{
  "best_global_step": 50000,
  "best_metric": 6.206859111785889,
  "best_model_checkpoint": "bert_base_km_5_v1/checkpoint-50000",
  "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": 4.277164936065674,
      "learning_rate": 5e-06,
      "loss": 10.7566,
      "step": 500
    },
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 18.51190185546875,
      "learning_rate": 1e-05,
      "loss": 8.6992,
      "step": 1000
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 6.114530086517334,
      "learning_rate": 1.5e-05,
      "loss": 7.8567,
      "step": 1500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 4.5745978355407715,
      "learning_rate": 2e-05,
      "loss": 7.4784,
      "step": 2000
    },
    {
      "epoch": 1.0495382031905962,
      "grad_norm": 13.715109825134277,
      "learning_rate": 2.5e-05,
      "loss": 7.2668,
      "step": 2500
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 7.92696475982666,
      "learning_rate": 3e-05,
      "loss": 7.1121,
      "step": 3000
    },
    {
      "epoch": 1.4693534844668346,
      "grad_norm": 5.563565731048584,
      "learning_rate": 3.5e-05,
      "loss": 6.9872,
      "step": 3500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 3.825592517852783,
      "learning_rate": 4e-05,
      "loss": 6.8841,
      "step": 4000
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 2.5625343322753906,
      "learning_rate": 4.5e-05,
      "loss": 6.8047,
      "step": 4500
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 6.973842144012451,
      "learning_rate": 5e-05,
      "loss": 6.7255,
      "step": 5000
    },
    {
      "epoch": 2.3089840470193117,
      "grad_norm": 3.0779194831848145,
      "learning_rate": 5.500000000000001e-05,
      "loss": 6.657,
      "step": 5500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 3.106268882751465,
      "learning_rate": 6e-05,
      "loss": 6.6231,
      "step": 6000
    },
    {
      "epoch": 2.72879932829555,
      "grad_norm": 5.569392681121826,
      "learning_rate": 6.500000000000001e-05,
      "loss": 6.5949,
      "step": 6500
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 4.0211567878723145,
      "learning_rate": 7e-05,
      "loss": 6.5598,
      "step": 7000
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 2.7737231254577637,
      "learning_rate": 7.500000000000001e-05,
      "loss": 6.5068,
      "step": 7500
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 2.328873872756958,
      "learning_rate": 8e-05,
      "loss": 6.4852,
      "step": 8000
    },
    {
      "epoch": 3.568429890848027,
      "grad_norm": 2.538013458251953,
      "learning_rate": 8.5e-05,
      "loss": 6.4674,
      "step": 8500
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 2.1208338737487793,
      "learning_rate": 9e-05,
      "loss": 6.4529,
      "step": 9000
    },
    {
      "epoch": 3.988245172124265,
      "grad_norm": 1.217817783355713,
      "learning_rate": 9.5e-05,
      "loss": 6.4308,
      "step": 9500
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 2.0154738426208496,
      "learning_rate": 0.0001,
      "loss": 6.3769,
      "step": 10000
    },
    {
      "epoch": 4.198152812762385,
      "eval_accuracy": 0.1513397701661148,
      "eval_loss": 6.433297157287598,
      "eval_runtime": 1.0607,
      "eval_samples_per_second": 451.574,
      "eval_steps_per_second": 4.714,
      "step": 10000
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 2.9351861476898193,
      "learning_rate": 9.899091826437942e-05,
      "loss": 6.3605,
      "step": 10500
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 2.177936315536499,
      "learning_rate": 9.798183652875883e-05,
      "loss": 6.3409,
      "step": 11000
    },
    {
      "epoch": 4.827875734676742,
      "grad_norm": 1.3544222116470337,
      "learning_rate": 9.697275479313825e-05,
      "loss": 6.3354,
      "step": 11500
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 1.8430594205856323,
      "learning_rate": 9.596367305751767e-05,
      "loss": 6.3057,
      "step": 12000
    },
    {
      "epoch": 5.247691015952981,
      "grad_norm": 3.520479202270508,
      "learning_rate": 9.495459132189707e-05,
      "loss": 6.2676,
      "step": 12500
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 1.6879801750183105,
      "learning_rate": 9.394550958627649e-05,
      "loss": 6.2528,
      "step": 13000
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 1.4952800273895264,
      "learning_rate": 9.293642785065591e-05,
      "loss": 6.248,
      "step": 13500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 3.3922054767608643,
      "learning_rate": 9.192734611503532e-05,
      "loss": 6.2394,
      "step": 14000
    },
    {
      "epoch": 6.087321578505458,
      "grad_norm": 1.4895763397216797,
      "learning_rate": 9.091826437941474e-05,
      "loss": 6.212,
      "step": 14500
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 1.7259314060211182,
      "learning_rate": 8.990918264379415e-05,
      "loss": 6.1869,
      "step": 15000
    },
    {
      "epoch": 6.507136859781696,
      "grad_norm": 2.0042507648468018,
      "learning_rate": 8.890010090817356e-05,
      "loss": 6.1802,
      "step": 15500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 2.030745267868042,
      "learning_rate": 8.789101917255298e-05,
      "loss": 6.1736,
      "step": 16000
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 1.0022932291030884,
      "learning_rate": 8.68819374369324e-05,
      "loss": 6.1623,
      "step": 16500
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 2.1215691566467285,
      "learning_rate": 8.58728557013118e-05,
      "loss": 6.1337,
      "step": 17000
    },
    {
      "epoch": 7.346767422334173,
      "grad_norm": 2.186061143875122,
      "learning_rate": 8.486377396569122e-05,
      "loss": 6.1135,
      "step": 17500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 1.947540044784546,
      "learning_rate": 8.385469223007064e-05,
      "loss": 6.1177,
      "step": 18000
    },
    {
      "epoch": 7.766582703610411,
      "grad_norm": 2.2378854751586914,
      "learning_rate": 8.284561049445006e-05,
      "loss": 6.1094,
      "step": 18500
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 1.5742141008377075,
      "learning_rate": 8.183652875882947e-05,
      "loss": 6.1076,
      "step": 19000
    },
    {
      "epoch": 8.18639798488665,
      "grad_norm": 2.0269694328308105,
      "learning_rate": 8.082744702320889e-05,
      "loss": 6.0696,
      "step": 19500
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 1.577865481376648,
      "learning_rate": 7.98183652875883e-05,
      "loss": 6.0652,
      "step": 20000
    },
    {
      "epoch": 8.39630562552477,
      "eval_accuracy": 0.15236117736187885,
      "eval_loss": 6.270164966583252,
      "eval_runtime": 1.0169,
      "eval_samples_per_second": 471.051,
      "eval_steps_per_second": 4.917,
      "step": 20000
    },
    {
      "epoch": 8.606213266162888,
      "grad_norm": 1.4776746034622192,
      "learning_rate": 7.880928355196771e-05,
      "loss": 6.0613,
      "step": 20500
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 1.1313718557357788,
      "learning_rate": 7.780020181634713e-05,
      "loss": 6.0612,
      "step": 21000
    },
    {
      "epoch": 9.026028547439127,
      "grad_norm": 3.472757339477539,
      "learning_rate": 7.679112008072655e-05,
      "loss": 6.0509,
      "step": 21500
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 1.722731590270996,
      "learning_rate": 7.578203834510595e-05,
      "loss": 6.0261,
      "step": 22000
    },
    {
      "epoch": 9.445843828715365,
      "grad_norm": 1.1875659227371216,
      "learning_rate": 7.477295660948537e-05,
      "loss": 6.0209,
      "step": 22500
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 2.201089382171631,
      "learning_rate": 7.376387487386479e-05,
      "loss": 6.0197,
      "step": 23000
    },
    {
      "epoch": 9.865659109991604,
      "grad_norm": 1.8647757768630981,
      "learning_rate": 7.27547931382442e-05,
      "loss": 6.0158,
      "step": 23500
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 1.4462037086486816,
      "learning_rate": 7.174571140262362e-05,
      "loss": 5.9976,
      "step": 24000
    },
    {
      "epoch": 10.285474391267842,
      "grad_norm": 2.066240072250366,
      "learning_rate": 7.073662966700304e-05,
      "loss": 5.9912,
      "step": 24500
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 1.6373943090438843,
      "learning_rate": 6.972754793138244e-05,
      "loss": 5.9795,
      "step": 25000
    },
    {
      "epoch": 10.705289672544081,
      "grad_norm": 0.8757071495056152,
      "learning_rate": 6.871846619576186e-05,
      "loss": 5.9815,
      "step": 25500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 1.0035611391067505,
      "learning_rate": 6.770938446014128e-05,
      "loss": 5.9785,
      "step": 26000
    },
    {
      "epoch": 11.125104953820319,
      "grad_norm": 0.9736186265945435,
      "learning_rate": 6.670030272452068e-05,
      "loss": 5.9666,
      "step": 26500
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 2.153017520904541,
      "learning_rate": 6.56912209889001e-05,
      "loss": 5.9564,
      "step": 27000
    },
    {
      "epoch": 11.544920235096557,
      "grad_norm": 1.180837631225586,
      "learning_rate": 6.468213925327952e-05,
      "loss": 5.9482,
      "step": 27500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 1.3063905239105225,
      "learning_rate": 6.367305751765893e-05,
      "loss": 5.9464,
      "step": 28000
    },
    {
      "epoch": 11.964735516372796,
      "grad_norm": 1.6498868465423584,
      "learning_rate": 6.266397578203835e-05,
      "loss": 5.9425,
      "step": 28500
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 2.0582432746887207,
      "learning_rate": 6.165489404641777e-05,
      "loss": 5.9321,
      "step": 29000
    },
    {
      "epoch": 12.384550797649034,
      "grad_norm": 1.4093681573867798,
      "learning_rate": 6.064581231079718e-05,
      "loss": 5.9213,
      "step": 29500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 1.5005286931991577,
      "learning_rate": 5.963673057517659e-05,
      "loss": 5.9267,
      "step": 30000
    },
    {
      "epoch": 12.594458438287154,
      "eval_accuracy": 0.15468561994571453,
      "eval_loss": 6.208686828613281,
      "eval_runtime": 1.0058,
      "eval_samples_per_second": 476.251,
      "eval_steps_per_second": 4.971,
      "step": 30000
    },
    {
      "epoch": 12.804366078925273,
      "grad_norm": 2.524463415145874,
      "learning_rate": 5.8627648839556e-05,
      "loss": 5.9196,
      "step": 30500
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 1.940150499343872,
      "learning_rate": 5.7618567103935416e-05,
      "loss": 5.92,
      "step": 31000
    },
    {
      "epoch": 13.224181360201511,
      "grad_norm": 2.490140438079834,
      "learning_rate": 5.660948536831484e-05,
      "loss": 5.9029,
      "step": 31500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 1.9462482929229736,
      "learning_rate": 5.5600403632694254e-05,
      "loss": 5.902,
      "step": 32000
    },
    {
      "epoch": 13.64399664147775,
      "grad_norm": 0.9252886176109314,
      "learning_rate": 5.4591321897073666e-05,
      "loss": 5.9052,
      "step": 32500
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 3.112215042114258,
      "learning_rate": 5.358224016145308e-05,
      "loss": 5.9002,
      "step": 33000
    },
    {
      "epoch": 14.063811922753988,
      "grad_norm": 0.7634838819503784,
      "learning_rate": 5.257315842583249e-05,
      "loss": 5.8915,
      "step": 33500
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 1.1513233184814453,
      "learning_rate": 5.15640766902119e-05,
      "loss": 5.8829,
      "step": 34000
    },
    {
      "epoch": 14.483627204030226,
      "grad_norm": 1.0833405256271362,
      "learning_rate": 5.055499495459133e-05,
      "loss": 5.8812,
      "step": 34500
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 3.1403777599334717,
      "learning_rate": 4.954591321897074e-05,
      "loss": 5.8784,
      "step": 35000
    },
    {
      "epoch": 14.903442485306465,
      "grad_norm": 1.3890024423599243,
      "learning_rate": 4.853683148335015e-05,
      "loss": 5.8819,
      "step": 35500
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 0.7398928999900818,
      "learning_rate": 4.7527749747729566e-05,
      "loss": 5.869,
      "step": 36000
    },
    {
      "epoch": 15.323257766582703,
      "grad_norm": 1.764589786529541,
      "learning_rate": 4.6518668012108985e-05,
      "loss": 5.8665,
      "step": 36500
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 2.010664224624634,
      "learning_rate": 4.55095862764884e-05,
      "loss": 5.8609,
      "step": 37000
    },
    {
      "epoch": 15.743073047858942,
      "grad_norm": 1.577263355255127,
      "learning_rate": 4.450050454086781e-05,
      "loss": 5.8631,
      "step": 37500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 2.1810100078582764,
      "learning_rate": 4.349142280524723e-05,
      "loss": 5.8596,
      "step": 38000
    },
    {
      "epoch": 16.162888329135182,
      "grad_norm": 0.7991992235183716,
      "learning_rate": 4.248234106962664e-05,
      "loss": 5.8525,
      "step": 38500
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 1.4151620864868164,
      "learning_rate": 4.147325933400605e-05,
      "loss": 5.845,
      "step": 39000
    },
    {
      "epoch": 16.582703610411418,
      "grad_norm": 1.941565990447998,
      "learning_rate": 4.046417759838547e-05,
      "loss": 5.8489,
      "step": 39500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 0.7828423380851746,
      "learning_rate": 3.9455095862764884e-05,
      "loss": 5.8377,
      "step": 40000
    },
    {
      "epoch": 16.79261125104954,
      "eval_accuracy": 0.1531064259280027,
      "eval_loss": 6.21019983291626,
      "eval_runtime": 1.0568,
      "eval_samples_per_second": 453.248,
      "eval_steps_per_second": 4.731,
      "step": 40000
    },
    {
      "epoch": 17.002518891687657,
      "grad_norm": 0.7765722274780273,
      "learning_rate": 3.84460141271443e-05,
      "loss": 5.8441,
      "step": 40500
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 0.7961872220039368,
      "learning_rate": 3.7436932391523715e-05,
      "loss": 5.8337,
      "step": 41000
    },
    {
      "epoch": 17.422334172963897,
      "grad_norm": 0.7686080932617188,
      "learning_rate": 3.642785065590313e-05,
      "loss": 5.8288,
      "step": 41500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 0.7522885799407959,
      "learning_rate": 3.541876892028255e-05,
      "loss": 5.8296,
      "step": 42000
    },
    {
      "epoch": 17.842149454240133,
      "grad_norm": 0.794641375541687,
      "learning_rate": 3.440968718466196e-05,
      "loss": 5.8302,
      "step": 42500
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 2.6729652881622314,
      "learning_rate": 3.340060544904137e-05,
      "loss": 5.8253,
      "step": 43000
    },
    {
      "epoch": 18.261964735516372,
      "grad_norm": 0.7775242924690247,
      "learning_rate": 3.239152371342079e-05,
      "loss": 5.8167,
      "step": 43500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 0.8612495064735413,
      "learning_rate": 3.13824419778002e-05,
      "loss": 5.8159,
      "step": 44000
    },
    {
      "epoch": 18.68178001679261,
      "grad_norm": 1.8233916759490967,
      "learning_rate": 3.0373360242179615e-05,
      "loss": 5.8166,
      "step": 44500
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 0.7620252966880798,
      "learning_rate": 2.9364278506559034e-05,
      "loss": 5.8189,
      "step": 45000
    },
    {
      "epoch": 19.10159529806885,
      "grad_norm": 2.8493731021881104,
      "learning_rate": 2.8355196770938446e-05,
      "loss": 5.8146,
      "step": 45500
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 2.4592082500457764,
      "learning_rate": 2.734611503531786e-05,
      "loss": 5.8019,
      "step": 46000
    },
    {
      "epoch": 19.521410579345087,
      "grad_norm": 0.7577051520347595,
      "learning_rate": 2.6337033299697278e-05,
      "loss": 5.8053,
      "step": 46500
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 0.7634961009025574,
      "learning_rate": 2.532795156407669e-05,
      "loss": 5.8059,
      "step": 47000
    },
    {
      "epoch": 19.941225860621326,
      "grad_norm": 0.7881820797920227,
      "learning_rate": 2.4318869828456106e-05,
      "loss": 5.8018,
      "step": 47500
    },
    {
      "epoch": 20.151133501259444,
      "grad_norm": 0.7461892366409302,
      "learning_rate": 2.330978809283552e-05,
      "loss": 5.7974,
      "step": 48000
    },
    {
      "epoch": 20.361041141897566,
      "grad_norm": 0.7963252067565918,
      "learning_rate": 2.2300706357214934e-05,
      "loss": 5.7936,
      "step": 48500
    },
    {
      "epoch": 20.570948782535684,
      "grad_norm": 0.9856062531471252,
      "learning_rate": 2.129162462159435e-05,
      "loss": 5.792,
      "step": 49000
    },
    {
      "epoch": 20.7808564231738,
      "grad_norm": 2.0694522857666016,
      "learning_rate": 2.0282542885973765e-05,
      "loss": 5.7908,
      "step": 49500
    },
    {
      "epoch": 20.990764063811923,
      "grad_norm": 0.8027323484420776,
      "learning_rate": 1.927346115035318e-05,
      "loss": 5.7915,
      "step": 50000
    },
    {
      "epoch": 20.990764063811923,
      "eval_accuracy": 0.15325090419715703,
      "eval_loss": 6.206859111785889,
      "eval_runtime": 1.014,
      "eval_samples_per_second": 472.404,
      "eval_steps_per_second": 4.931,
      "step": 50000
    },
    {
      "epoch": 21.20067170445004,
      "grad_norm": 0.775394856929779,
      "learning_rate": 1.8264379414732593e-05,
      "loss": 5.7834,
      "step": 50500
    },
    {
      "epoch": 21.410579345088163,
      "grad_norm": 2.305814027786255,
      "learning_rate": 1.725529767911201e-05,
      "loss": 5.7856,
      "step": 51000
    },
    {
      "epoch": 21.62048698572628,
      "grad_norm": 1.7238717079162598,
      "learning_rate": 1.6246215943491424e-05,
      "loss": 5.7849,
      "step": 51500
    },
    {
      "epoch": 21.8303946263644,
      "grad_norm": 0.7742457389831543,
      "learning_rate": 1.5237134207870837e-05,
      "loss": 5.7828,
      "step": 52000
    },
    {
      "epoch": 22.04030226700252,
      "grad_norm": 3.5560109615325928,
      "learning_rate": 1.4228052472250252e-05,
      "loss": 5.7819,
      "step": 52500
    },
    {
      "epoch": 22.250209907640638,
      "grad_norm": 1.0969030857086182,
      "learning_rate": 1.3218970736629668e-05,
      "loss": 5.7712,
      "step": 53000
    },
    {
      "epoch": 22.460117548278756,
      "grad_norm": 0.7730518579483032,
      "learning_rate": 1.2209889001009082e-05,
      "loss": 5.776,
      "step": 53500
    },
    {
      "epoch": 22.670025188916878,
      "grad_norm": 0.7667786478996277,
      "learning_rate": 1.1200807265388496e-05,
      "loss": 5.7775,
      "step": 54000
    },
    {
      "epoch": 22.879932829554996,
      "grad_norm": 1.0061250925064087,
      "learning_rate": 1.0191725529767912e-05,
      "loss": 5.7741,
      "step": 54500
    },
    {
      "epoch": 23.089840470193113,
      "grad_norm": 2.6139068603515625,
      "learning_rate": 9.182643794147326e-06,
      "loss": 5.7702,
      "step": 55000
    },
    {
      "epoch": 23.299748110831235,
      "grad_norm": 0.7558300495147705,
      "learning_rate": 8.173562058526741e-06,
      "loss": 5.7704,
      "step": 55500
    },
    {
      "epoch": 23.509655751469353,
      "grad_norm": 0.7883274555206299,
      "learning_rate": 7.164480322906156e-06,
      "loss": 5.7693,
      "step": 56000
    },
    {
      "epoch": 23.719563392107474,
      "grad_norm": 0.7563859820365906,
      "learning_rate": 6.155398587285571e-06,
      "loss": 5.7652,
      "step": 56500
    },
    {
      "epoch": 23.929471032745592,
      "grad_norm": 0.767353892326355,
      "learning_rate": 5.146316851664986e-06,
      "loss": 5.7671,
      "step": 57000
    },
    {
      "epoch": 24.13937867338371,
      "grad_norm": 0.8509750366210938,
      "learning_rate": 4.1372351160444e-06,
      "loss": 5.7645,
      "step": 57500
    },
    {
      "epoch": 24.349286314021832,
      "grad_norm": 0.7755681872367859,
      "learning_rate": 3.1281533804238145e-06,
      "loss": 5.7655,
      "step": 58000
    },
    {
      "epoch": 24.55919395465995,
      "grad_norm": 0.7704532146453857,
      "learning_rate": 2.1190716448032293e-06,
      "loss": 5.76,
      "step": 58500
    },
    {
      "epoch": 24.769101595298068,
      "grad_norm": 0.7518042325973511,
      "learning_rate": 1.109989909182644e-06,
      "loss": 5.7621,
      "step": 59000
    },
    {
      "epoch": 24.97900923593619,
      "grad_norm": 0.7653847932815552,
      "learning_rate": 1.0090817356205853e-07,
      "loss": 5.7606,
      "step": 59500
    },
    {
      "epoch": 25.0,
      "step": 59550,
      "total_flos": 1.5149101316039424e+18,
      "train_loss": 6.1251428740250775,
      "train_runtime": 24629.2757,
      "train_samples_per_second": 232.081,
      "train_steps_per_second": 2.418
    }
  ],
  "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
}