{
  "best_metric": 6.90674352645874,
  "best_model_checkpoint": "tiny_bert_km_50_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": 2.527822732925415,
      "learning_rate": 5e-06,
      "loss": 13.6879,
      "step": 500
    },
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 1.832667350769043,
      "learning_rate": 1e-05,
      "loss": 12.0904,
      "step": 1000
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 2.442608594894409,
      "learning_rate": 1.5e-05,
      "loss": 10.9219,
      "step": 1500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 2.884775400161743,
      "learning_rate": 2e-05,
      "loss": 9.9398,
      "step": 2000
    },
    {
      "epoch": 1.0495382031905962,
      "grad_norm": 4.984293460845947,
      "learning_rate": 2.5e-05,
      "loss": 8.9743,
      "step": 2500
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 4.8589653968811035,
      "learning_rate": 3e-05,
      "loss": 8.376,
      "step": 3000
    },
    {
      "epoch": 1.4693534844668346,
      "grad_norm": 4.101855278015137,
      "learning_rate": 3.5e-05,
      "loss": 8.0367,
      "step": 3500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 5.231034278869629,
      "learning_rate": 4e-05,
      "loss": 7.8284,
      "step": 4000
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 8.200454711914062,
      "learning_rate": 4.5e-05,
      "loss": 7.6835,
      "step": 4500
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 4.4725189208984375,
      "learning_rate": 5e-05,
      "loss": 7.5338,
      "step": 5000
    },
    {
      "epoch": 2.3089840470193117,
      "grad_norm": 7.595385551452637,
      "learning_rate": 5.500000000000001e-05,
      "loss": 7.4372,
      "step": 5500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 9.865002632141113,
      "learning_rate": 6e-05,
      "loss": 7.4078,
      "step": 6000
    },
    {
      "epoch": 2.72879932829555,
      "grad_norm": 6.366688251495361,
      "learning_rate": 6.500000000000001e-05,
      "loss": 7.3709,
      "step": 6500
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 4.332571029663086,
      "learning_rate": 7e-05,
      "loss": 7.3183,
      "step": 7000
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 7.531284809112549,
      "learning_rate": 7.500000000000001e-05,
      "loss": 7.225,
      "step": 7500
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 4.725491046905518,
      "learning_rate": 8e-05,
      "loss": 7.1982,
      "step": 8000
    },
    {
      "epoch": 3.568429890848027,
      "grad_norm": 5.494848728179932,
      "learning_rate": 8.5e-05,
      "loss": 7.205,
      "step": 8500
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 6.40097713470459,
      "learning_rate": 9e-05,
      "loss": 7.1751,
      "step": 9000
    },
    {
      "epoch": 3.988245172124265,
      "grad_norm": 3.8896799087524414,
      "learning_rate": 9.5e-05,
      "loss": 7.1688,
      "step": 9500
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 4.6238179206848145,
      "learning_rate": 0.0001,
      "loss": 7.0546,
      "step": 10000
    },
    {
      "epoch": 4.198152812762385,
      "eval_accuracy": 0.146804307298478,
      "eval_loss": 7.054701328277588,
      "eval_runtime": 0.6922,
      "eval_samples_per_second": 692.024,
      "eval_steps_per_second": 7.224,
      "step": 10000
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 3.847954511642456,
      "learning_rate": 9.899091826437942e-05,
      "loss": 7.0649,
      "step": 10500
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 4.213462829589844,
      "learning_rate": 9.798183652875883e-05,
      "loss": 7.0638,
      "step": 11000
    },
    {
      "epoch": 4.827875734676742,
      "grad_norm": 3.9441540241241455,
      "learning_rate": 9.697275479313825e-05,
      "loss": 7.0479,
      "step": 11500
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 3.684396266937256,
      "learning_rate": 9.596367305751767e-05,
      "loss": 7.013,
      "step": 12000
    },
    {
      "epoch": 5.247691015952981,
      "grad_norm": 3.9965009689331055,
      "learning_rate": 9.495459132189707e-05,
      "loss": 6.9222,
      "step": 12500
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 4.845287322998047,
      "learning_rate": 9.394550958627649e-05,
      "loss": 6.928,
      "step": 13000
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 4.661120891571045,
      "learning_rate": 9.293642785065591e-05,
      "loss": 6.9347,
      "step": 13500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 3.7022831439971924,
      "learning_rate": 9.192734611503532e-05,
      "loss": 6.9283,
      "step": 14000
    },
    {
      "epoch": 6.087321578505458,
      "grad_norm": 4.2522430419921875,
      "learning_rate": 9.091826437941474e-05,
      "loss": 6.8775,
      "step": 14500
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 5.626385688781738,
      "learning_rate": 8.990918264379415e-05,
      "loss": 6.8201,
      "step": 15000
    },
    {
      "epoch": 6.507136859781696,
      "grad_norm": 4.78915548324585,
      "learning_rate": 8.890010090817356e-05,
      "loss": 6.8253,
      "step": 15500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 6.639094829559326,
      "learning_rate": 8.789101917255298e-05,
      "loss": 6.828,
      "step": 16000
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 5.143166542053223,
      "learning_rate": 8.68819374369324e-05,
      "loss": 6.8247,
      "step": 16500
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 3.9872820377349854,
      "learning_rate": 8.58728557013118e-05,
      "loss": 6.7473,
      "step": 17000
    },
    {
      "epoch": 7.346767422334173,
      "grad_norm": 4.957212924957275,
      "learning_rate": 8.486377396569122e-05,
      "loss": 6.7193,
      "step": 17500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 3.8618111610412598,
      "learning_rate": 8.385469223007064e-05,
      "loss": 6.7323,
      "step": 18000
    },
    {
      "epoch": 7.766582703610411,
      "grad_norm": 4.4679059982299805,
      "learning_rate": 8.284561049445006e-05,
      "loss": 6.7491,
      "step": 18500
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 3.817917823791504,
      "learning_rate": 8.183652875882947e-05,
      "loss": 6.7514,
      "step": 19000
    },
    {
      "epoch": 8.18639798488665,
      "grad_norm": 3.414637804031372,
      "learning_rate": 8.082744702320889e-05,
      "loss": 6.6528,
      "step": 19500
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 3.67791485786438,
      "learning_rate": 7.98183652875883e-05,
      "loss": 6.6575,
      "step": 20000
    },
    {
      "epoch": 8.39630562552477,
      "eval_accuracy": 0.15188417183422656,
      "eval_loss": 6.90674352645874,
      "eval_runtime": 0.6721,
      "eval_samples_per_second": 712.664,
      "eval_steps_per_second": 7.439,
      "step": 20000
    },
    {
      "epoch": 8.606213266162888,
      "grad_norm": 6.641595840454102,
      "learning_rate": 7.880928355196771e-05,
      "loss": 6.6642,
      "step": 20500
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 3.579845666885376,
      "learning_rate": 7.780020181634713e-05,
      "loss": 6.6703,
      "step": 21000
    },
    {
      "epoch": 9.026028547439127,
      "grad_norm": 2.7270305156707764,
      "learning_rate": 7.679112008072655e-05,
      "loss": 6.6514,
      "step": 21500
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 4.12720251083374,
      "learning_rate": 7.578203834510595e-05,
      "loss": 6.5674,
      "step": 22000
    },
    {
      "epoch": 9.445843828715365,
      "grad_norm": 4.198493480682373,
      "learning_rate": 7.477295660948537e-05,
      "loss": 6.5816,
      "step": 22500
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 3.270420789718628,
      "learning_rate": 7.376387487386479e-05,
      "loss": 6.5928,
      "step": 23000
    },
    {
      "epoch": 9.865659109991604,
      "grad_norm": 3.2376928329467773,
      "learning_rate": 7.27547931382442e-05,
      "loss": 6.586,
      "step": 23500
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 3.6241872310638428,
      "learning_rate": 7.174571140262362e-05,
      "loss": 6.5576,
      "step": 24000
    },
    {
      "epoch": 10.285474391267842,
      "grad_norm": 3.476975202560425,
      "learning_rate": 7.073662966700304e-05,
      "loss": 6.5142,
      "step": 24500
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 4.205807685852051,
      "learning_rate": 6.972754793138244e-05,
      "loss": 6.5225,
      "step": 25000
    },
    {
      "epoch": 10.705289672544081,
      "grad_norm": 5.018836975097656,
      "learning_rate": 6.871846619576186e-05,
      "loss": 6.5327,
      "step": 25500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 3.1629929542541504,
      "learning_rate": 6.770938446014128e-05,
      "loss": 6.5332,
      "step": 26000
    },
    {
      "epoch": 11.125104953820319,
      "grad_norm": 3.34938645362854,
      "learning_rate": 6.670030272452068e-05,
      "loss": 6.4867,
      "step": 26500
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 4.3821187019348145,
      "learning_rate": 6.56912209889001e-05,
      "loss": 6.46,
      "step": 27000
    },
    {
      "epoch": 11.544920235096557,
      "grad_norm": 3.1705963611602783,
      "learning_rate": 6.468213925327952e-05,
      "loss": 6.4691,
      "step": 27500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 3.833258867263794,
      "learning_rate": 6.367305751765893e-05,
      "loss": 6.48,
      "step": 28000
    },
    {
      "epoch": 11.964735516372796,
      "grad_norm": 3.203676223754883,
      "learning_rate": 6.266397578203835e-05,
      "loss": 6.48,
      "step": 28500
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 3.013791799545288,
      "learning_rate": 6.165489404641777e-05,
      "loss": 6.4161,
      "step": 29000
    },
    {
      "epoch": 12.384550797649034,
      "grad_norm": 3.252168893814087,
      "learning_rate": 6.064581231079718e-05,
      "loss": 6.4132,
      "step": 29500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 3.701188087463379,
      "learning_rate": 5.963673057517659e-05,
      "loss": 6.4303,
      "step": 30000
    },
    {
      "epoch": 12.594458438287154,
      "eval_accuracy": 0.15393010045610991,
      "eval_loss": 6.945522785186768,
      "eval_runtime": 0.6612,
      "eval_samples_per_second": 724.485,
      "eval_steps_per_second": 7.562,
      "step": 30000
    },
    {
      "epoch": 12.804366078925273,
      "grad_norm": 5.079094886779785,
      "learning_rate": 5.8627648839556e-05,
      "loss": 6.4281,
      "step": 30500
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 4.823525428771973,
      "learning_rate": 5.7618567103935416e-05,
      "loss": 6.4262,
      "step": 31000
    },
    {
      "epoch": 13.224181360201511,
      "grad_norm": 3.2761666774749756,
      "learning_rate": 5.660948536831484e-05,
      "loss": 6.3686,
      "step": 31500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 3.2379093170166016,
      "learning_rate": 5.5600403632694254e-05,
      "loss": 6.3733,
      "step": 32000
    },
    {
      "epoch": 13.64399664147775,
      "grad_norm": 5.344300270080566,
      "learning_rate": 5.4591321897073666e-05,
      "loss": 6.3839,
      "step": 32500
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 4.2472333908081055,
      "learning_rate": 5.358224016145308e-05,
      "loss": 6.3871,
      "step": 33000
    },
    {
      "epoch": 14.063811922753988,
      "grad_norm": 2.8953723907470703,
      "learning_rate": 5.257315842583249e-05,
      "loss": 6.3709,
      "step": 33500
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 3.744457244873047,
      "learning_rate": 5.15640766902119e-05,
      "loss": 6.3287,
      "step": 34000
    },
    {
      "epoch": 14.483627204030226,
      "grad_norm": 4.1032938957214355,
      "learning_rate": 5.055499495459133e-05,
      "loss": 6.3368,
      "step": 34500
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 4.128444671630859,
      "learning_rate": 4.954591321897074e-05,
      "loss": 6.3444,
      "step": 35000
    },
    {
      "epoch": 14.903442485306465,
      "grad_norm": 4.473135948181152,
      "learning_rate": 4.853683148335015e-05,
      "loss": 6.3499,
      "step": 35500
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 4.16132116317749,
      "learning_rate": 4.7527749747729566e-05,
      "loss": 6.3176,
      "step": 36000
    },
    {
      "epoch": 15.323257766582703,
      "grad_norm": 3.1756906509399414,
      "learning_rate": 4.6518668012108985e-05,
      "loss": 6.3037,
      "step": 36500
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 4.053339958190918,
      "learning_rate": 4.55095862764884e-05,
      "loss": 6.3014,
      "step": 37000
    },
    {
      "epoch": 15.743073047858942,
      "grad_norm": 4.244011402130127,
      "learning_rate": 4.450050454086781e-05,
      "loss": 6.3117,
      "step": 37500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 4.481657028198242,
      "learning_rate": 4.349142280524723e-05,
      "loss": 6.3123,
      "step": 38000
    },
    {
      "epoch": 16.162888329135182,
      "grad_norm": 4.420573711395264,
      "learning_rate": 4.248234106962664e-05,
      "loss": 6.2776,
      "step": 38500
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 4.302750110626221,
      "learning_rate": 4.147325933400605e-05,
      "loss": 6.267,
      "step": 39000
    },
    {
      "epoch": 16.582703610411418,
      "grad_norm": 2.761948347091675,
      "learning_rate": 4.046417759838547e-05,
      "loss": 6.281,
      "step": 39500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 3.3701863288879395,
      "learning_rate": 3.9455095862764884e-05,
      "loss": 6.2678,
      "step": 40000
    },
    {
      "epoch": 16.79261125104954,
      "eval_accuracy": 0.15190086351912077,
      "eval_loss": 7.036715030670166,
      "eval_runtime": 0.6676,
      "eval_samples_per_second": 717.518,
      "eval_steps_per_second": 7.49,
      "step": 40000
    },
    {
      "epoch": 17.002518891687657,
      "grad_norm": 4.5265092849731445,
      "learning_rate": 3.84460141271443e-05,
      "loss": 6.2764,
      "step": 40500
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 3.0750575065612793,
      "learning_rate": 3.7436932391523715e-05,
      "loss": 6.2409,
      "step": 41000
    },
    {
      "epoch": 17.422334172963897,
      "grad_norm": 4.004626274108887,
      "learning_rate": 3.642785065590313e-05,
      "loss": 6.238,
      "step": 41500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 2.2061779499053955,
      "learning_rate": 3.541876892028255e-05,
      "loss": 6.2384,
      "step": 42000
    },
    {
      "epoch": 17.842149454240133,
      "grad_norm": 2.8878047466278076,
      "learning_rate": 3.440968718466196e-05,
      "loss": 6.2478,
      "step": 42500
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 2.6562340259552,
      "learning_rate": 3.340060544904137e-05,
      "loss": 6.2383,
      "step": 43000
    },
    {
      "epoch": 18.261964735516372,
      "grad_norm": 3.6874148845672607,
      "learning_rate": 3.239152371342079e-05,
      "loss": 6.2119,
      "step": 43500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 4.094950199127197,
      "learning_rate": 3.13824419778002e-05,
      "loss": 6.2215,
      "step": 44000
    },
    {
      "epoch": 18.68178001679261,
      "grad_norm": 3.1106882095336914,
      "learning_rate": 3.0373360242179615e-05,
      "loss": 6.217,
      "step": 44500
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 4.529526710510254,
      "learning_rate": 2.9364278506559034e-05,
      "loss": 6.2237,
      "step": 45000
    },
    {
      "epoch": 19.10159529806885,
      "grad_norm": 4.674984931945801,
      "learning_rate": 2.8355196770938446e-05,
      "loss": 6.213,
      "step": 45500
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 5.12192440032959,
      "learning_rate": 2.734611503531786e-05,
      "loss": 6.1936,
      "step": 46000
    },
    {
      "epoch": 19.521410579345087,
      "grad_norm": 3.6852598190307617,
      "learning_rate": 2.6337033299697278e-05,
      "loss": 6.1967,
      "step": 46500
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 3.273944616317749,
      "learning_rate": 2.532795156407669e-05,
      "loss": 6.1976,
      "step": 47000
    },
    {
      "epoch": 19.941225860621326,
      "grad_norm": 2.061715841293335,
      "learning_rate": 2.4318869828456106e-05,
      "loss": 6.196,
      "step": 47500
    },
    {
      "epoch": 20.151133501259444,
      "grad_norm": 3.486471652984619,
      "learning_rate": 2.330978809283552e-05,
      "loss": 6.18,
      "step": 48000
    },
    {
      "epoch": 20.361041141897566,
      "grad_norm": 4.1520586013793945,
      "learning_rate": 2.2300706357214934e-05,
      "loss": 6.1752,
      "step": 48500
    },
    {
      "epoch": 20.570948782535684,
      "grad_norm": 4.145242214202881,
      "learning_rate": 2.129162462159435e-05,
      "loss": 6.1739,
      "step": 49000
    },
    {
      "epoch": 20.7808564231738,
      "grad_norm": 3.667370557785034,
      "learning_rate": 2.0282542885973765e-05,
      "loss": 6.1737,
      "step": 49500
    },
    {
      "epoch": 20.990764063811923,
      "grad_norm": 3.066197156906128,
      "learning_rate": 1.927346115035318e-05,
      "loss": 6.1768,
      "step": 50000
    },
    {
      "epoch": 20.990764063811923,
      "eval_accuracy": 0.15036308071887178,
      "eval_loss": 7.0473480224609375,
      "eval_runtime": 0.6644,
      "eval_samples_per_second": 720.954,
      "eval_steps_per_second": 7.526,
      "step": 50000
    },
    {
      "epoch": 21.20067170445004,
      "grad_norm": 3.2085506916046143,
      "learning_rate": 1.8264379414732593e-05,
      "loss": 6.157,
      "step": 50500
    },
    {
      "epoch": 21.410579345088163,
      "grad_norm": 5.945531368255615,
      "learning_rate": 1.725529767911201e-05,
      "loss": 6.1593,
      "step": 51000
    },
    {
      "epoch": 21.62048698572628,
      "grad_norm": 3.410444498062134,
      "learning_rate": 1.6246215943491424e-05,
      "loss": 6.1552,
      "step": 51500
    },
    {
      "epoch": 21.8303946263644,
      "grad_norm": 3.0463058948516846,
      "learning_rate": 1.5237134207870837e-05,
      "loss": 6.1585,
      "step": 52000
    },
    {
      "epoch": 22.04030226700252,
      "grad_norm": 3.5028927326202393,
      "learning_rate": 1.4228052472250252e-05,
      "loss": 6.1511,
      "step": 52500
    },
    {
      "epoch": 22.250209907640638,
      "grad_norm": 4.44357967376709,
      "learning_rate": 1.3218970736629668e-05,
      "loss": 6.1387,
      "step": 53000
    },
    {
      "epoch": 22.460117548278756,
      "grad_norm": 2.49257755279541,
      "learning_rate": 1.2209889001009082e-05,
      "loss": 6.1415,
      "step": 53500
    },
    {
      "epoch": 22.670025188916878,
      "grad_norm": 2.910200595855713,
      "learning_rate": 1.1200807265388496e-05,
      "loss": 6.1419,
      "step": 54000
    },
    {
      "epoch": 22.879932829554996,
      "grad_norm": 1.404018521308899,
      "learning_rate": 1.0191725529767912e-05,
      "loss": 6.1403,
      "step": 54500
    },
    {
      "epoch": 23.089840470193113,
      "grad_norm": 3.4004967212677,
      "learning_rate": 9.182643794147326e-06,
      "loss": 6.1368,
      "step": 55000
    },
    {
      "epoch": 23.299748110831235,
      "grad_norm": 2.04544997215271,
      "learning_rate": 8.173562058526741e-06,
      "loss": 6.1299,
      "step": 55500
    },
    {
      "epoch": 23.509655751469353,
      "grad_norm": 3.9433722496032715,
      "learning_rate": 7.164480322906156e-06,
      "loss": 6.1323,
      "step": 56000
    },
    {
      "epoch": 23.719563392107474,
      "grad_norm": 3.012051820755005,
      "learning_rate": 6.155398587285571e-06,
      "loss": 6.1255,
      "step": 56500
    },
    {
      "epoch": 23.929471032745592,
      "grad_norm": 4.664351463317871,
      "learning_rate": 5.146316851664986e-06,
      "loss": 6.1315,
      "step": 57000
    },
    {
      "epoch": 24.13937867338371,
      "grad_norm": 4.059649467468262,
      "learning_rate": 4.1372351160444e-06,
      "loss": 6.1238,
      "step": 57500
    },
    {
      "epoch": 24.349286314021832,
      "grad_norm": 3.9240424633026123,
      "learning_rate": 3.1281533804238145e-06,
      "loss": 6.1213,
      "step": 58000
    },
    {
      "epoch": 24.55919395465995,
      "grad_norm": 4.281371116638184,
      "learning_rate": 2.1190716448032293e-06,
      "loss": 6.1177,
      "step": 58500
    },
    {
      "epoch": 24.769101595298068,
      "grad_norm": 2.9702208042144775,
      "learning_rate": 1.109989909182644e-06,
      "loss": 6.1155,
      "step": 59000
    },
    {
      "epoch": 24.97900923593619,
      "grad_norm": 2.3278939723968506,
      "learning_rate": 1.0090817356205853e-07,
      "loss": 6.1164,
      "step": 59500
    },
    {
      "epoch": 25.0,
      "step": 59550,
      "total_flos": 3.053850510375936e+17,
      "train_loss": 6.746473053004139,
      "train_runtime": 11362.3718,
      "train_samples_per_second": 503.062,
      "train_steps_per_second": 5.241
    }
  ],
  "logging_steps": 500,
  "max_steps": 59550,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 10000,
  "total_flos": 3.053850510375936e+17,
  "train_batch_size": 96,
  "trial_name": null,
  "trial_params": null
}