{
  "best_metric": 6.54965877532959,
  "best_model_checkpoint": "tiny_bert_km_20_v2/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": 2.71126389503479,
      "learning_rate": 5e-06,
      "loss": 12.7506,
      "step": 500
    },
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 1.9715861082077026,
      "learning_rate": 1e-05,
      "loss": 11.1644,
      "step": 1000
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 5.0529069900512695,
      "learning_rate": 1.5e-05,
      "loss": 9.7797,
      "step": 1500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 4.67584228515625,
      "learning_rate": 2e-05,
      "loss": 8.7486,
      "step": 2000
    },
    {
      "epoch": 1.0495382031905962,
      "grad_norm": 4.135430335998535,
      "learning_rate": 2.5e-05,
      "loss": 8.1275,
      "step": 2500
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 5.515730857849121,
      "learning_rate": 3e-05,
      "loss": 7.8004,
      "step": 3000
    },
    {
      "epoch": 1.4693534844668346,
      "grad_norm": 3.708866834640503,
      "learning_rate": 3.5e-05,
      "loss": 7.6218,
      "step": 3500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 4.445878505706787,
      "learning_rate": 4e-05,
      "loss": 7.4752,
      "step": 4000
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 3.3583083152770996,
      "learning_rate": 4.5e-05,
      "loss": 7.3382,
      "step": 4500
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 3.607733964920044,
      "learning_rate": 5e-05,
      "loss": 7.2035,
      "step": 5000
    },
    {
      "epoch": 2.3089840470193117,
      "grad_norm": 4.7366204261779785,
      "learning_rate": 5.500000000000001e-05,
      "loss": 7.0941,
      "step": 5500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 4.448856830596924,
      "learning_rate": 6e-05,
      "loss": 7.044,
      "step": 6000
    },
    {
      "epoch": 2.72879932829555,
      "grad_norm": 3.4905285835266113,
      "learning_rate": 6.500000000000001e-05,
      "loss": 6.9928,
      "step": 6500
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 2.962218761444092,
      "learning_rate": 7e-05,
      "loss": 6.9435,
      "step": 7000
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 2.484970808029175,
      "learning_rate": 7.500000000000001e-05,
      "loss": 6.8375,
      "step": 7500
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 5.155412673950195,
      "learning_rate": 8e-05,
      "loss": 6.7932,
      "step": 8000
    },
    {
      "epoch": 3.568429890848027,
      "grad_norm": 2.8160481452941895,
      "learning_rate": 8.5e-05,
      "loss": 6.7949,
      "step": 8500
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 2.7087278366088867,
      "learning_rate": 9e-05,
      "loss": 6.7745,
      "step": 9000
    },
    {
      "epoch": 3.988245172124265,
      "grad_norm": 2.5564920902252197,
      "learning_rate": 9.5e-05,
      "loss": 6.759,
      "step": 9500
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 2.553255796432495,
      "learning_rate": 0.0001,
      "loss": 6.6514,
      "step": 10000
    },
    {
      "epoch": 4.198152812762385,
      "eval_accuracy": 0.14850162775814574,
      "eval_loss": 6.7570953369140625,
      "eval_runtime": 0.7121,
      "eval_samples_per_second": 672.648,
      "eval_steps_per_second": 7.021,
      "step": 10000
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 4.206502437591553,
      "learning_rate": 9.899091826437942e-05,
      "loss": 6.6586,
      "step": 10500
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 3.871795415878296,
      "learning_rate": 9.798183652875883e-05,
      "loss": 6.6555,
      "step": 11000
    },
    {
      "epoch": 4.827875734676742,
      "grad_norm": 2.5797126293182373,
      "learning_rate": 9.697275479313825e-05,
      "loss": 6.6546,
      "step": 11500
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 2.1714375019073486,
      "learning_rate": 9.596367305751767e-05,
      "loss": 6.6229,
      "step": 12000
    },
    {
      "epoch": 5.247691015952981,
      "grad_norm": 2.7962558269500732,
      "learning_rate": 9.495459132189707e-05,
      "loss": 6.5348,
      "step": 12500
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 3.548794984817505,
      "learning_rate": 9.394550958627649e-05,
      "loss": 6.5446,
      "step": 13000
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 2.6331114768981934,
      "learning_rate": 9.293642785065591e-05,
      "loss": 6.5454,
      "step": 13500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 4.004150390625,
      "learning_rate": 9.192734611503532e-05,
      "loss": 6.54,
      "step": 14000
    },
    {
      "epoch": 6.087321578505458,
      "grad_norm": 2.731539249420166,
      "learning_rate": 9.091826437941474e-05,
      "loss": 6.499,
      "step": 14500
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 2.7149033546447754,
      "learning_rate": 8.990918264379415e-05,
      "loss": 6.4526,
      "step": 15000
    },
    {
      "epoch": 6.507136859781696,
      "grad_norm": 2.085320472717285,
      "learning_rate": 8.890010090817356e-05,
      "loss": 6.4567,
      "step": 15500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 3.539597272872925,
      "learning_rate": 8.789101917255298e-05,
      "loss": 6.4563,
      "step": 16000
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 2.0627574920654297,
      "learning_rate": 8.68819374369324e-05,
      "loss": 6.4603,
      "step": 16500
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 2.7053449153900146,
      "learning_rate": 8.58728557013118e-05,
      "loss": 6.3975,
      "step": 17000
    },
    {
      "epoch": 7.346767422334173,
      "grad_norm": 2.3060944080352783,
      "learning_rate": 8.486377396569122e-05,
      "loss": 6.3735,
      "step": 17500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 2.0664925575256348,
      "learning_rate": 8.385469223007064e-05,
      "loss": 6.3892,
      "step": 18000
    },
    {
      "epoch": 7.766582703610411,
      "grad_norm": 3.5700786113739014,
      "learning_rate": 8.284561049445006e-05,
      "loss": 6.3867,
      "step": 18500
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 2.6246838569641113,
      "learning_rate": 8.183652875882947e-05,
      "loss": 6.3911,
      "step": 19000
    },
    {
      "epoch": 8.18639798488665,
      "grad_norm": 2.0920815467834473,
      "learning_rate": 8.082744702320889e-05,
      "loss": 6.3265,
      "step": 19500
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 2.441361665725708,
      "learning_rate": 7.98183652875883e-05,
      "loss": 6.326,
      "step": 20000
    },
    {
      "epoch": 8.39630562552477,
      "eval_accuracy": 0.15264176884873312,
      "eval_loss": 6.54965877532959,
      "eval_runtime": 0.6605,
      "eval_samples_per_second": 725.235,
      "eval_steps_per_second": 7.57,
      "step": 20000
    },
    {
      "epoch": 8.606213266162888,
      "grad_norm": 2.759524345397949,
      "learning_rate": 7.880928355196771e-05,
      "loss": 6.3265,
      "step": 20500
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 3.454348564147949,
      "learning_rate": 7.780020181634713e-05,
      "loss": 6.3344,
      "step": 21000
    },
    {
      "epoch": 9.026028547439127,
      "grad_norm": 2.499746322631836,
      "learning_rate": 7.679112008072655e-05,
      "loss": 6.3274,
      "step": 21500
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 2.2949016094207764,
      "learning_rate": 7.578203834510595e-05,
      "loss": 6.2767,
      "step": 22000
    },
    {
      "epoch": 9.445843828715365,
      "grad_norm": 1.830916166305542,
      "learning_rate": 7.477295660948537e-05,
      "loss": 6.2786,
      "step": 22500
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 3.42680025100708,
      "learning_rate": 7.376387487386479e-05,
      "loss": 6.2788,
      "step": 23000
    },
    {
      "epoch": 9.865659109991604,
      "grad_norm": 1.889884114265442,
      "learning_rate": 7.27547931382442e-05,
      "loss": 6.2782,
      "step": 23500
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 1.8403366804122925,
      "learning_rate": 7.174571140262362e-05,
      "loss": 6.2553,
      "step": 24000
    },
    {
      "epoch": 10.285474391267842,
      "grad_norm": 2.413959503173828,
      "learning_rate": 7.073662966700304e-05,
      "loss": 6.2381,
      "step": 24500
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 2.354292631149292,
      "learning_rate": 6.972754793138244e-05,
      "loss": 6.2367,
      "step": 25000
    },
    {
      "epoch": 10.705289672544081,
      "grad_norm": 2.1689646244049072,
      "learning_rate": 6.871846619576186e-05,
      "loss": 6.2348,
      "step": 25500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 2.645822286605835,
      "learning_rate": 6.770938446014128e-05,
      "loss": 6.2401,
      "step": 26000
    },
    {
      "epoch": 11.125104953820319,
      "grad_norm": 2.8303282260894775,
      "learning_rate": 6.670030272452068e-05,
      "loss": 6.212,
      "step": 26500
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 3.3768234252929688,
      "learning_rate": 6.56912209889001e-05,
      "loss": 6.2001,
      "step": 27000
    },
    {
      "epoch": 11.544920235096557,
      "grad_norm": 3.458333730697632,
      "learning_rate": 6.468213925327952e-05,
      "loss": 6.2017,
      "step": 27500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 2.546410322189331,
      "learning_rate": 6.367305751765893e-05,
      "loss": 6.2012,
      "step": 28000
    },
    {
      "epoch": 11.964735516372796,
      "grad_norm": 2.704577922821045,
      "learning_rate": 6.266397578203835e-05,
      "loss": 6.2013,
      "step": 28500
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 1.857717752456665,
      "learning_rate": 6.165489404641777e-05,
      "loss": 6.1723,
      "step": 29000
    },
    {
      "epoch": 12.384550797649034,
      "grad_norm": 2.8042995929718018,
      "learning_rate": 6.064581231079718e-05,
      "loss": 6.1654,
      "step": 29500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 2.007753849029541,
      "learning_rate": 5.963673057517659e-05,
      "loss": 6.1671,
      "step": 30000
    },
    {
      "epoch": 12.594458438287154,
      "eval_accuracy": 0.15440579791252762,
      "eval_loss": 6.653005599975586,
      "eval_runtime": 0.6695,
      "eval_samples_per_second": 715.463,
      "eval_steps_per_second": 7.468,
      "step": 30000
    },
    {
      "epoch": 12.804366078925273,
      "grad_norm": 2.3782219886779785,
      "learning_rate": 5.8627648839556e-05,
      "loss": 6.1671,
      "step": 30500
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 2.7273099422454834,
      "learning_rate": 5.7618567103935416e-05,
      "loss": 6.1705,
      "step": 31000
    },
    {
      "epoch": 13.224181360201511,
      "grad_norm": 1.9647295475006104,
      "learning_rate": 5.660948536831484e-05,
      "loss": 6.1404,
      "step": 31500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 1.891290307044983,
      "learning_rate": 5.5600403632694254e-05,
      "loss": 6.1373,
      "step": 32000
    },
    {
      "epoch": 13.64399664147775,
      "grad_norm": 1.9195438623428345,
      "learning_rate": 5.4591321897073666e-05,
      "loss": 6.1417,
      "step": 32500
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 2.3004908561706543,
      "learning_rate": 5.358224016145308e-05,
      "loss": 6.1431,
      "step": 33000
    },
    {
      "epoch": 14.063811922753988,
      "grad_norm": 2.5193397998809814,
      "learning_rate": 5.257315842583249e-05,
      "loss": 6.131,
      "step": 33500
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 2.3650710582733154,
      "learning_rate": 5.15640766902119e-05,
      "loss": 6.117,
      "step": 34000
    },
    {
      "epoch": 14.483627204030226,
      "grad_norm": 2.837047576904297,
      "learning_rate": 5.055499495459133e-05,
      "loss": 6.116,
      "step": 34500
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 1.7063215970993042,
      "learning_rate": 4.954591321897074e-05,
      "loss": 6.116,
      "step": 35000
    },
    {
      "epoch": 14.903442485306465,
      "grad_norm": 1.8279320001602173,
      "learning_rate": 4.853683148335015e-05,
      "loss": 6.1207,
      "step": 35500
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 1.15080988407135,
      "learning_rate": 4.7527749747729566e-05,
      "loss": 6.1035,
      "step": 36000
    },
    {
      "epoch": 15.323257766582703,
      "grad_norm": 2.4123611450195312,
      "learning_rate": 4.6518668012108985e-05,
      "loss": 6.1014,
      "step": 36500
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 3.021183490753174,
      "learning_rate": 4.55095862764884e-05,
      "loss": 6.0981,
      "step": 37000
    },
    {
      "epoch": 15.743073047858942,
      "grad_norm": 3.244154930114746,
      "learning_rate": 4.450050454086781e-05,
      "loss": 6.0933,
      "step": 37500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 2.689694881439209,
      "learning_rate": 4.349142280524723e-05,
      "loss": 6.0943,
      "step": 38000
    },
    {
      "epoch": 16.162888329135182,
      "grad_norm": 1.9141645431518555,
      "learning_rate": 4.248234106962664e-05,
      "loss": 6.0836,
      "step": 38500
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 2.8854215145111084,
      "learning_rate": 4.147325933400605e-05,
      "loss": 6.0775,
      "step": 39000
    },
    {
      "epoch": 16.582703610411418,
      "grad_norm": 4.890077590942383,
      "learning_rate": 4.046417759838547e-05,
      "loss": 6.0798,
      "step": 39500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 2.3635575771331787,
      "learning_rate": 3.9455095862764884e-05,
      "loss": 6.0706,
      "step": 40000
    },
    {
      "epoch": 16.79261125104954,
      "eval_accuracy": 0.1513962094874958,
      "eval_loss": 6.6304779052734375,
      "eval_runtime": 0.6596,
      "eval_samples_per_second": 726.182,
      "eval_steps_per_second": 7.58,
      "step": 40000
    },
    {
      "epoch": 17.002518891687657,
      "grad_norm": 1.3440437316894531,
      "learning_rate": 3.84460141271443e-05,
      "loss": 6.076,
      "step": 40500
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 1.7824230194091797,
      "learning_rate": 3.7436932391523715e-05,
      "loss": 6.0612,
      "step": 41000
    },
    {
      "epoch": 17.422334172963897,
      "grad_norm": 3.6087536811828613,
      "learning_rate": 3.642785065590313e-05,
      "loss": 6.0588,
      "step": 41500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 2.307131767272949,
      "learning_rate": 3.541876892028255e-05,
      "loss": 6.061,
      "step": 42000
    },
    {
      "epoch": 17.842149454240133,
      "grad_norm": 1.3711094856262207,
      "learning_rate": 3.440968718466196e-05,
      "loss": 6.0569,
      "step": 42500
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 1.2603286504745483,
      "learning_rate": 3.340060544904137e-05,
      "loss": 6.0503,
      "step": 43000
    },
    {
      "epoch": 18.261964735516372,
      "grad_norm": 1.4356471300125122,
      "learning_rate": 3.239152371342079e-05,
      "loss": 6.0412,
      "step": 43500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 3.093308925628662,
      "learning_rate": 3.13824419778002e-05,
      "loss": 6.0432,
      "step": 44000
    },
    {
      "epoch": 18.68178001679261,
      "grad_norm": 2.986450433731079,
      "learning_rate": 3.0373360242179615e-05,
      "loss": 6.0436,
      "step": 44500
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 3.6752383708953857,
      "learning_rate": 2.9364278506559034e-05,
      "loss": 6.0456,
      "step": 45000
    },
    {
      "epoch": 19.10159529806885,
      "grad_norm": 2.0228183269500732,
      "learning_rate": 2.8355196770938446e-05,
      "loss": 6.0376,
      "step": 45500
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 2.9569408893585205,
      "learning_rate": 2.734611503531786e-05,
      "loss": 6.0257,
      "step": 46000
    },
    {
      "epoch": 19.521410579345087,
      "grad_norm": 1.9467380046844482,
      "learning_rate": 2.6337033299697278e-05,
      "loss": 6.0264,
      "step": 46500
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 2.2577433586120605,
      "learning_rate": 2.532795156407669e-05,
      "loss": 6.0273,
      "step": 47000
    },
    {
      "epoch": 19.941225860621326,
      "grad_norm": 1.0765509605407715,
      "learning_rate": 2.4318869828456106e-05,
      "loss": 6.0226,
      "step": 47500
    },
    {
      "epoch": 20.151133501259444,
      "grad_norm": 0.9119879603385925,
      "learning_rate": 2.330978809283552e-05,
      "loss": 6.0192,
      "step": 48000
    },
    {
      "epoch": 20.361041141897566,
      "grad_norm": 3.756011724472046,
      "learning_rate": 2.2300706357214934e-05,
      "loss": 6.0165,
      "step": 48500
    },
    {
      "epoch": 20.570948782535684,
      "grad_norm": 2.2077417373657227,
      "learning_rate": 2.129162462159435e-05,
      "loss": 6.0134,
      "step": 49000
    },
    {
      "epoch": 20.7808564231738,
      "grad_norm": 0.7892888784408569,
      "learning_rate": 2.0282542885973765e-05,
      "loss": 6.0117,
      "step": 49500
    },
    {
      "epoch": 20.990764063811923,
      "grad_norm": 2.5961363315582275,
      "learning_rate": 1.927346115035318e-05,
      "loss": 6.0123,
      "step": 50000
    },
    {
      "epoch": 20.990764063811923,
      "eval_accuracy": 0.1510920458687302,
      "eval_loss": 6.707541465759277,
      "eval_runtime": 0.6579,
      "eval_samples_per_second": 728.118,
      "eval_steps_per_second": 7.6,
      "step": 50000
    },
    {
      "epoch": 21.20067170445004,
      "grad_norm": 3.521597385406494,
      "learning_rate": 1.8264379414732593e-05,
      "loss": 6.001,
      "step": 50500
    },
    {
      "epoch": 21.410579345088163,
      "grad_norm": 1.4102064371109009,
      "learning_rate": 1.725529767911201e-05,
      "loss": 6.0054,
      "step": 51000
    },
    {
      "epoch": 21.62048698572628,
      "grad_norm": 1.0189828872680664,
      "learning_rate": 1.6246215943491424e-05,
      "loss": 6.0023,
      "step": 51500
    },
    {
      "epoch": 21.8303946263644,
      "grad_norm": 3.020364999771118,
      "learning_rate": 1.5237134207870837e-05,
      "loss": 6.0019,
      "step": 52000
    },
    {
      "epoch": 22.04030226700252,
      "grad_norm": 3.071946859359741,
      "learning_rate": 1.4228052472250252e-05,
      "loss": 6.0005,
      "step": 52500
    },
    {
      "epoch": 22.250209907640638,
      "grad_norm": 2.5311036109924316,
      "learning_rate": 1.3218970736629668e-05,
      "loss": 5.9933,
      "step": 53000
    },
    {
      "epoch": 22.460117548278756,
      "grad_norm": 1.2899010181427002,
      "learning_rate": 1.2209889001009082e-05,
      "loss": 5.9943,
      "step": 53500
    },
    {
      "epoch": 22.670025188916878,
      "grad_norm": 2.720703363418579,
      "learning_rate": 1.1200807265388496e-05,
      "loss": 5.9969,
      "step": 54000
    },
    {
      "epoch": 22.879932829554996,
      "grad_norm": 1.5236635208129883,
      "learning_rate": 1.0191725529767912e-05,
      "loss": 5.9938,
      "step": 54500
    },
    {
      "epoch": 23.089840470193113,
      "grad_norm": 0.9731554985046387,
      "learning_rate": 9.182643794147326e-06,
      "loss": 5.9893,
      "step": 55000
    },
    {
      "epoch": 23.299748110831235,
      "grad_norm": 0.8480926752090454,
      "learning_rate": 8.173562058526741e-06,
      "loss": 5.9875,
      "step": 55500
    },
    {
      "epoch": 23.509655751469353,
      "grad_norm": 2.845705032348633,
      "learning_rate": 7.164480322906156e-06,
      "loss": 5.9885,
      "step": 56000
    },
    {
      "epoch": 23.719563392107474,
      "grad_norm": 0.7594144940376282,
      "learning_rate": 6.155398587285571e-06,
      "loss": 5.9807,
      "step": 56500
    },
    {
      "epoch": 23.929471032745592,
      "grad_norm": 0.811726987361908,
      "learning_rate": 5.146316851664986e-06,
      "loss": 5.9865,
      "step": 57000
    },
    {
      "epoch": 24.13937867338371,
      "grad_norm": 1.1903959512710571,
      "learning_rate": 4.1372351160444e-06,
      "loss": 5.9809,
      "step": 57500
    },
    {
      "epoch": 24.349286314021832,
      "grad_norm": 2.064216375350952,
      "learning_rate": 3.1281533804238145e-06,
      "loss": 5.9826,
      "step": 58000
    },
    {
      "epoch": 24.55919395465995,
      "grad_norm": 0.8064579367637634,
      "learning_rate": 2.1190716448032293e-06,
      "loss": 5.9788,
      "step": 58500
    },
    {
      "epoch": 24.769101595298068,
      "grad_norm": 0.7879181504249573,
      "learning_rate": 1.109989909182644e-06,
      "loss": 5.9791,
      "step": 59000
    },
    {
      "epoch": 24.97900923593619,
      "grad_norm": 1.4866392612457275,
      "learning_rate": 1.0090817356205853e-07,
      "loss": 5.9796,
      "step": 59500
    },
    {
      "epoch": 25.0,
      "step": 59550,
      "total_flos": 3.051148107142656e+17,
      "train_loss": 6.453962587345357,
      "train_runtime": 11281.7551,
      "train_samples_per_second": 506.657,
      "train_steps_per_second": 5.278
    }
  ],
  "logging_steps": 500,
  "max_steps": 59550,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 10000,
  "total_flos": 3.051148107142656e+17,
  "train_batch_size": 96,
  "trial_name": null,
  "trial_params": null
}