{
  "best_global_step": 40000,
  "best_metric": 6.1867241859436035,
  "best_model_checkpoint": "bert_base_km_5_v2/checkpoint-40000",
  "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": 6.0301618576049805,
      "learning_rate": 5e-06,
      "loss": 10.757,
      "step": 500
    },
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 16.769906997680664,
      "learning_rate": 1e-05,
      "loss": 8.7451,
      "step": 1000
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 7.1726531982421875,
      "learning_rate": 1.5e-05,
      "loss": 7.8269,
      "step": 1500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 6.693809509277344,
      "learning_rate": 2e-05,
      "loss": 7.4514,
      "step": 2000
    },
    {
      "epoch": 1.0495382031905962,
      "grad_norm": 9.91014289855957,
      "learning_rate": 2.5e-05,
      "loss": 7.2691,
      "step": 2500
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 10.000188827514648,
      "learning_rate": 3e-05,
      "loss": 7.0885,
      "step": 3000
    },
    {
      "epoch": 1.4693534844668346,
      "grad_norm": 3.8296265602111816,
      "learning_rate": 3.5e-05,
      "loss": 6.9553,
      "step": 3500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 1.9357906579971313,
      "learning_rate": 4e-05,
      "loss": 6.8586,
      "step": 4000
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 2.881507396697998,
      "learning_rate": 4.5e-05,
      "loss": 6.7796,
      "step": 4500
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 3.3740458488464355,
      "learning_rate": 5e-05,
      "loss": 6.6888,
      "step": 5000
    },
    {
      "epoch": 2.3089840470193117,
      "grad_norm": 2.4316248893737793,
      "learning_rate": 5.500000000000001e-05,
      "loss": 6.633,
      "step": 5500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 2.016279697418213,
      "learning_rate": 6e-05,
      "loss": 6.5985,
      "step": 6000
    },
    {
      "epoch": 2.72879932829555,
      "grad_norm": 1.9014559984207153,
      "learning_rate": 6.500000000000001e-05,
      "loss": 6.5631,
      "step": 6500
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 3.7373297214508057,
      "learning_rate": 7e-05,
      "loss": 6.5429,
      "step": 7000
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 2.4893252849578857,
      "learning_rate": 7.500000000000001e-05,
      "loss": 6.4796,
      "step": 7500
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 1.6236482858657837,
      "learning_rate": 8e-05,
      "loss": 6.4555,
      "step": 8000
    },
    {
      "epoch": 3.568429890848027,
      "grad_norm": 1.7691621780395508,
      "learning_rate": 8.5e-05,
      "loss": 6.4502,
      "step": 8500
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 2.8690245151519775,
      "learning_rate": 9e-05,
      "loss": 6.4298,
      "step": 9000
    },
    {
      "epoch": 3.988245172124265,
      "grad_norm": 2.252593994140625,
      "learning_rate": 9.5e-05,
      "loss": 6.4103,
      "step": 9500
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 2.1625120639801025,
      "learning_rate": 0.0001,
      "loss": 6.352,
      "step": 10000
    },
    {
      "epoch": 4.198152812762385,
      "eval_accuracy": 0.1499485238876986,
      "eval_loss": 6.380714416503906,
      "eval_runtime": 1.0285,
      "eval_samples_per_second": 465.727,
      "eval_steps_per_second": 4.861,
      "step": 10000
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 1.6482921838760376,
      "learning_rate": 9.899091826437942e-05,
      "loss": 6.3442,
      "step": 10500
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 4.066359996795654,
      "learning_rate": 9.798183652875883e-05,
      "loss": 6.3277,
      "step": 11000
    },
    {
      "epoch": 4.827875734676742,
      "grad_norm": 2.1703338623046875,
      "learning_rate": 9.697275479313825e-05,
      "loss": 6.3219,
      "step": 11500
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 2.1197056770324707,
      "learning_rate": 9.596367305751767e-05,
      "loss": 6.2973,
      "step": 12000
    },
    {
      "epoch": 5.247691015952981,
      "grad_norm": 1.4740880727767944,
      "learning_rate": 9.495459132189707e-05,
      "loss": 6.2432,
      "step": 12500
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 1.7484091520309448,
      "learning_rate": 9.394550958627649e-05,
      "loss": 6.2353,
      "step": 13000
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 3.0082826614379883,
      "learning_rate": 9.293642785065591e-05,
      "loss": 6.2378,
      "step": 13500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 1.2044631242752075,
      "learning_rate": 9.192734611503532e-05,
      "loss": 6.2257,
      "step": 14000
    },
    {
      "epoch": 6.087321578505458,
      "grad_norm": 2.8152501583099365,
      "learning_rate": 9.091826437941474e-05,
      "loss": 6.204,
      "step": 14500
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 1.4020590782165527,
      "learning_rate": 8.990918264379415e-05,
      "loss": 6.1749,
      "step": 15000
    },
    {
      "epoch": 6.507136859781696,
      "grad_norm": 1.2158809900283813,
      "learning_rate": 8.890010090817356e-05,
      "loss": 6.1668,
      "step": 15500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 1.8693963289260864,
      "learning_rate": 8.789101917255298e-05,
      "loss": 6.1626,
      "step": 16000
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 1.0617125034332275,
      "learning_rate": 8.68819374369324e-05,
      "loss": 6.1578,
      "step": 16500
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 0.9940109252929688,
      "learning_rate": 8.58728557013118e-05,
      "loss": 6.1247,
      "step": 17000
    },
    {
      "epoch": 7.346767422334173,
      "grad_norm": 1.0228919982910156,
      "learning_rate": 8.486377396569122e-05,
      "loss": 6.1061,
      "step": 17500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 2.0107200145721436,
      "learning_rate": 8.385469223007064e-05,
      "loss": 6.1092,
      "step": 18000
    },
    {
      "epoch": 7.766582703610411,
      "grad_norm": 3.047442674636841,
      "learning_rate": 8.284561049445006e-05,
      "loss": 6.098,
      "step": 18500
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 1.0168153047561646,
      "learning_rate": 8.183652875882947e-05,
      "loss": 6.1014,
      "step": 19000
    },
    {
      "epoch": 8.18639798488665,
      "grad_norm": 2.082069158554077,
      "learning_rate": 8.082744702320889e-05,
      "loss": 6.063,
      "step": 19500
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 2.421567678451538,
      "learning_rate": 7.98183652875883e-05,
      "loss": 6.0656,
      "step": 20000
    },
    {
      "epoch": 8.39630562552477,
      "eval_accuracy": 0.15266982799741854,
      "eval_loss": 6.24129056930542,
      "eval_runtime": 1.0077,
      "eval_samples_per_second": 475.335,
      "eval_steps_per_second": 4.962,
      "step": 20000
    },
    {
      "epoch": 8.606213266162888,
      "grad_norm": 3.8225059509277344,
      "learning_rate": 7.880928355196771e-05,
      "loss": 6.0564,
      "step": 20500
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 1.3910123109817505,
      "learning_rate": 7.780020181634713e-05,
      "loss": 6.0583,
      "step": 21000
    },
    {
      "epoch": 9.026028547439127,
      "grad_norm": 0.8005045056343079,
      "learning_rate": 7.679112008072655e-05,
      "loss": 6.0456,
      "step": 21500
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 1.752871036529541,
      "learning_rate": 7.578203834510595e-05,
      "loss": 6.0205,
      "step": 22000
    },
    {
      "epoch": 9.445843828715365,
      "grad_norm": 1.8086637258529663,
      "learning_rate": 7.477295660948537e-05,
      "loss": 6.0171,
      "step": 22500
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 1.1677097082138062,
      "learning_rate": 7.376387487386479e-05,
      "loss": 6.0156,
      "step": 23000
    },
    {
      "epoch": 9.865659109991604,
      "grad_norm": 0.7790222764015198,
      "learning_rate": 7.27547931382442e-05,
      "loss": 6.0096,
      "step": 23500
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 1.6951969861984253,
      "learning_rate": 7.174571140262362e-05,
      "loss": 5.9932,
      "step": 24000
    },
    {
      "epoch": 10.285474391267842,
      "grad_norm": 0.7763303518295288,
      "learning_rate": 7.073662966700304e-05,
      "loss": 5.9879,
      "step": 24500
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 2.300548553466797,
      "learning_rate": 6.972754793138244e-05,
      "loss": 5.977,
      "step": 25000
    },
    {
      "epoch": 10.705289672544081,
      "grad_norm": 1.5366549491882324,
      "learning_rate": 6.871846619576186e-05,
      "loss": 5.9785,
      "step": 25500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 2.3130111694335938,
      "learning_rate": 6.770938446014128e-05,
      "loss": 5.9786,
      "step": 26000
    },
    {
      "epoch": 11.125104953820319,
      "grad_norm": 2.656207323074341,
      "learning_rate": 6.670030272452068e-05,
      "loss": 5.9583,
      "step": 26500
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 1.0905320644378662,
      "learning_rate": 6.56912209889001e-05,
      "loss": 5.9525,
      "step": 27000
    },
    {
      "epoch": 11.544920235096557,
      "grad_norm": 2.9984869956970215,
      "learning_rate": 6.468213925327952e-05,
      "loss": 5.9487,
      "step": 27500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 2.012657403945923,
      "learning_rate": 6.367305751765893e-05,
      "loss": 5.9448,
      "step": 28000
    },
    {
      "epoch": 11.964735516372796,
      "grad_norm": 1.4100701808929443,
      "learning_rate": 6.266397578203835e-05,
      "loss": 5.9437,
      "step": 28500
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 2.330237627029419,
      "learning_rate": 6.165489404641777e-05,
      "loss": 5.9292,
      "step": 29000
    },
    {
      "epoch": 12.384550797649034,
      "grad_norm": 0.9418126940727234,
      "learning_rate": 6.064581231079718e-05,
      "loss": 5.9212,
      "step": 29500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 0.7583677768707275,
      "learning_rate": 5.963673057517659e-05,
      "loss": 5.9257,
      "step": 30000
    },
    {
      "epoch": 12.594458438287154,
      "eval_accuracy": 0.15387413604947253,
      "eval_loss": 6.210723400115967,
      "eval_runtime": 1.0127,
      "eval_samples_per_second": 473.005,
      "eval_steps_per_second": 4.937,
      "step": 30000
    },
    {
      "epoch": 12.804366078925273,
      "grad_norm": 1.4112985134124756,
      "learning_rate": 5.8627648839556e-05,
      "loss": 5.9188,
      "step": 30500
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 1.1568245887756348,
      "learning_rate": 5.7618567103935416e-05,
      "loss": 5.9216,
      "step": 31000
    },
    {
      "epoch": 13.224181360201511,
      "grad_norm": 0.8279080986976624,
      "learning_rate": 5.660948536831484e-05,
      "loss": 5.9045,
      "step": 31500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 1.3724620342254639,
      "learning_rate": 5.5600403632694254e-05,
      "loss": 5.8996,
      "step": 32000
    },
    {
      "epoch": 13.64399664147775,
      "grad_norm": 2.230830430984497,
      "learning_rate": 5.4591321897073666e-05,
      "loss": 5.9041,
      "step": 32500
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 0.8925848007202148,
      "learning_rate": 5.358224016145308e-05,
      "loss": 5.8975,
      "step": 33000
    },
    {
      "epoch": 14.063811922753988,
      "grad_norm": 0.7499693036079407,
      "learning_rate": 5.257315842583249e-05,
      "loss": 5.8893,
      "step": 33500
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 0.97462397813797,
      "learning_rate": 5.15640766902119e-05,
      "loss": 5.8815,
      "step": 34000
    },
    {
      "epoch": 14.483627204030226,
      "grad_norm": 1.4555714130401611,
      "learning_rate": 5.055499495459133e-05,
      "loss": 5.8789,
      "step": 34500
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 0.7514633536338806,
      "learning_rate": 4.954591321897074e-05,
      "loss": 5.8786,
      "step": 35000
    },
    {
      "epoch": 14.903442485306465,
      "grad_norm": 1.0099568367004395,
      "learning_rate": 4.853683148335015e-05,
      "loss": 5.8779,
      "step": 35500
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 0.7769736647605896,
      "learning_rate": 4.7527749747729566e-05,
      "loss": 5.8676,
      "step": 36000
    },
    {
      "epoch": 15.323257766582703,
      "grad_norm": 1.894301176071167,
      "learning_rate": 4.6518668012108985e-05,
      "loss": 5.8676,
      "step": 36500
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 1.4337573051452637,
      "learning_rate": 4.55095862764884e-05,
      "loss": 5.8611,
      "step": 37000
    },
    {
      "epoch": 15.743073047858942,
      "grad_norm": 1.3730665445327759,
      "learning_rate": 4.450050454086781e-05,
      "loss": 5.8587,
      "step": 37500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 1.0250178575515747,
      "learning_rate": 4.349142280524723e-05,
      "loss": 5.8556,
      "step": 38000
    },
    {
      "epoch": 16.162888329135182,
      "grad_norm": 2.5980072021484375,
      "learning_rate": 4.248234106962664e-05,
      "loss": 5.8543,
      "step": 38500
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 0.729290246963501,
      "learning_rate": 4.147325933400605e-05,
      "loss": 5.8439,
      "step": 39000
    },
    {
      "epoch": 16.582703610411418,
      "grad_norm": 0.7251316905021667,
      "learning_rate": 4.046417759838547e-05,
      "loss": 5.8477,
      "step": 39500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 0.9639233350753784,
      "learning_rate": 3.9455095862764884e-05,
      "loss": 5.8375,
      "step": 40000
    },
    {
      "epoch": 16.79261125104954,
      "eval_accuracy": 0.1535830436245374,
      "eval_loss": 6.1867241859436035,
      "eval_runtime": 1.0139,
      "eval_samples_per_second": 472.432,
      "eval_steps_per_second": 4.931,
      "step": 40000
    },
    {
      "epoch": 17.002518891687657,
      "grad_norm": 0.748256266117096,
      "learning_rate": 3.84460141271443e-05,
      "loss": 5.8411,
      "step": 40500
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 1.0267351865768433,
      "learning_rate": 3.7436932391523715e-05,
      "loss": 5.8325,
      "step": 41000
    },
    {
      "epoch": 17.422334172963897,
      "grad_norm": 0.7656574845314026,
      "learning_rate": 3.642785065590313e-05,
      "loss": 5.8291,
      "step": 41500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 1.3526027202606201,
      "learning_rate": 3.541876892028255e-05,
      "loss": 5.8296,
      "step": 42000
    },
    {
      "epoch": 17.842149454240133,
      "grad_norm": 0.7542754411697388,
      "learning_rate": 3.440968718466196e-05,
      "loss": 5.8297,
      "step": 42500
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 1.61220121383667,
      "learning_rate": 3.340060544904137e-05,
      "loss": 5.8253,
      "step": 43000
    },
    {
      "epoch": 18.261964735516372,
      "grad_norm": 3.7016022205352783,
      "learning_rate": 3.239152371342079e-05,
      "loss": 5.8162,
      "step": 43500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 0.7713373303413391,
      "learning_rate": 3.13824419778002e-05,
      "loss": 5.8174,
      "step": 44000
    },
    {
      "epoch": 18.68178001679261,
      "grad_norm": 0.7449604272842407,
      "learning_rate": 3.0373360242179615e-05,
      "loss": 5.8168,
      "step": 44500
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 2.460657835006714,
      "learning_rate": 2.9364278506559034e-05,
      "loss": 5.8187,
      "step": 45000
    },
    {
      "epoch": 19.10159529806885,
      "grad_norm": 1.8433833122253418,
      "learning_rate": 2.8355196770938446e-05,
      "loss": 5.8128,
      "step": 45500
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 1.4397203922271729,
      "learning_rate": 2.734611503531786e-05,
      "loss": 5.8014,
      "step": 46000
    },
    {
      "epoch": 19.521410579345087,
      "grad_norm": 0.7590939402580261,
      "learning_rate": 2.6337033299697278e-05,
      "loss": 5.8043,
      "step": 46500
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 0.7301506996154785,
      "learning_rate": 2.532795156407669e-05,
      "loss": 5.8048,
      "step": 47000
    },
    {
      "epoch": 19.941225860621326,
      "grad_norm": 0.7734084725379944,
      "learning_rate": 2.4318869828456106e-05,
      "loss": 5.8007,
      "step": 47500
    },
    {
      "epoch": 20.151133501259444,
      "grad_norm": 0.8802815675735474,
      "learning_rate": 2.330978809283552e-05,
      "loss": 5.7973,
      "step": 48000
    },
    {
      "epoch": 20.361041141897566,
      "grad_norm": 2.6403074264526367,
      "learning_rate": 2.2300706357214934e-05,
      "loss": 5.795,
      "step": 48500
    },
    {
      "epoch": 20.570948782535684,
      "grad_norm": 0.7639625072479248,
      "learning_rate": 2.129162462159435e-05,
      "loss": 5.7903,
      "step": 49000
    },
    {
      "epoch": 20.7808564231738,
      "grad_norm": 0.7550709843635559,
      "learning_rate": 2.0282542885973765e-05,
      "loss": 5.7899,
      "step": 49500
    },
    {
      "epoch": 20.990764063811923,
      "grad_norm": 1.2644368410110474,
      "learning_rate": 1.927346115035318e-05,
      "loss": 5.7901,
      "step": 50000
    },
    {
      "epoch": 20.990764063811923,
      "eval_accuracy": 0.15263408753189223,
      "eval_loss": 6.244796276092529,
      "eval_runtime": 1.0049,
      "eval_samples_per_second": 476.662,
      "eval_steps_per_second": 4.976,
      "step": 50000
    },
    {
      "epoch": 21.20067170445004,
      "grad_norm": 0.7343103885650635,
      "learning_rate": 1.8264379414732593e-05,
      "loss": 5.784,
      "step": 50500
    },
    {
      "epoch": 21.410579345088163,
      "grad_norm": 0.9881123304367065,
      "learning_rate": 1.725529767911201e-05,
      "loss": 5.7864,
      "step": 51000
    },
    {
      "epoch": 21.62048698572628,
      "grad_norm": 0.8555614948272705,
      "learning_rate": 1.6246215943491424e-05,
      "loss": 5.7837,
      "step": 51500
    },
    {
      "epoch": 21.8303946263644,
      "grad_norm": 0.7525923252105713,
      "learning_rate": 1.5237134207870837e-05,
      "loss": 5.7824,
      "step": 52000
    },
    {
      "epoch": 22.04030226700252,
      "grad_norm": 0.9377904534339905,
      "learning_rate": 1.4228052472250252e-05,
      "loss": 5.7825,
      "step": 52500
    },
    {
      "epoch": 22.250209907640638,
      "grad_norm": 0.7610573172569275,
      "learning_rate": 1.3218970736629668e-05,
      "loss": 5.7722,
      "step": 53000
    },
    {
      "epoch": 22.460117548278756,
      "grad_norm": 0.7758990526199341,
      "learning_rate": 1.2209889001009082e-05,
      "loss": 5.7746,
      "step": 53500
    },
    {
      "epoch": 22.670025188916878,
      "grad_norm": 3.209929943084717,
      "learning_rate": 1.1200807265388496e-05,
      "loss": 5.7776,
      "step": 54000
    },
    {
      "epoch": 22.879932829554996,
      "grad_norm": 2.9033498764038086,
      "learning_rate": 1.0191725529767912e-05,
      "loss": 5.7748,
      "step": 54500
    },
    {
      "epoch": 23.089840470193113,
      "grad_norm": 2.4328505992889404,
      "learning_rate": 9.182643794147326e-06,
      "loss": 5.7709,
      "step": 55000
    },
    {
      "epoch": 23.299748110831235,
      "grad_norm": 0.7488394975662231,
      "learning_rate": 8.173562058526741e-06,
      "loss": 5.771,
      "step": 55500
    },
    {
      "epoch": 23.509655751469353,
      "grad_norm": 0.7825860381126404,
      "learning_rate": 7.164480322906156e-06,
      "loss": 5.7685,
      "step": 56000
    },
    {
      "epoch": 23.719563392107474,
      "grad_norm": 3.3306174278259277,
      "learning_rate": 6.155398587285571e-06,
      "loss": 5.7641,
      "step": 56500
    },
    {
      "epoch": 23.929471032745592,
      "grad_norm": 0.7582070827484131,
      "learning_rate": 5.146316851664986e-06,
      "loss": 5.7681,
      "step": 57000
    },
    {
      "epoch": 24.13937867338371,
      "grad_norm": 0.7624479532241821,
      "learning_rate": 4.1372351160444e-06,
      "loss": 5.7653,
      "step": 57500
    },
    {
      "epoch": 24.349286314021832,
      "grad_norm": 0.768104076385498,
      "learning_rate": 3.1281533804238145e-06,
      "loss": 5.7651,
      "step": 58000
    },
    {
      "epoch": 24.55919395465995,
      "grad_norm": 0.7916618585586548,
      "learning_rate": 2.1190716448032293e-06,
      "loss": 5.7616,
      "step": 58500
    },
    {
      "epoch": 24.769101595298068,
      "grad_norm": 0.7430672645568848,
      "learning_rate": 1.109989909182644e-06,
      "loss": 5.7613,
      "step": 59000
    },
    {
      "epoch": 24.97900923593619,
      "grad_norm": 0.7497892379760742,
      "learning_rate": 1.0090817356205853e-07,
      "loss": 5.7617,
      "step": 59500
    },
    {
      "epoch": 25.0,
      "step": 59550,
      "total_flos": 1.5149101316039424e+18,
      "train_loss": 6.119275358441895,
      "train_runtime": 24674.4736,
      "train_samples_per_second": 231.655,
      "train_steps_per_second": 2.413
    }
  ],
  "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": 1.5149101316039424e+18,
  "train_batch_size": 96,
  "trial_name": null,
  "trial_params": null
}