{
  "best_metric": 6.738729953765869,
  "best_model_checkpoint": "tiny_bert_km_20_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": 2.626917600631714,
      "learning_rate": 5e-06,
      "loss": 12.7573,
      "step": 500
    },
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 1.9100428819656372,
      "learning_rate": 1e-05,
      "loss": 11.1676,
      "step": 1000
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 2.924760341644287,
      "learning_rate": 1.5e-05,
      "loss": 9.7025,
      "step": 1500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 5.516970157623291,
      "learning_rate": 2e-05,
      "loss": 8.6156,
      "step": 2000
    },
    {
      "epoch": 1.0495382031905962,
      "grad_norm": 8.952390670776367,
      "learning_rate": 2.5e-05,
      "loss": 7.9966,
      "step": 2500
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 3.484361410140991,
      "learning_rate": 3e-05,
      "loss": 7.6978,
      "step": 3000
    },
    {
      "epoch": 1.4693534844668346,
      "grad_norm": 7.40541934967041,
      "learning_rate": 3.5e-05,
      "loss": 7.5305,
      "step": 3500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 8.520888328552246,
      "learning_rate": 4e-05,
      "loss": 7.4397,
      "step": 4000
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 6.013493061065674,
      "learning_rate": 4.5e-05,
      "loss": 7.3428,
      "step": 4500
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 6.658591270446777,
      "learning_rate": 5e-05,
      "loss": 7.2303,
      "step": 5000
    },
    {
      "epoch": 2.3089840470193117,
      "grad_norm": 3.478933334350586,
      "learning_rate": 5.500000000000001e-05,
      "loss": 7.1514,
      "step": 5500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 5.886797904968262,
      "learning_rate": 6e-05,
      "loss": 7.1208,
      "step": 6000
    },
    {
      "epoch": 2.72879932829555,
      "grad_norm": 6.351716995239258,
      "learning_rate": 6.500000000000001e-05,
      "loss": 7.0765,
      "step": 6500
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 9.912898063659668,
      "learning_rate": 7e-05,
      "loss": 7.0506,
      "step": 7000
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 5.107701778411865,
      "learning_rate": 7.500000000000001e-05,
      "loss": 6.9614,
      "step": 7500
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 5.066652297973633,
      "learning_rate": 8e-05,
      "loss": 6.9374,
      "step": 8000
    },
    {
      "epoch": 3.568429890848027,
      "grad_norm": 4.600908279418945,
      "learning_rate": 8.5e-05,
      "loss": 6.9234,
      "step": 8500
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 6.573398590087891,
      "learning_rate": 9e-05,
      "loss": 6.9026,
      "step": 9000
    },
    {
      "epoch": 3.988245172124265,
      "grad_norm": 4.131033897399902,
      "learning_rate": 9.5e-05,
      "loss": 6.8857,
      "step": 9500
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 3.8546698093414307,
      "learning_rate": 0.0001,
      "loss": 6.7928,
      "step": 10000
    },
    {
      "epoch": 4.198152812762385,
      "eval_accuracy": 0.1481120788001892,
      "eval_loss": 6.862878799438477,
      "eval_runtime": 0.699,
      "eval_samples_per_second": 685.226,
      "eval_steps_per_second": 7.153,
      "step": 10000
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 3.9628775119781494,
      "learning_rate": 9.899091826437942e-05,
      "loss": 6.7997,
      "step": 10500
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 2.652864694595337,
      "learning_rate": 9.798183652875883e-05,
      "loss": 6.7921,
      "step": 11000
    },
    {
      "epoch": 4.827875734676742,
      "grad_norm": 2.7218310832977295,
      "learning_rate": 9.697275479313825e-05,
      "loss": 6.7781,
      "step": 11500
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 3.650275945663452,
      "learning_rate": 9.596367305751767e-05,
      "loss": 6.7365,
      "step": 12000
    },
    {
      "epoch": 5.247691015952981,
      "grad_norm": 2.891012191772461,
      "learning_rate": 9.495459132189707e-05,
      "loss": 6.6725,
      "step": 12500
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 3.2625622749328613,
      "learning_rate": 9.394550958627649e-05,
      "loss": 6.6693,
      "step": 13000
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 2.9605414867401123,
      "learning_rate": 9.293642785065591e-05,
      "loss": 6.6694,
      "step": 13500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 2.8077549934387207,
      "learning_rate": 9.192734611503532e-05,
      "loss": 6.6607,
      "step": 14000
    },
    {
      "epoch": 6.087321578505458,
      "grad_norm": 4.0943121910095215,
      "learning_rate": 9.091826437941474e-05,
      "loss": 6.6173,
      "step": 14500
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 2.0766336917877197,
      "learning_rate": 8.990918264379415e-05,
      "loss": 6.5676,
      "step": 15000
    },
    {
      "epoch": 6.507136859781696,
      "grad_norm": 2.148397445678711,
      "learning_rate": 8.890010090817356e-05,
      "loss": 6.5737,
      "step": 15500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 2.837857961654663,
      "learning_rate": 8.789101917255298e-05,
      "loss": 6.5691,
      "step": 16000
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 2.902109384536743,
      "learning_rate": 8.68819374369324e-05,
      "loss": 6.5689,
      "step": 16500
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 1.992867350578308,
      "learning_rate": 8.58728557013118e-05,
      "loss": 6.5103,
      "step": 17000
    },
    {
      "epoch": 7.346767422334173,
      "grad_norm": 4.094528675079346,
      "learning_rate": 8.486377396569122e-05,
      "loss": 6.4893,
      "step": 17500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 3.531911849975586,
      "learning_rate": 8.385469223007064e-05,
      "loss": 6.4962,
      "step": 18000
    },
    {
      "epoch": 7.766582703610411,
      "grad_norm": 3.1731412410736084,
      "learning_rate": 8.284561049445006e-05,
      "loss": 6.4986,
      "step": 18500
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 2.611910820007324,
      "learning_rate": 8.183652875882947e-05,
      "loss": 6.4964,
      "step": 19000
    },
    {
      "epoch": 8.18639798488665,
      "grad_norm": 2.186275005340576,
      "learning_rate": 8.082744702320889e-05,
      "loss": 6.4235,
      "step": 19500
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 2.4784603118896484,
      "learning_rate": 7.98183652875883e-05,
      "loss": 6.4284,
      "step": 20000
    },
    {
      "epoch": 8.39630562552477,
      "eval_accuracy": 0.15283818288953113,
      "eval_loss": 6.738729953765869,
      "eval_runtime": 0.6762,
      "eval_samples_per_second": 708.384,
      "eval_steps_per_second": 7.394,
      "step": 20000
    },
    {
      "epoch": 8.606213266162888,
      "grad_norm": 3.508121967315674,
      "learning_rate": 7.880928355196771e-05,
      "loss": 6.4296,
      "step": 20500
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 3.2439723014831543,
      "learning_rate": 7.780020181634713e-05,
      "loss": 6.4308,
      "step": 21000
    },
    {
      "epoch": 9.026028547439127,
      "grad_norm": 2.687647819519043,
      "learning_rate": 7.679112008072655e-05,
      "loss": 6.4231,
      "step": 21500
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 2.5599114894866943,
      "learning_rate": 7.578203834510595e-05,
      "loss": 6.3691,
      "step": 22000
    },
    {
      "epoch": 9.445843828715365,
      "grad_norm": 2.2765402793884277,
      "learning_rate": 7.477295660948537e-05,
      "loss": 6.374,
      "step": 22500
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 5.340300559997559,
      "learning_rate": 7.376387487386479e-05,
      "loss": 6.3733,
      "step": 23000
    },
    {
      "epoch": 9.865659109991604,
      "grad_norm": 3.3845908641815186,
      "learning_rate": 7.27547931382442e-05,
      "loss": 6.3774,
      "step": 23500
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 2.933563709259033,
      "learning_rate": 7.174571140262362e-05,
      "loss": 6.3517,
      "step": 24000
    },
    {
      "epoch": 10.285474391267842,
      "grad_norm": 4.057340145111084,
      "learning_rate": 7.073662966700304e-05,
      "loss": 6.3229,
      "step": 24500
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 2.4048781394958496,
      "learning_rate": 6.972754793138244e-05,
      "loss": 6.326,
      "step": 25000
    },
    {
      "epoch": 10.705289672544081,
      "grad_norm": 3.3523154258728027,
      "learning_rate": 6.871846619576186e-05,
      "loss": 6.3318,
      "step": 25500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 3.0950045585632324,
      "learning_rate": 6.770938446014128e-05,
      "loss": 6.3367,
      "step": 26000
    },
    {
      "epoch": 11.125104953820319,
      "grad_norm": 2.1324470043182373,
      "learning_rate": 6.670030272452068e-05,
      "loss": 6.2982,
      "step": 26500
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 3.2120983600616455,
      "learning_rate": 6.56912209889001e-05,
      "loss": 6.2809,
      "step": 27000
    },
    {
      "epoch": 11.544920235096557,
      "grad_norm": 3.337730884552002,
      "learning_rate": 6.468213925327952e-05,
      "loss": 6.2798,
      "step": 27500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 3.8394360542297363,
      "learning_rate": 6.367305751765893e-05,
      "loss": 6.2886,
      "step": 28000
    },
    {
      "epoch": 11.964735516372796,
      "grad_norm": 1.6199002265930176,
      "learning_rate": 6.266397578203835e-05,
      "loss": 6.2853,
      "step": 28500
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 3.1857144832611084,
      "learning_rate": 6.165489404641777e-05,
      "loss": 6.2519,
      "step": 29000
    },
    {
      "epoch": 12.384550797649034,
      "grad_norm": 1.888143539428711,
      "learning_rate": 6.064581231079718e-05,
      "loss": 6.2412,
      "step": 29500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 2.276371479034424,
      "learning_rate": 5.963673057517659e-05,
      "loss": 6.2528,
      "step": 30000
    },
    {
      "epoch": 12.594458438287154,
      "eval_accuracy": 0.15337045638973612,
      "eval_loss": 6.793837070465088,
      "eval_runtime": 0.6792,
      "eval_samples_per_second": 705.229,
      "eval_steps_per_second": 7.361,
      "step": 30000
    },
    {
      "epoch": 12.804366078925273,
      "grad_norm": 2.553093671798706,
      "learning_rate": 5.8627648839556e-05,
      "loss": 6.2482,
      "step": 30500
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 2.7331161499023438,
      "learning_rate": 5.7618567103935416e-05,
      "loss": 6.2494,
      "step": 31000
    },
    {
      "epoch": 13.224181360201511,
      "grad_norm": 1.9387927055358887,
      "learning_rate": 5.660948536831484e-05,
      "loss": 6.2118,
      "step": 31500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 2.0029964447021484,
      "learning_rate": 5.5600403632694254e-05,
      "loss": 6.21,
      "step": 32000
    },
    {
      "epoch": 13.64399664147775,
      "grad_norm": 2.053971529006958,
      "learning_rate": 5.4591321897073666e-05,
      "loss": 6.2202,
      "step": 32500
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 2.6166188716888428,
      "learning_rate": 5.358224016145308e-05,
      "loss": 6.22,
      "step": 33000
    },
    {
      "epoch": 14.063811922753988,
      "grad_norm": 3.489931106567383,
      "learning_rate": 5.257315842583249e-05,
      "loss": 6.2067,
      "step": 33500
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 1.5623815059661865,
      "learning_rate": 5.15640766902119e-05,
      "loss": 6.1869,
      "step": 34000
    },
    {
      "epoch": 14.483627204030226,
      "grad_norm": 1.6946079730987549,
      "learning_rate": 5.055499495459133e-05,
      "loss": 6.1897,
      "step": 34500
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 2.893101215362549,
      "learning_rate": 4.954591321897074e-05,
      "loss": 6.1905,
      "step": 35000
    },
    {
      "epoch": 14.903442485306465,
      "grad_norm": 3.0789566040039062,
      "learning_rate": 4.853683148335015e-05,
      "loss": 6.1918,
      "step": 35500
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 2.1385488510131836,
      "learning_rate": 4.7527749747729566e-05,
      "loss": 6.1725,
      "step": 36000
    },
    {
      "epoch": 15.323257766582703,
      "grad_norm": 1.7395439147949219,
      "learning_rate": 4.6518668012108985e-05,
      "loss": 6.1673,
      "step": 36500
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 3.8336212635040283,
      "learning_rate": 4.55095862764884e-05,
      "loss": 6.1665,
      "step": 37000
    },
    {
      "epoch": 15.743073047858942,
      "grad_norm": 2.4808847904205322,
      "learning_rate": 4.450050454086781e-05,
      "loss": 6.1688,
      "step": 37500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 3.9198453426361084,
      "learning_rate": 4.349142280524723e-05,
      "loss": 6.1669,
      "step": 38000
    },
    {
      "epoch": 16.162888329135182,
      "grad_norm": 3.0992448329925537,
      "learning_rate": 4.248234106962664e-05,
      "loss": 6.1483,
      "step": 38500
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 1.923215389251709,
      "learning_rate": 4.147325933400605e-05,
      "loss": 6.1389,
      "step": 39000
    },
    {
      "epoch": 16.582703610411418,
      "grad_norm": 2.8576881885528564,
      "learning_rate": 4.046417759838547e-05,
      "loss": 6.1462,
      "step": 39500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 2.547161817550659,
      "learning_rate": 3.9455095862764884e-05,
      "loss": 6.135,
      "step": 40000
    },
    {
      "epoch": 16.79261125104954,
      "eval_accuracy": 0.15220926320511383,
      "eval_loss": 6.990812301635742,
      "eval_runtime": 0.6803,
      "eval_samples_per_second": 704.08,
      "eval_steps_per_second": 7.349,
      "step": 40000
    },
    {
      "epoch": 17.002518891687657,
      "grad_norm": 2.547924041748047,
      "learning_rate": 3.84460141271443e-05,
      "loss": 6.1427,
      "step": 40500
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 1.9287402629852295,
      "learning_rate": 3.7436932391523715e-05,
      "loss": 6.119,
      "step": 41000
    },
    {
      "epoch": 17.422334172963897,
      "grad_norm": 3.7875235080718994,
      "learning_rate": 3.642785065590313e-05,
      "loss": 6.1214,
      "step": 41500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 2.060349941253662,
      "learning_rate": 3.541876892028255e-05,
      "loss": 6.1224,
      "step": 42000
    },
    {
      "epoch": 17.842149454240133,
      "grad_norm": 1.9006288051605225,
      "learning_rate": 3.440968718466196e-05,
      "loss": 6.1238,
      "step": 42500
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 3.339731216430664,
      "learning_rate": 3.340060544904137e-05,
      "loss": 6.1152,
      "step": 43000
    },
    {
      "epoch": 18.261964735516372,
      "grad_norm": 3.107632637023926,
      "learning_rate": 3.239152371342079e-05,
      "loss": 6.1032,
      "step": 43500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 3.526423215866089,
      "learning_rate": 3.13824419778002e-05,
      "loss": 6.1028,
      "step": 44000
    },
    {
      "epoch": 18.68178001679261,
      "grad_norm": 2.2601630687713623,
      "learning_rate": 3.0373360242179615e-05,
      "loss": 6.1045,
      "step": 44500
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 3.369821071624756,
      "learning_rate": 2.9364278506559034e-05,
      "loss": 6.1077,
      "step": 45000
    },
    {
      "epoch": 19.10159529806885,
      "grad_norm": 2.336292266845703,
      "learning_rate": 2.8355196770938446e-05,
      "loss": 6.0984,
      "step": 45500
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 1.6681281328201294,
      "learning_rate": 2.734611503531786e-05,
      "loss": 6.0841,
      "step": 46000
    },
    {
      "epoch": 19.521410579345087,
      "grad_norm": 1.2452404499053955,
      "learning_rate": 2.6337033299697278e-05,
      "loss": 6.0884,
      "step": 46500
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 2.2260518074035645,
      "learning_rate": 2.532795156407669e-05,
      "loss": 6.0906,
      "step": 47000
    },
    {
      "epoch": 19.941225860621326,
      "grad_norm": 2.4123737812042236,
      "learning_rate": 2.4318869828456106e-05,
      "loss": 6.0857,
      "step": 47500
    },
    {
      "epoch": 20.151133501259444,
      "grad_norm": 0.9671133160591125,
      "learning_rate": 2.330978809283552e-05,
      "loss": 6.0772,
      "step": 48000
    },
    {
      "epoch": 20.361041141897566,
      "grad_norm": 3.584923028945923,
      "learning_rate": 2.2300706357214934e-05,
      "loss": 6.076,
      "step": 48500
    },
    {
      "epoch": 20.570948782535684,
      "grad_norm": 2.9383702278137207,
      "learning_rate": 2.129162462159435e-05,
      "loss": 6.0682,
      "step": 49000
    },
    {
      "epoch": 20.7808564231738,
      "grad_norm": 2.90152645111084,
      "learning_rate": 2.0282542885973765e-05,
      "loss": 6.07,
      "step": 49500
    },
    {
      "epoch": 20.990764063811923,
      "grad_norm": 2.645386219024658,
      "learning_rate": 1.927346115035318e-05,
      "loss": 6.0725,
      "step": 50000
    },
    {
      "epoch": 20.990764063811923,
      "eval_accuracy": 0.1516247511705498,
      "eval_loss": 7.07234525680542,
      "eval_runtime": 0.7081,
      "eval_samples_per_second": 676.413,
      "eval_steps_per_second": 7.061,
      "step": 50000
    },
    {
      "epoch": 21.20067170445004,
      "grad_norm": 2.878293752670288,
      "learning_rate": 1.8264379414732593e-05,
      "loss": 6.0613,
      "step": 50500
    },
    {
      "epoch": 21.410579345088163,
      "grad_norm": 3.8497214317321777,
      "learning_rate": 1.725529767911201e-05,
      "loss": 6.0621,
      "step": 51000
    },
    {
      "epoch": 21.62048698572628,
      "grad_norm": 1.451300859451294,
      "learning_rate": 1.6246215943491424e-05,
      "loss": 6.0617,
      "step": 51500
    },
    {
      "epoch": 21.8303946263644,
      "grad_norm": 2.1480000019073486,
      "learning_rate": 1.5237134207870837e-05,
      "loss": 6.0614,
      "step": 52000
    },
    {
      "epoch": 22.04030226700252,
      "grad_norm": 1.1892906427383423,
      "learning_rate": 1.4228052472250252e-05,
      "loss": 6.0609,
      "step": 52500
    },
    {
      "epoch": 22.250209907640638,
      "grad_norm": 0.9119521379470825,
      "learning_rate": 1.3218970736629668e-05,
      "loss": 6.0453,
      "step": 53000
    },
    {
      "epoch": 22.460117548278756,
      "grad_norm": 1.1588727235794067,
      "learning_rate": 1.2209889001009082e-05,
      "loss": 6.0493,
      "step": 53500
    },
    {
      "epoch": 22.670025188916878,
      "grad_norm": 0.8290802240371704,
      "learning_rate": 1.1200807265388496e-05,
      "loss": 6.0531,
      "step": 54000
    },
    {
      "epoch": 22.879932829554996,
      "grad_norm": 2.3373863697052,
      "learning_rate": 1.0191725529767912e-05,
      "loss": 6.0472,
      "step": 54500
    },
    {
      "epoch": 23.089840470193113,
      "grad_norm": 1.9802433252334595,
      "learning_rate": 9.182643794147326e-06,
      "loss": 6.0441,
      "step": 55000
    },
    {
      "epoch": 23.299748110831235,
      "grad_norm": 0.8753450512886047,
      "learning_rate": 8.173562058526741e-06,
      "loss": 6.0393,
      "step": 55500
    },
    {
      "epoch": 23.509655751469353,
      "grad_norm": 2.932295560836792,
      "learning_rate": 7.164480322906156e-06,
      "loss": 6.0449,
      "step": 56000
    },
    {
      "epoch": 23.719563392107474,
      "grad_norm": 3.6671857833862305,
      "learning_rate": 6.155398587285571e-06,
      "loss": 6.0376,
      "step": 56500
    },
    {
      "epoch": 23.929471032745592,
      "grad_norm": 1.280537724494934,
      "learning_rate": 5.146316851664986e-06,
      "loss": 6.0395,
      "step": 57000
    },
    {
      "epoch": 24.13937867338371,
      "grad_norm": 1.0981942415237427,
      "learning_rate": 4.1372351160444e-06,
      "loss": 6.0356,
      "step": 57500
    },
    {
      "epoch": 24.349286314021832,
      "grad_norm": 2.460625410079956,
      "learning_rate": 3.1281533804238145e-06,
      "loss": 6.0385,
      "step": 58000
    },
    {
      "epoch": 24.55919395465995,
      "grad_norm": 2.348940372467041,
      "learning_rate": 2.1190716448032293e-06,
      "loss": 6.0341,
      "step": 58500
    },
    {
      "epoch": 24.769101595298068,
      "grad_norm": 1.964220404624939,
      "learning_rate": 1.109989909182644e-06,
      "loss": 6.0323,
      "step": 59000
    },
    {
      "epoch": 24.97900923593619,
      "grad_norm": 1.1357017755508423,
      "learning_rate": 1.0090817356205853e-07,
      "loss": 6.031,
      "step": 59500
    },
    {
      "epoch": 25.0,
      "step": 59550,
      "total_flos": 3.051148107142656e+17,
      "train_loss": 6.524913863046744,
      "train_runtime": 11619.3207,
      "train_samples_per_second": 491.937,
      "train_steps_per_second": 5.125
    }
  ],
  "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
}