{
  "best_global_step": 40000,
  "best_metric": 6.320341110229492,
  "best_model_checkpoint": "bert_base_km_10_v1/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": 2.957629442214966,
      "learning_rate": 5e-06,
      "loss": 11.0235,
      "step": 500
    },
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 26.89862632751465,
      "learning_rate": 1e-05,
      "loss": 8.8314,
      "step": 1000
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 28.85153579711914,
      "learning_rate": 1.5e-05,
      "loss": 8.0017,
      "step": 1500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 12.106171607971191,
      "learning_rate": 2e-05,
      "loss": 7.6686,
      "step": 2000
    },
    {
      "epoch": 1.0495382031905962,
      "grad_norm": 5.503027439117432,
      "learning_rate": 2.5e-05,
      "loss": 7.4478,
      "step": 2500
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 9.906328201293945,
      "learning_rate": 3e-05,
      "loss": 7.2833,
      "step": 3000
    },
    {
      "epoch": 1.4693534844668346,
      "grad_norm": 3.4275739192962646,
      "learning_rate": 3.5e-05,
      "loss": 7.1498,
      "step": 3500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 6.326077938079834,
      "learning_rate": 4e-05,
      "loss": 7.0441,
      "step": 4000
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 4.961556434631348,
      "learning_rate": 4.5e-05,
      "loss": 6.9557,
      "step": 4500
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 3.7932581901550293,
      "learning_rate": 5e-05,
      "loss": 6.866,
      "step": 5000
    },
    {
      "epoch": 2.3089840470193117,
      "grad_norm": 2.1573853492736816,
      "learning_rate": 5.500000000000001e-05,
      "loss": 6.798,
      "step": 5500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 2.618248701095581,
      "learning_rate": 6e-05,
      "loss": 6.7604,
      "step": 6000
    },
    {
      "epoch": 2.72879932829555,
      "grad_norm": 1.6221585273742676,
      "learning_rate": 6.500000000000001e-05,
      "loss": 6.7298,
      "step": 6500
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 4.635636329650879,
      "learning_rate": 7e-05,
      "loss": 6.6964,
      "step": 7000
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 2.5305886268615723,
      "learning_rate": 7.500000000000001e-05,
      "loss": 6.6333,
      "step": 7500
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 3.351393461227417,
      "learning_rate": 8e-05,
      "loss": 6.6159,
      "step": 8000
    },
    {
      "epoch": 3.568429890848027,
      "grad_norm": 2.145704746246338,
      "learning_rate": 8.5e-05,
      "loss": 6.608,
      "step": 8500
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 3.13683819770813,
      "learning_rate": 9e-05,
      "loss": 6.586,
      "step": 9000
    },
    {
      "epoch": 3.988245172124265,
      "grad_norm": 4.483334541320801,
      "learning_rate": 9.5e-05,
      "loss": 6.5661,
      "step": 9500
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 4.050334453582764,
      "learning_rate": 0.0001,
      "loss": 6.4992,
      "step": 10000
    },
    {
      "epoch": 4.198152812762385,
      "eval_accuracy": 0.15033807284565515,
      "eval_loss": 6.548629283905029,
      "eval_runtime": 1.0419,
      "eval_samples_per_second": 459.745,
      "eval_steps_per_second": 4.799,
      "step": 10000
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 1.9287068843841553,
      "learning_rate": 9.899091826437942e-05,
      "loss": 6.4902,
      "step": 10500
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 1.9653023481369019,
      "learning_rate": 9.798183652875883e-05,
      "loss": 6.4781,
      "step": 11000
    },
    {
      "epoch": 4.827875734676742,
      "grad_norm": 2.9297804832458496,
      "learning_rate": 9.697275479313825e-05,
      "loss": 6.4646,
      "step": 11500
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 1.531516194343567,
      "learning_rate": 9.596367305751767e-05,
      "loss": 6.435,
      "step": 12000
    },
    {
      "epoch": 5.247691015952981,
      "grad_norm": 2.84403920173645,
      "learning_rate": 9.495459132189707e-05,
      "loss": 6.3782,
      "step": 12500
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 2.274853229522705,
      "learning_rate": 9.394550958627649e-05,
      "loss": 6.3701,
      "step": 13000
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 1.798396348953247,
      "learning_rate": 9.293642785065591e-05,
      "loss": 6.3692,
      "step": 13500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 2.764941692352295,
      "learning_rate": 9.192734611503532e-05,
      "loss": 6.3615,
      "step": 14000
    },
    {
      "epoch": 6.087321578505458,
      "grad_norm": 2.3369522094726562,
      "learning_rate": 9.091826437941474e-05,
      "loss": 6.3254,
      "step": 14500
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 1.6648894548416138,
      "learning_rate": 8.990918264379415e-05,
      "loss": 6.2913,
      "step": 15000
    },
    {
      "epoch": 6.507136859781696,
      "grad_norm": 1.080825686454773,
      "learning_rate": 8.890010090817356e-05,
      "loss": 6.2829,
      "step": 15500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 2.333254337310791,
      "learning_rate": 8.789101917255298e-05,
      "loss": 6.2828,
      "step": 16000
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 1.4054838418960571,
      "learning_rate": 8.68819374369324e-05,
      "loss": 6.2819,
      "step": 16500
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 1.2248133420944214,
      "learning_rate": 8.58728557013118e-05,
      "loss": 6.2322,
      "step": 17000
    },
    {
      "epoch": 7.346767422334173,
      "grad_norm": 2.101506471633911,
      "learning_rate": 8.486377396569122e-05,
      "loss": 6.2153,
      "step": 17500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 2.5985188484191895,
      "learning_rate": 8.385469223007064e-05,
      "loss": 6.2187,
      "step": 18000
    },
    {
      "epoch": 7.766582703610411,
      "grad_norm": 2.4871652126312256,
      "learning_rate": 8.284561049445006e-05,
      "loss": 6.2188,
      "step": 18500
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 1.7885987758636475,
      "learning_rate": 8.183652875882947e-05,
      "loss": 6.2176,
      "step": 19000
    },
    {
      "epoch": 8.18639798488665,
      "grad_norm": 3.0471668243408203,
      "learning_rate": 8.082744702320889e-05,
      "loss": 6.1637,
      "step": 19500
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 2.808570384979248,
      "learning_rate": 7.98183652875883e-05,
      "loss": 6.1612,
      "step": 20000
    },
    {
      "epoch": 8.39630562552477,
      "eval_accuracy": 0.15286624203821655,
      "eval_loss": 6.334768295288086,
      "eval_runtime": 1.0178,
      "eval_samples_per_second": 470.643,
      "eval_steps_per_second": 4.913,
      "step": 20000
    },
    {
      "epoch": 8.606213266162888,
      "grad_norm": 1.9498950242996216,
      "learning_rate": 7.880928355196771e-05,
      "loss": 6.1631,
      "step": 20500
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 2.398033618927002,
      "learning_rate": 7.780020181634713e-05,
      "loss": 6.1616,
      "step": 21000
    },
    {
      "epoch": 9.026028547439127,
      "grad_norm": 1.159834623336792,
      "learning_rate": 7.679112008072655e-05,
      "loss": 6.1489,
      "step": 21500
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 2.115370273590088,
      "learning_rate": 7.578203834510595e-05,
      "loss": 6.1156,
      "step": 22000
    },
    {
      "epoch": 9.445843828715365,
      "grad_norm": 2.8205251693725586,
      "learning_rate": 7.477295660948537e-05,
      "loss": 6.1134,
      "step": 22500
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 1.9374617338180542,
      "learning_rate": 7.376387487386479e-05,
      "loss": 6.111,
      "step": 23000
    },
    {
      "epoch": 9.865659109991604,
      "grad_norm": 1.631994366645813,
      "learning_rate": 7.27547931382442e-05,
      "loss": 6.1113,
      "step": 23500
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 1.778627872467041,
      "learning_rate": 7.174571140262362e-05,
      "loss": 6.0904,
      "step": 24000
    },
    {
      "epoch": 10.285474391267842,
      "grad_norm": 2.0517311096191406,
      "learning_rate": 7.073662966700304e-05,
      "loss": 6.0768,
      "step": 24500
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 1.2375179529190063,
      "learning_rate": 6.972754793138244e-05,
      "loss": 6.07,
      "step": 25000
    },
    {
      "epoch": 10.705289672544081,
      "grad_norm": 2.8565666675567627,
      "learning_rate": 6.871846619576186e-05,
      "loss": 6.069,
      "step": 25500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 2.1667635440826416,
      "learning_rate": 6.770938446014128e-05,
      "loss": 6.0699,
      "step": 26000
    },
    {
      "epoch": 11.125104953820319,
      "grad_norm": 2.9108147621154785,
      "learning_rate": 6.670030272452068e-05,
      "loss": 6.0449,
      "step": 26500
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 1.5597903728485107,
      "learning_rate": 6.56912209889001e-05,
      "loss": 6.0357,
      "step": 27000
    },
    {
      "epoch": 11.544920235096557,
      "grad_norm": 2.0552663803100586,
      "learning_rate": 6.468213925327952e-05,
      "loss": 6.0323,
      "step": 27500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 2.4605319499969482,
      "learning_rate": 6.367305751765893e-05,
      "loss": 6.0322,
      "step": 28000
    },
    {
      "epoch": 11.964735516372796,
      "grad_norm": 1.68356454372406,
      "learning_rate": 6.266397578203835e-05,
      "loss": 6.0283,
      "step": 28500
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 2.0448694229125977,
      "learning_rate": 6.165489404641777e-05,
      "loss": 6.0117,
      "step": 29000
    },
    {
      "epoch": 12.384550797649034,
      "grad_norm": 1.508269190788269,
      "learning_rate": 6.064581231079718e-05,
      "loss": 5.998,
      "step": 29500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 1.7553138732910156,
      "learning_rate": 5.963673057517659e-05,
      "loss": 6.0025,
      "step": 30000
    },
    {
      "epoch": 12.594458438287154,
      "eval_accuracy": 0.15443378011584633,
      "eval_loss": 6.3795857429504395,
      "eval_runtime": 1.0017,
      "eval_samples_per_second": 478.178,
      "eval_steps_per_second": 4.991,
      "step": 30000
    },
    {
      "epoch": 12.804366078925273,
      "grad_norm": 1.422471284866333,
      "learning_rate": 5.8627648839556e-05,
      "loss": 5.9995,
      "step": 30500
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 1.6773265600204468,
      "learning_rate": 5.7618567103935416e-05,
      "loss": 6.0012,
      "step": 31000
    },
    {
      "epoch": 13.224181360201511,
      "grad_norm": 2.812241554260254,
      "learning_rate": 5.660948536831484e-05,
      "loss": 5.9747,
      "step": 31500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 0.7657954692840576,
      "learning_rate": 5.5600403632694254e-05,
      "loss": 5.9713,
      "step": 32000
    },
    {
      "epoch": 13.64399664147775,
      "grad_norm": 1.8706390857696533,
      "learning_rate": 5.4591321897073666e-05,
      "loss": 5.9766,
      "step": 32500
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 3.601344108581543,
      "learning_rate": 5.358224016145308e-05,
      "loss": 5.973,
      "step": 33000
    },
    {
      "epoch": 14.063811922753988,
      "grad_norm": 1.8090931177139282,
      "learning_rate": 5.257315842583249e-05,
      "loss": 5.9615,
      "step": 33500
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 2.1972224712371826,
      "learning_rate": 5.15640766902119e-05,
      "loss": 5.9491,
      "step": 34000
    },
    {
      "epoch": 14.483627204030226,
      "grad_norm": 1.5615252256393433,
      "learning_rate": 5.055499495459133e-05,
      "loss": 5.9488,
      "step": 34500
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 0.906795859336853,
      "learning_rate": 4.954591321897074e-05,
      "loss": 5.9487,
      "step": 35000
    },
    {
      "epoch": 14.903442485306465,
      "grad_norm": 0.8129790425300598,
      "learning_rate": 4.853683148335015e-05,
      "loss": 5.947,
      "step": 35500
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 0.781025767326355,
      "learning_rate": 4.7527749747729566e-05,
      "loss": 5.9334,
      "step": 36000
    },
    {
      "epoch": 15.323257766582703,
      "grad_norm": 0.9468545317649841,
      "learning_rate": 4.6518668012108985e-05,
      "loss": 5.9324,
      "step": 36500
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 1.2426480054855347,
      "learning_rate": 4.55095862764884e-05,
      "loss": 5.9263,
      "step": 37000
    },
    {
      "epoch": 15.743073047858942,
      "grad_norm": 2.5635149478912354,
      "learning_rate": 4.450050454086781e-05,
      "loss": 5.9272,
      "step": 37500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 1.6879910230636597,
      "learning_rate": 4.349142280524723e-05,
      "loss": 5.9236,
      "step": 38000
    },
    {
      "epoch": 16.162888329135182,
      "grad_norm": 3.6608622074127197,
      "learning_rate": 4.248234106962664e-05,
      "loss": 5.9149,
      "step": 38500
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 2.552720069885254,
      "learning_rate": 4.147325933400605e-05,
      "loss": 5.9086,
      "step": 39000
    },
    {
      "epoch": 16.582703610411418,
      "grad_norm": 0.8735772371292114,
      "learning_rate": 4.046417759838547e-05,
      "loss": 5.9091,
      "step": 39500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 1.7069435119628906,
      "learning_rate": 3.9455095862764884e-05,
      "loss": 5.8992,
      "step": 40000
    },
    {
      "epoch": 16.79261125104954,
      "eval_accuracy": 0.1534708982841763,
      "eval_loss": 6.320341110229492,
      "eval_runtime": 0.9983,
      "eval_samples_per_second": 479.824,
      "eval_steps_per_second": 5.009,
      "step": 40000
    },
    {
      "epoch": 17.002518891687657,
      "grad_norm": 2.1813061237335205,
      "learning_rate": 3.84460141271443e-05,
      "loss": 5.9098,
      "step": 40500
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 2.3921892642974854,
      "learning_rate": 3.7436932391523715e-05,
      "loss": 5.8914,
      "step": 41000
    },
    {
      "epoch": 17.422334172963897,
      "grad_norm": 0.8605856895446777,
      "learning_rate": 3.642785065590313e-05,
      "loss": 5.8898,
      "step": 41500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 2.711857318878174,
      "learning_rate": 3.541876892028255e-05,
      "loss": 5.8906,
      "step": 42000
    },
    {
      "epoch": 17.842149454240133,
      "grad_norm": 1.428740382194519,
      "learning_rate": 3.440968718466196e-05,
      "loss": 5.8879,
      "step": 42500
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 2.1238832473754883,
      "learning_rate": 3.340060544904137e-05,
      "loss": 5.8815,
      "step": 43000
    },
    {
      "epoch": 18.261964735516372,
      "grad_norm": 2.44793438911438,
      "learning_rate": 3.239152371342079e-05,
      "loss": 5.8733,
      "step": 43500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 2.8119001388549805,
      "learning_rate": 3.13824419778002e-05,
      "loss": 5.8742,
      "step": 44000
    },
    {
      "epoch": 18.68178001679261,
      "grad_norm": 2.058626174926758,
      "learning_rate": 3.0373360242179615e-05,
      "loss": 5.8733,
      "step": 44500
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 1.3226954936981201,
      "learning_rate": 2.9364278506559034e-05,
      "loss": 5.8755,
      "step": 45000
    },
    {
      "epoch": 19.10159529806885,
      "grad_norm": 2.563274621963501,
      "learning_rate": 2.8355196770938446e-05,
      "loss": 5.87,
      "step": 45500
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 4.54210901260376,
      "learning_rate": 2.734611503531786e-05,
      "loss": 5.8578,
      "step": 46000
    },
    {
      "epoch": 19.521410579345087,
      "grad_norm": 4.294485092163086,
      "learning_rate": 2.6337033299697278e-05,
      "loss": 5.8592,
      "step": 46500
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 0.7946419715881348,
      "learning_rate": 2.532795156407669e-05,
      "loss": 5.8603,
      "step": 47000
    },
    {
      "epoch": 19.941225860621326,
      "grad_norm": 2.0153515338897705,
      "learning_rate": 2.4318869828456106e-05,
      "loss": 5.8576,
      "step": 47500
    },
    {
      "epoch": 20.151133501259444,
      "grad_norm": 0.7703822255134583,
      "learning_rate": 2.330978809283552e-05,
      "loss": 5.8507,
      "step": 48000
    },
    {
      "epoch": 20.361041141897566,
      "grad_norm": 1.1402866840362549,
      "learning_rate": 2.2300706357214934e-05,
      "loss": 5.8481,
      "step": 48500
    },
    {
      "epoch": 20.570948782535684,
      "grad_norm": 0.8359605073928833,
      "learning_rate": 2.129162462159435e-05,
      "loss": 5.8428,
      "step": 49000
    },
    {
      "epoch": 20.7808564231738,
      "grad_norm": 0.7477090358734131,
      "learning_rate": 2.0282542885973765e-05,
      "loss": 5.8435,
      "step": 49500
    },
    {
      "epoch": 20.990764063811923,
      "grad_norm": 3.758740186691284,
      "learning_rate": 1.927346115035318e-05,
      "loss": 5.8428,
      "step": 50000
    },
    {
      "epoch": 20.990764063811923,
      "eval_accuracy": 0.1521013822300726,
      "eval_loss": 6.431597709655762,
      "eval_runtime": 1.036,
      "eval_samples_per_second": 462.334,
      "eval_steps_per_second": 4.826,
      "step": 50000
    },
    {
      "epoch": 21.20067170445004,
      "grad_norm": 2.1753249168395996,
      "learning_rate": 1.8264379414732593e-05,
      "loss": 5.8366,
      "step": 50500
    },
    {
      "epoch": 21.410579345088163,
      "grad_norm": 0.9047945737838745,
      "learning_rate": 1.725529767911201e-05,
      "loss": 5.8361,
      "step": 51000
    },
    {
      "epoch": 21.62048698572628,
      "grad_norm": 1.6985797882080078,
      "learning_rate": 1.6246215943491424e-05,
      "loss": 5.8355,
      "step": 51500
    },
    {
      "epoch": 21.8303946263644,
      "grad_norm": 1.0182300806045532,
      "learning_rate": 1.5237134207870837e-05,
      "loss": 5.8348,
      "step": 52000
    },
    {
      "epoch": 22.04030226700252,
      "grad_norm": 0.7883506417274475,
      "learning_rate": 1.4228052472250252e-05,
      "loss": 5.8321,
      "step": 52500
    },
    {
      "epoch": 22.250209907640638,
      "grad_norm": 2.4436087608337402,
      "learning_rate": 1.3218970736629668e-05,
      "loss": 5.8215,
      "step": 53000
    },
    {
      "epoch": 22.460117548278756,
      "grad_norm": 1.6637613773345947,
      "learning_rate": 1.2209889001009082e-05,
      "loss": 5.8264,
      "step": 53500
    },
    {
      "epoch": 22.670025188916878,
      "grad_norm": 2.597198247909546,
      "learning_rate": 1.1200807265388496e-05,
      "loss": 5.8267,
      "step": 54000
    },
    {
      "epoch": 22.879932829554996,
      "grad_norm": 0.8695548176765442,
      "learning_rate": 1.0191725529767912e-05,
      "loss": 5.8211,
      "step": 54500
    },
    {
      "epoch": 23.089840470193113,
      "grad_norm": 0.7805715203285217,
      "learning_rate": 9.182643794147326e-06,
      "loss": 5.8193,
      "step": 55000
    },
    {
      "epoch": 23.299748110831235,
      "grad_norm": 0.8124175071716309,
      "learning_rate": 8.173562058526741e-06,
      "loss": 5.8195,
      "step": 55500
    },
    {
      "epoch": 23.509655751469353,
      "grad_norm": 1.671306848526001,
      "learning_rate": 7.164480322906156e-06,
      "loss": 5.8179,
      "step": 56000
    },
    {
      "epoch": 23.719563392107474,
      "grad_norm": 0.9749959111213684,
      "learning_rate": 6.155398587285571e-06,
      "loss": 5.8103,
      "step": 56500
    },
    {
      "epoch": 23.929471032745592,
      "grad_norm": 0.7807030081748962,
      "learning_rate": 5.146316851664986e-06,
      "loss": 5.8156,
      "step": 57000
    },
    {
      "epoch": 24.13937867338371,
      "grad_norm": 0.8342841863632202,
      "learning_rate": 4.1372351160444e-06,
      "loss": 5.8113,
      "step": 57500
    },
    {
      "epoch": 24.349286314021832,
      "grad_norm": 0.780062735080719,
      "learning_rate": 3.1281533804238145e-06,
      "loss": 5.8132,
      "step": 58000
    },
    {
      "epoch": 24.55919395465995,
      "grad_norm": 0.86262047290802,
      "learning_rate": 2.1190716448032293e-06,
      "loss": 5.808,
      "step": 58500
    },
    {
      "epoch": 24.769101595298068,
      "grad_norm": 0.8456590175628662,
      "learning_rate": 1.109989909182644e-06,
      "loss": 5.8078,
      "step": 59000
    },
    {
      "epoch": 24.97900923593619,
      "grad_norm": 0.7847692370414734,
      "learning_rate": 1.0090817356205853e-07,
      "loss": 5.8083,
      "step": 59500
    },
    {
      "epoch": 25.0,
      "step": 59550,
      "total_flos": 1.5149776477860864e+18,
      "train_loss": 6.213085491651252,
      "train_runtime": 24622.688,
      "train_samples_per_second": 232.143,
      "train_steps_per_second": 2.419
    }
  ],
  "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": 1.5149776477860864e+18,
  "train_batch_size": 96,
  "trial_name": null,
  "trial_params": null
}