{
  "best_global_step": 10000,
  "best_metric": 6.876064300537109,
  "best_model_checkpoint": "distilbert_km_50_v2/checkpoint-10000",
  "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": 2.5942015647888184,
      "learning_rate": 5e-06,
      "loss": 13.2109,
      "step": 500
    },
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 3.0829241275787354,
      "learning_rate": 1e-05,
      "loss": 11.41,
      "step": 1000
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 3.9222471714019775,
      "learning_rate": 1.5e-05,
      "loss": 10.0738,
      "step": 1500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 13.29185676574707,
      "learning_rate": 2e-05,
      "loss": 9.1115,
      "step": 2000
    },
    {
      "epoch": 1.0495382031905962,
      "grad_norm": 7.1462860107421875,
      "learning_rate": 2.5e-05,
      "loss": 8.5507,
      "step": 2500
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 6.569362163543701,
      "learning_rate": 3e-05,
      "loss": 8.1902,
      "step": 3000
    },
    {
      "epoch": 1.4693534844668346,
      "grad_norm": 10.645280838012695,
      "learning_rate": 3.5e-05,
      "loss": 7.8821,
      "step": 3500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 4.802201271057129,
      "learning_rate": 4e-05,
      "loss": 7.6653,
      "step": 4000
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 4.122864246368408,
      "learning_rate": 4.5e-05,
      "loss": 7.5087,
      "step": 4500
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 10.079947471618652,
      "learning_rate": 5e-05,
      "loss": 7.3566,
      "step": 5000
    },
    {
      "epoch": 2.3089840470193117,
      "grad_norm": 5.222455024719238,
      "learning_rate": 5.500000000000001e-05,
      "loss": 7.2419,
      "step": 5500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 5.6451497077941895,
      "learning_rate": 6e-05,
      "loss": 7.2088,
      "step": 6000
    },
    {
      "epoch": 2.72879932829555,
      "grad_norm": 5.035846710205078,
      "learning_rate": 6.500000000000001e-05,
      "loss": 7.1723,
      "step": 6500
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 4.1157097816467285,
      "learning_rate": 7e-05,
      "loss": 7.1444,
      "step": 7000
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 5.347066879272461,
      "learning_rate": 7.500000000000001e-05,
      "loss": 7.0179,
      "step": 7500
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 6.044243335723877,
      "learning_rate": 8e-05,
      "loss": 6.9765,
      "step": 8000
    },
    {
      "epoch": 3.568429890848027,
      "grad_norm": 4.617038726806641,
      "learning_rate": 8.5e-05,
      "loss": 6.9976,
      "step": 8500
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 5.0755085945129395,
      "learning_rate": 9e-05,
      "loss": 6.9865,
      "step": 9000
    },
    {
      "epoch": 3.988245172124265,
      "grad_norm": 2.692549228668213,
      "learning_rate": 9.5e-05,
      "loss": 6.9816,
      "step": 9500
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 3.265491008758545,
      "learning_rate": 0.0001,
      "loss": 6.8364,
      "step": 10000
    },
    {
      "epoch": 4.198152812762385,
      "eval_accuracy": 0.15036589777122347,
      "eval_loss": 6.876064300537109,
      "eval_runtime": 0.8375,
      "eval_samples_per_second": 571.954,
      "eval_steps_per_second": 5.97,
      "step": 10000
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 4.12289571762085,
      "learning_rate": 9.899091826437942e-05,
      "loss": 6.8537,
      "step": 10500
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 5.4416680335998535,
      "learning_rate": 9.798183652875883e-05,
      "loss": 6.862,
      "step": 11000
    },
    {
      "epoch": 4.827875734676742,
      "grad_norm": 3.5069587230682373,
      "learning_rate": 9.697275479313825e-05,
      "loss": 6.8438,
      "step": 11500
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 3.045193910598755,
      "learning_rate": 9.596367305751767e-05,
      "loss": 6.8096,
      "step": 12000
    },
    {
      "epoch": 5.247691015952981,
      "grad_norm": 5.9380998611450195,
      "learning_rate": 9.495459132189707e-05,
      "loss": 6.6923,
      "step": 12500
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 4.184321403503418,
      "learning_rate": 9.394550958627649e-05,
      "loss": 6.6988,
      "step": 13000
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 4.107109546661377,
      "learning_rate": 9.293642785065591e-05,
      "loss": 6.7132,
      "step": 13500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 3.741273880004883,
      "learning_rate": 9.192734611503532e-05,
      "loss": 6.7124,
      "step": 14000
    },
    {
      "epoch": 6.087321578505458,
      "grad_norm": 2.995544195175171,
      "learning_rate": 9.091826437941474e-05,
      "loss": 6.6456,
      "step": 14500
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 3.872407913208008,
      "learning_rate": 8.990918264379415e-05,
      "loss": 6.5725,
      "step": 15000
    },
    {
      "epoch": 6.507136859781696,
      "grad_norm": 2.6585052013397217,
      "learning_rate": 8.890010090817356e-05,
      "loss": 6.5881,
      "step": 15500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 3.190882921218872,
      "learning_rate": 8.789101917255298e-05,
      "loss": 6.5996,
      "step": 16000
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 3.574570894241333,
      "learning_rate": 8.68819374369324e-05,
      "loss": 6.5989,
      "step": 16500
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 3.1599719524383545,
      "learning_rate": 8.58728557013118e-05,
      "loss": 6.5096,
      "step": 17000
    },
    {
      "epoch": 7.346767422334173,
      "grad_norm": 3.929410457611084,
      "learning_rate": 8.486377396569122e-05,
      "loss": 6.4787,
      "step": 17500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 3.347877264022827,
      "learning_rate": 8.385469223007064e-05,
      "loss": 6.4918,
      "step": 18000
    },
    {
      "epoch": 7.766582703610411,
      "grad_norm": 2.7393689155578613,
      "learning_rate": 8.284561049445006e-05,
      "loss": 6.4952,
      "step": 18500
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 4.251912593841553,
      "learning_rate": 8.183652875882947e-05,
      "loss": 6.5015,
      "step": 19000
    },
    {
      "epoch": 8.18639798488665,
      "grad_norm": 2.8671343326568604,
      "learning_rate": 8.082744702320889e-05,
      "loss": 6.3982,
      "step": 19500
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 3.1719858646392822,
      "learning_rate": 7.98183652875883e-05,
      "loss": 6.4025,
      "step": 20000
    },
    {
      "epoch": 8.39630562552477,
      "eval_accuracy": 0.15160358034737226,
      "eval_loss": 6.9073638916015625,
      "eval_runtime": 0.8356,
      "eval_samples_per_second": 573.232,
      "eval_steps_per_second": 5.984,
      "step": 20000
    },
    {
      "epoch": 8.606213266162888,
      "grad_norm": 3.0642356872558594,
      "learning_rate": 7.880928355196771e-05,
      "loss": 6.4029,
      "step": 20500
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 4.562249660491943,
      "learning_rate": 7.780020181634713e-05,
      "loss": 6.4171,
      "step": 21000
    },
    {
      "epoch": 9.026028547439127,
      "grad_norm": 3.9007492065429688,
      "learning_rate": 7.679112008072655e-05,
      "loss": 6.4083,
      "step": 21500
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 3.7150356769561768,
      "learning_rate": 7.578203834510595e-05,
      "loss": 6.3229,
      "step": 22000
    },
    {
      "epoch": 9.445843828715365,
      "grad_norm": 4.066181659698486,
      "learning_rate": 7.477295660948537e-05,
      "loss": 6.3272,
      "step": 22500
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 3.8810086250305176,
      "learning_rate": 7.376387487386479e-05,
      "loss": 6.3347,
      "step": 23000
    },
    {
      "epoch": 9.865659109991604,
      "grad_norm": 3.259021759033203,
      "learning_rate": 7.27547931382442e-05,
      "loss": 6.3426,
      "step": 23500
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 3.6049447059631348,
      "learning_rate": 7.174571140262362e-05,
      "loss": 6.3148,
      "step": 24000
    },
    {
      "epoch": 10.285474391267842,
      "grad_norm": 4.684219837188721,
      "learning_rate": 7.073662966700304e-05,
      "loss": 6.2732,
      "step": 24500
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 3.5771358013153076,
      "learning_rate": 6.972754793138244e-05,
      "loss": 6.2751,
      "step": 25000
    },
    {
      "epoch": 10.705289672544081,
      "grad_norm": 5.011735916137695,
      "learning_rate": 6.871846619576186e-05,
      "loss": 6.2767,
      "step": 25500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 3.1461920738220215,
      "learning_rate": 6.770938446014128e-05,
      "loss": 6.2834,
      "step": 26000
    },
    {
      "epoch": 11.125104953820319,
      "grad_norm": 5.479197025299072,
      "learning_rate": 6.670030272452068e-05,
      "loss": 6.2464,
      "step": 26500
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 4.422645568847656,
      "learning_rate": 6.56912209889001e-05,
      "loss": 6.2232,
      "step": 27000
    },
    {
      "epoch": 11.544920235096557,
      "grad_norm": 4.456131458282471,
      "learning_rate": 6.468213925327952e-05,
      "loss": 6.2254,
      "step": 27500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 4.289433479309082,
      "learning_rate": 6.367305751765893e-05,
      "loss": 6.2265,
      "step": 28000
    },
    {
      "epoch": 11.964735516372796,
      "grad_norm": 4.341806411743164,
      "learning_rate": 6.266397578203835e-05,
      "loss": 6.2293,
      "step": 28500
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 2.852874755859375,
      "learning_rate": 6.165489404641777e-05,
      "loss": 6.1935,
      "step": 29000
    },
    {
      "epoch": 12.384550797649034,
      "grad_norm": 3.42668080329895,
      "learning_rate": 6.064581231079718e-05,
      "loss": 6.1776,
      "step": 29500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 3.047215700149536,
      "learning_rate": 5.963673057517659e-05,
      "loss": 6.1886,
      "step": 30000
    },
    {
      "epoch": 12.594458438287154,
      "eval_accuracy": 0.15434983350589027,
      "eval_loss": 6.9720139503479,
      "eval_runtime": 0.8333,
      "eval_samples_per_second": 574.789,
      "eval_steps_per_second": 6.0,
      "step": 30000
    },
    {
      "epoch": 12.804366078925273,
      "grad_norm": 4.062623977661133,
      "learning_rate": 5.8627648839556e-05,
      "loss": 6.1831,
      "step": 30500
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 2.6942687034606934,
      "learning_rate": 5.7618567103935416e-05,
      "loss": 6.1829,
      "step": 31000
    },
    {
      "epoch": 13.224181360201511,
      "grad_norm": 5.287597179412842,
      "learning_rate": 5.660948536831484e-05,
      "loss": 6.1502,
      "step": 31500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 3.7545247077941895,
      "learning_rate": 5.5600403632694254e-05,
      "loss": 6.1447,
      "step": 32000
    },
    {
      "epoch": 13.64399664147775,
      "grad_norm": 2.194673538208008,
      "learning_rate": 5.4591321897073666e-05,
      "loss": 6.1526,
      "step": 32500
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 3.4157662391662598,
      "learning_rate": 5.358224016145308e-05,
      "loss": 6.1503,
      "step": 33000
    },
    {
      "epoch": 14.063811922753988,
      "grad_norm": 3.2840874195098877,
      "learning_rate": 5.257315842583249e-05,
      "loss": 6.1373,
      "step": 33500
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 3.1655707359313965,
      "learning_rate": 5.15640766902119e-05,
      "loss": 6.1129,
      "step": 34000
    },
    {
      "epoch": 14.483627204030226,
      "grad_norm": 3.0090761184692383,
      "learning_rate": 5.055499495459133e-05,
      "loss": 6.1174,
      "step": 34500
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 3.265897750854492,
      "learning_rate": 4.954591321897074e-05,
      "loss": 6.1142,
      "step": 35000
    },
    {
      "epoch": 14.903442485306465,
      "grad_norm": 2.526585340499878,
      "learning_rate": 4.853683148335015e-05,
      "loss": 6.1158,
      "step": 35500
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 3.0448718070983887,
      "learning_rate": 4.7527749747729566e-05,
      "loss": 6.1022,
      "step": 36000
    },
    {
      "epoch": 15.323257766582703,
      "grad_norm": 2.9147238731384277,
      "learning_rate": 4.6518668012108985e-05,
      "loss": 6.0919,
      "step": 36500
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 2.7290289402008057,
      "learning_rate": 4.55095862764884e-05,
      "loss": 6.0861,
      "step": 37000
    },
    {
      "epoch": 15.743073047858942,
      "grad_norm": 4.233330726623535,
      "learning_rate": 4.450050454086781e-05,
      "loss": 6.0909,
      "step": 37500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 2.52457594871521,
      "learning_rate": 4.349142280524723e-05,
      "loss": 6.0872,
      "step": 38000
    },
    {
      "epoch": 16.162888329135182,
      "grad_norm": 4.0439133644104,
      "learning_rate": 4.248234106962664e-05,
      "loss": 6.0715,
      "step": 38500
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 1.935972809791565,
      "learning_rate": 4.147325933400605e-05,
      "loss": 6.0652,
      "step": 39000
    },
    {
      "epoch": 16.582703610411418,
      "grad_norm": 3.7831125259399414,
      "learning_rate": 4.046417759838547e-05,
      "loss": 6.0712,
      "step": 39500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 2.656057834625244,
      "learning_rate": 3.9455095862764884e-05,
      "loss": 6.0593,
      "step": 40000
    },
    {
      "epoch": 16.79261125104954,
      "eval_accuracy": 0.15220926320511383,
      "eval_loss": 7.052610874176025,
      "eval_runtime": 0.9038,
      "eval_samples_per_second": 530.01,
      "eval_steps_per_second": 5.532,
      "step": 40000
    },
    {
      "epoch": 17.002518891687657,
      "grad_norm": 1.914518117904663,
      "learning_rate": 3.84460141271443e-05,
      "loss": 6.0658,
      "step": 40500
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 2.8532705307006836,
      "learning_rate": 3.7436932391523715e-05,
      "loss": 6.0438,
      "step": 41000
    },
    {
      "epoch": 17.422334172963897,
      "grad_norm": 3.562371015548706,
      "learning_rate": 3.642785065590313e-05,
      "loss": 6.0399,
      "step": 41500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 2.576478958129883,
      "learning_rate": 3.541876892028255e-05,
      "loss": 6.0412,
      "step": 42000
    },
    {
      "epoch": 17.842149454240133,
      "grad_norm": 2.1456174850463867,
      "learning_rate": 3.440968718466196e-05,
      "loss": 6.0393,
      "step": 42500
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 2.9254274368286133,
      "learning_rate": 3.340060544904137e-05,
      "loss": 6.0328,
      "step": 43000
    },
    {
      "epoch": 18.261964735516372,
      "grad_norm": 1.6633203029632568,
      "learning_rate": 3.239152371342079e-05,
      "loss": 6.0224,
      "step": 43500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 3.052461624145508,
      "learning_rate": 3.13824419778002e-05,
      "loss": 6.0194,
      "step": 44000
    },
    {
      "epoch": 18.68178001679261,
      "grad_norm": 5.925746917724609,
      "learning_rate": 3.0373360242179615e-05,
      "loss": 6.0228,
      "step": 44500
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 3.373609781265259,
      "learning_rate": 2.9364278506559034e-05,
      "loss": 6.0236,
      "step": 45000
    },
    {
      "epoch": 19.10159529806885,
      "grad_norm": 3.2891108989715576,
      "learning_rate": 2.8355196770938446e-05,
      "loss": 6.0156,
      "step": 45500
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 1.40165376663208,
      "learning_rate": 2.734611503531786e-05,
      "loss": 6.0019,
      "step": 46000
    },
    {
      "epoch": 19.521410579345087,
      "grad_norm": 3.0847766399383545,
      "learning_rate": 2.6337033299697278e-05,
      "loss": 6.0044,
      "step": 46500
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 1.0164066553115845,
      "learning_rate": 2.532795156407669e-05,
      "loss": 6.0028,
      "step": 47000
    },
    {
      "epoch": 19.941225860621326,
      "grad_norm": 3.7578606605529785,
      "learning_rate": 2.4318869828456106e-05,
      "loss": 5.9992,
      "step": 47500
    },
    {
      "epoch": 20.151133501259444,
      "grad_norm": 1.7513774633407593,
      "learning_rate": 2.330978809283552e-05,
      "loss": 5.9925,
      "step": 48000
    },
    {
      "epoch": 20.361041141897566,
      "grad_norm": 3.49128794670105,
      "learning_rate": 2.2300706357214934e-05,
      "loss": 5.9884,
      "step": 48500
    },
    {
      "epoch": 20.570948782535684,
      "grad_norm": 1.9832980632781982,
      "learning_rate": 2.129162462159435e-05,
      "loss": 5.983,
      "step": 49000
    },
    {
      "epoch": 20.7808564231738,
      "grad_norm": 1.4352755546569824,
      "learning_rate": 2.0282542885973765e-05,
      "loss": 5.9829,
      "step": 49500
    },
    {
      "epoch": 20.990764063811923,
      "grad_norm": 2.9469616413116455,
      "learning_rate": 1.927346115035318e-05,
      "loss": 5.9809,
      "step": 50000
    },
    {
      "epoch": 20.990764063811923,
      "eval_accuracy": 0.15156867692825302,
      "eval_loss": 7.026661396026611,
      "eval_runtime": 0.8225,
      "eval_samples_per_second": 582.38,
      "eval_steps_per_second": 6.079,
      "step": 50000
    },
    {
      "epoch": 21.20067170445004,
      "grad_norm": 0.8712583184242249,
      "learning_rate": 1.8264379414732593e-05,
      "loss": 5.9747,
      "step": 50500
    },
    {
      "epoch": 21.410579345088163,
      "grad_norm": 5.054527282714844,
      "learning_rate": 1.725529767911201e-05,
      "loss": 5.9737,
      "step": 51000
    },
    {
      "epoch": 21.62048698572628,
      "grad_norm": 0.8384255170822144,
      "learning_rate": 1.6246215943491424e-05,
      "loss": 5.9717,
      "step": 51500
    },
    {
      "epoch": 21.8303946263644,
      "grad_norm": 2.0753424167633057,
      "learning_rate": 1.5237134207870837e-05,
      "loss": 5.9714,
      "step": 52000
    },
    {
      "epoch": 22.04030226700252,
      "grad_norm": 1.4410223960876465,
      "learning_rate": 1.4228052472250252e-05,
      "loss": 5.9691,
      "step": 52500
    },
    {
      "epoch": 22.250209907640638,
      "grad_norm": 3.0846803188323975,
      "learning_rate": 1.3218970736629668e-05,
      "loss": 5.9584,
      "step": 53000
    },
    {
      "epoch": 22.460117548278756,
      "grad_norm": 2.658273220062256,
      "learning_rate": 1.2209889001009082e-05,
      "loss": 5.9589,
      "step": 53500
    },
    {
      "epoch": 22.670025188916878,
      "grad_norm": 0.9044545888900757,
      "learning_rate": 1.1200807265388496e-05,
      "loss": 5.9611,
      "step": 54000
    },
    {
      "epoch": 22.879932829554996,
      "grad_norm": 3.3916687965393066,
      "learning_rate": 1.0191725529767912e-05,
      "loss": 5.9593,
      "step": 54500
    },
    {
      "epoch": 23.089840470193113,
      "grad_norm": 1.2624566555023193,
      "learning_rate": 9.182643794147326e-06,
      "loss": 5.9536,
      "step": 55000
    },
    {
      "epoch": 23.299748110831235,
      "grad_norm": 1.0515649318695068,
      "learning_rate": 8.173562058526741e-06,
      "loss": 5.9499,
      "step": 55500
    },
    {
      "epoch": 23.509655751469353,
      "grad_norm": 0.9997740387916565,
      "learning_rate": 7.164480322906156e-06,
      "loss": 5.9518,
      "step": 56000
    },
    {
      "epoch": 23.719563392107474,
      "grad_norm": 1.5580836534500122,
      "learning_rate": 6.155398587285571e-06,
      "loss": 5.9447,
      "step": 56500
    },
    {
      "epoch": 23.929471032745592,
      "grad_norm": 2.255404233932495,
      "learning_rate": 5.146316851664986e-06,
      "loss": 5.9476,
      "step": 57000
    },
    {
      "epoch": 24.13937867338371,
      "grad_norm": 1.1771949529647827,
      "learning_rate": 4.1372351160444e-06,
      "loss": 5.9435,
      "step": 57500
    },
    {
      "epoch": 24.349286314021832,
      "grad_norm": 2.311717987060547,
      "learning_rate": 3.1281533804238145e-06,
      "loss": 5.9452,
      "step": 58000
    },
    {
      "epoch": 24.55919395465995,
      "grad_norm": 1.5345447063446045,
      "learning_rate": 2.1190716448032293e-06,
      "loss": 5.9411,
      "step": 58500
    },
    {
      "epoch": 24.769101595298068,
      "grad_norm": 1.1828807592391968,
      "learning_rate": 1.109989909182644e-06,
      "loss": 5.9403,
      "step": 59000
    },
    {
      "epoch": 24.97900923593619,
      "grad_norm": 1.2583199739456177,
      "learning_rate": 1.0090817356205853e-07,
      "loss": 5.9391,
      "step": 59500
    },
    {
      "epoch": 25.0,
      "step": 59550,
      "total_flos": 7.68761901614592e+17,
      "train_loss": 6.515696689403928,
      "train_runtime": 16851.789,
      "train_samples_per_second": 339.191,
      "train_steps_per_second": 3.534
    }
  ],
  "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
}