loss": 5.9726,
      "step": 54500
    },
    {
      "epoch": 23.089840470193113,
      "grad_norm": 0.9858363270759583,
      "learning_rate": 9.182643794147326e-06,
      "loss": 5.9699,
      "step": 55000
    },
    {
      "epoch": 23.299748110831235,
      "grad_norm": 4.528422832489014,
      "learning_rate": 8.173562058526741e-06,
      "loss": 5.9672,
      "step": 55500
    },
    {
      "epoch": 23.509655751469353,
      "grad_norm": 4.258862495422363,
      "learning_rate": 7.164480322906156e-06,
      "loss": 5.966,
      "step": 56000
    },
    {
      "epoch": 23.719563392107474,
      "grad_norm": 3.3200228214263916,
      "learning_rate": 6.155398587285571e-06,
      "loss": 5.9601,
      "step": 56500
    },
    {
      "epoch": 23.929471032745592,
      "grad_norm": 2.4026434421539307,
      "learning_rate": 5.146316851664986e-06,
      "loss": 5.965,
      "step": 57000
    },
    {
      "epoch": 24.13937867338371,
      "grad_norm": 2.701397180557251,
      "learning_rate": 4.1372351160444e-06,
      "loss": 5.9575,
      "step": 57500
    },
    {
      "epoch": 24.349286314021832,
      "grad_norm": 3.0246880054473877,
      "learning_rate": 3.1281533804238145e-06,
      "loss": 5.9596,
      "step": 58000
    },
    {
      "epoch": 24.55919395465995,
      "grad_norm": 2.0782816410064697,
      "learning_rate": 2.1190716448032293e-06,
      "loss": 5.9539,
      "step": 58500
    },
    {
      "epoch": 24.769101595298068,
      "grad_norm": 1.3083659410476685,
      "learning_rate": 1.109989909182644e-06,
      "loss": 5.9524,
      "step": 59000
    },
    {
      "epoch": 24.97900923593619,
      "grad_norm": 3.437366485595703,
      "learning_rate": 1.0090817356205853e-07,
      "loss": 5.953,
      "step": 59500
    },
    {
      "epoch": 25.0,
      "step": 59550,
      "total_flos": 7.68761901614592e+17,
      "train_loss": 6.575524925404892,
      "train_runtime": 17044.2581,
      "train_samples_per_second": 335.361,
      "train_steps_per_second": 3.494
    }
  ],
  "logging_steps": 500,
  "max_steps": 59550,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 10000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.68761901614592e+17,
  "train_batch_size": 96,
  "trial_name": null,
  "trial_params": null
}