{
  "best_global_step": 20000,
  "best_metric": 6.480999946594238,
  "best_model_checkpoint": "bert_base_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": 3.3506252765655518,
      "learning_rate": 5e-06,
      "loss": 12.1648,
      "step": 500
    },
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 5.457421779632568,
      "learning_rate": 1e-05,
      "loss": 10.0986,
      "step": 1000
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 23.159326553344727,
      "learning_rate": 1.5e-05,
      "loss": 8.7021,
      "step": 1500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 8.975157737731934,
      "learning_rate": 2e-05,
      "loss": 8.1464,
      "step": 2000
    },
    {
      "epoch": 1.0495382031905962,
      "grad_norm": 10.226245880126953,
      "learning_rate": 2.5e-05,
      "loss": 7.8208,
      "step": 2500
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 9.012547492980957,
      "learning_rate": 3e-05,
      "loss": 7.5779,
      "step": 3000
    },
    {
      "epoch": 1.4693534844668346,
      "grad_norm": 8.201004028320312,
      "learning_rate": 3.5e-05,
      "loss": 7.4349,
      "step": 3500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 6.068079471588135,
      "learning_rate": 4e-05,
      "loss": 7.3029,
      "step": 4000
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 4.390971660614014,
      "learning_rate": 4.5e-05,
      "loss": 7.189,
      "step": 4500
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 4.879016399383545,
      "learning_rate": 5e-05,
      "loss": 7.0702,
      "step": 5000
    },
    {
      "epoch": 2.3089840470193117,
      "grad_norm": 4.469964981079102,
      "learning_rate": 5.500000000000001e-05,
      "loss": 6.9738,
      "step": 5500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 4.512015342712402,
      "learning_rate": 6e-05,
      "loss": 6.9463,
      "step": 6000
    },
    {
      "epoch": 2.72879932829555,
      "grad_norm": 3.9740757942199707,
      "learning_rate": 6.500000000000001e-05,
      "loss": 6.8989,
      "step": 6500
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 4.45874547958374,
      "learning_rate": 7e-05,
      "loss": 6.8596,
      "step": 7000
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 3.5402514934539795,
      "learning_rate": 7.500000000000001e-05,
      "loss": 6.7614,
      "step": 7500
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 3.2395708560943604,
      "learning_rate": 8e-05,
      "loss": 6.7273,
      "step": 8000
    },
    {
      "epoch": 3.568429890848027,
      "grad_norm": 2.3133621215820312,
      "learning_rate": 8.5e-05,
      "loss": 6.737,
      "step": 8500
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 2.676318883895874,
      "learning_rate": 9e-05,
      "loss": 6.7239,
      "step": 9000
    },
    {
      "epoch": 3.988245172124265,
      "grad_norm": 2.319993734359741,
      "learning_rate": 9.5e-05,
      "loss": 6.7111,
      "step": 9500
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 4.243929386138916,
      "learning_rate": 0.0001,
      "loss": 6.6058,
      "step": 10000
    },
    {
      "epoch": 4.198152812762385,
      "eval_accuracy": 0.1499485238876986,
      "eval_loss": 6.702857971191406,
      "eval_runtime": 1.0645,
      "eval_samples_per_second": 449.956,
      "eval_steps_per_second": 4.697,
      "step": 10000
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 4.093377590179443,
      "learning_rate": 9.899091826437942e-05,
      "loss": 6.6124,
      "step": 10500
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 3.2889018058776855,
      "learning_rate": 9.798183652875883e-05,
      "loss": 6.6077,
      "step": 11000
    },
    {
      "epoch": 4.827875734676742,
      "grad_norm": 2.30513334274292,
      "learning_rate": 9.697275479313825e-05,
      "loss": 6.6029,
      "step": 11500
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 2.541769027709961,
      "learning_rate": 9.596367305751767e-05,
      "loss": 6.5729,
      "step": 12000
    },
    {
      "epoch": 5.247691015952981,
      "grad_norm": 2.6730828285217285,
      "learning_rate": 9.495459132189707e-05,
      "loss": 6.4829,
      "step": 12500
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 2.574977159500122,
      "learning_rate": 9.394550958627649e-05,
      "loss": 6.4891,
      "step": 13000
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 2.663959264755249,
      "learning_rate": 9.293642785065591e-05,
      "loss": 6.4903,
      "step": 13500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 3.6568427085876465,
      "learning_rate": 9.192734611503532e-05,
      "loss": 6.4806,
      "step": 14000
    },
    {
      "epoch": 6.087321578505458,
      "grad_norm": 2.5574238300323486,
      "learning_rate": 9.091826437941474e-05,
      "loss": 6.4354,
      "step": 14500
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 2.4129903316497803,
      "learning_rate": 8.990918264379415e-05,
      "loss": 6.3888,
      "step": 15000
    },
    {
      "epoch": 6.507136859781696,
      "grad_norm": 2.638054132461548,
      "learning_rate": 8.890010090817356e-05,
      "loss": 6.3903,
      "step": 15500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 2.9566128253936768,
      "learning_rate": 8.789101917255298e-05,
      "loss": 6.3851,
      "step": 16000
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 2.3400323390960693,
      "learning_rate": 8.68819374369324e-05,
      "loss": 6.3894,
      "step": 16500
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 2.8594205379486084,
      "learning_rate": 8.58728557013118e-05,
      "loss": 6.325,
      "step": 17000
    },
    {
      "epoch": 7.346767422334173,
      "grad_norm": 1.7151496410369873,
      "learning_rate": 8.486377396569122e-05,
      "loss": 6.3014,
      "step": 17500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 2.114755630493164,
      "learning_rate": 8.385469223007064e-05,
      "loss": 6.3106,
      "step": 18000
    },
    {
      "epoch": 7.766582703610411,
      "grad_norm": 3.3576650619506836,
      "learning_rate": 8.284561049445006e-05,
      "loss": 6.31,
      "step": 18500
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 2.4794225692749023,
      "learning_rate": 8.183652875882947e-05,
      "loss": 6.3136,
      "step": 19000
    },
    {
      "epoch": 8.18639798488665,
      "grad_norm": 2.520782470703125,
      "learning_rate": 8.082744702320889e-05,
      "loss": 6.2433,
      "step": 19500
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 2.2647593021392822,
      "learning_rate": 7.98183652875883e-05,
      "loss": 6.2431,
      "step": 20000
    },
    {
      "epoch": 8.39630562552477,
      "eval_accuracy": 0.15323101097112712,
      "eval_loss": 6.480999946594238,
      "eval_runtime": 1.0251,
      "eval_samples_per_second": 467.277,
      "eval_steps_per_second": 4.878,
      "step": 20000
    },
    {
      "epoch": 8.606213266162888,
      "grad_norm": 2.691263198852539,
      "learning_rate": 7.880928355196771e-05,
      "loss": 6.2411,
      "step": 20500
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 3.366475820541382,
      "learning_rate": 7.780020181634713e-05,
      "loss": 6.2467,
      "step": 21000
    },
    {
      "epoch": 9.026028547439127,
      "grad_norm": 2.6554746627807617,
      "learning_rate": 7.679112008072655e-05,
      "loss": 6.2374,
      "step": 21500
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 2.4414002895355225,
      "learning_rate": 7.578203834510595e-05,
      "loss": 6.185,
      "step": 22000
    },
    {
      "epoch": 9.445843828715365,
      "grad_norm": 2.6609840393066406,
      "learning_rate": 7.477295660948537e-05,
      "loss": 6.1858,
      "step": 22500
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 2.6620678901672363,
      "learning_rate": 7.376387487386479e-05,
      "loss": 6.1865,
      "step": 23000
    },
    {
      "epoch": 9.865659109991604,
      "grad_norm": 1.7680195569992065,
      "learning_rate": 7.27547931382442e-05,
      "loss": 6.1879,
      "step": 23500
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 2.300396680831909,
      "learning_rate": 7.174571140262362e-05,
      "loss": 6.159,
      "step": 24000
    },
    {
      "epoch": 10.285474391267842,
      "grad_norm": 2.125763177871704,
      "learning_rate": 7.073662966700304e-05,
      "loss": 6.1433,
      "step": 24500
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 3.4411227703094482,
      "learning_rate": 6.972754793138244e-05,
      "loss": 6.1372,
      "step": 25000
    },
    {
      "epoch": 10.705289672544081,
      "grad_norm": 2.8306920528411865,
      "learning_rate": 6.871846619576186e-05,
      "loss": 6.1403,
      "step": 25500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 2.745607614517212,
      "learning_rate": 6.770938446014128e-05,
      "loss": 6.1449,
      "step": 26000
    },
    {
      "epoch": 11.125104953820319,
      "grad_norm": 3.7489781379699707,
      "learning_rate": 6.670030272452068e-05,
      "loss": 6.114,
      "step": 26500
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 2.851313352584839,
      "learning_rate": 6.56912209889001e-05,
      "loss": 6.1028,
      "step": 27000
    },
    {
      "epoch": 11.544920235096557,
      "grad_norm": 2.392357349395752,
      "learning_rate": 6.468213925327952e-05,
      "loss": 6.0983,
      "step": 27500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 3.2635204792022705,
      "learning_rate": 6.367305751765893e-05,
      "loss": 6.0985,
      "step": 28000
    },
    {
      "epoch": 11.964735516372796,
      "grad_norm": 1.8090300559997559,
      "learning_rate": 6.266397578203835e-05,
      "loss": 6.0976,
      "step": 28500
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 2.6949682235717773,
      "learning_rate": 6.165489404641777e-05,
      "loss": 6.0729,
      "step": 29000
    },
    {
      "epoch": 12.384550797649034,
      "grad_norm": 2.046905755996704,
      "learning_rate": 6.064581231079718e-05,
      "loss": 6.0621,
      "step": 29500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 2.0488154888153076,
      "learning_rate": 5.963673057517659e-05,
      "loss": 6.0675,
      "step": 30000
    },
    {
      "epoch": 12.594458438287154,
      "eval_accuracy": 0.15462965553907715,
      "eval_loss": 6.531682014465332,
      "eval_runtime": 1.0347,
      "eval_samples_per_second": 462.951,
      "eval_steps_per_second": 4.832,
      "step": 30000
    },
    {
      "epoch": 12.804366078925273,
      "grad_norm": 1.8668218851089478,
      "learning_rate": 5.8627648839556e-05,
      "loss": 6.0623,
      "step": 30500
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 3.1596615314483643,
      "learning_rate": 5.7618567103935416e-05,
      "loss": 6.0697,
      "step": 31000
    },
    {
      "epoch": 13.224181360201511,
      "grad_norm": 2.794318437576294,
      "learning_rate": 5.660948536831484e-05,
      "loss": 6.037,
      "step": 31500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 3.688591718673706,
      "learning_rate": 5.5600403632694254e-05,
      "loss": 6.0389,
      "step": 32000
    },
    {
      "epoch": 13.64399664147775,
      "grad_norm": 3.4934194087982178,
      "learning_rate": 5.4591321897073666e-05,
      "loss": 6.0363,
      "step": 32500
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 4.7184977531433105,
      "learning_rate": 5.358224016145308e-05,
      "loss": 6.0373,
      "step": 33000
    },
    {
      "epoch": 14.063811922753988,
      "grad_norm": 2.1444015502929688,
      "learning_rate": 5.257315842583249e-05,
      "loss": 6.0246,
      "step": 33500
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 3.46429443359375,
      "learning_rate": 5.15640766902119e-05,
      "loss": 6.0098,
      "step": 34000
    },
    {
      "epoch": 14.483627204030226,
      "grad_norm": 1.9107475280761719,
      "learning_rate": 5.055499495459133e-05,
      "loss": 6.0094,
      "step": 34500
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 1.8858654499053955,
      "learning_rate": 4.954591321897074e-05,
      "loss": 6.008,
      "step": 35000
    },
    {
      "epoch": 14.903442485306465,
      "grad_norm": 1.9284015893936157,
      "learning_rate": 4.853683148335015e-05,
      "loss": 6.0067,
      "step": 35500
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 1.400378942489624,
      "learning_rate": 4.7527749747729566e-05,
      "loss": 5.9918,
      "step": 36000
    },
    {
      "epoch": 15.323257766582703,
      "grad_norm": 1.1706284284591675,
      "learning_rate": 4.6518668012108985e-05,
      "loss": 5.9912,
      "step": 36500
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 2.7408406734466553,
      "learning_rate": 4.55095862764884e-05,
      "loss": 5.9854,
      "step": 37000
    },
    {
      "epoch": 15.743073047858942,
      "grad_norm": 2.697648048400879,
      "learning_rate": 4.450050454086781e-05,
      "loss": 5.9834,
      "step": 37500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 3.865612268447876,
      "learning_rate": 4.349142280524723e-05,
      "loss": 5.9813,
      "step": 38000
    },
    {
      "epoch": 16.162888329135182,
      "grad_norm": 0.8426138758659363,
      "learning_rate": 4.248234106962664e-05,
      "loss": 5.9715,
      "step": 38500
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 3.171299934387207,
      "learning_rate": 4.147325933400605e-05,
      "loss": 5.9639,
      "step": 39000
    },
    {
      "epoch": 16.582703610411418,
      "grad_norm": 4.164156913757324,
      "learning_rate": 4.046417759838547e-05,
      "loss": 5.9665,
      "step": 39500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 2.041938066482544,
      "learning_rate": 3.9455095862764884e-05,
      "loss": 5.9569,
      "step": 40000
    },
    {
      "epoch": 16.79261125104954,
      "eval_accuracy": 0.15302231692273185,
      "eval_loss": 6.651042938232422,
      "eval_runtime": 1.021,
      "eval_samples_per_second": 469.166,
      "eval_steps_per_second": 4.897,
      "step": 40000
    },
    {
      "epoch": 17.002518891687657,
      "grad_norm": 3.150923013687134,
      "learning_rate": 3.84460141271443e-05,
      "loss": 5.9617,
      "step": 40500
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 0.7818689346313477,
      "learning_rate": 3.7436932391523715e-05,
      "loss": 5.9453,
      "step": 41000
    },
    {
      "epoch": 17.422334172963897,
      "grad_norm": 4.910539150238037,
      "learning_rate": 3.642785065590313e-05,
      "loss": 5.9428,
      "step": 41500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 2.48958683013916,
      "learning_rate": 3.541876892028255e-05,
      "loss": 5.9432,
      "step": 42000
    },
    {
      "epoch": 17.842149454240133,
      "grad_norm": 2.8172192573547363,
      "learning_rate": 3.440968718466196e-05,
      "loss": 5.9413,
      "step": 42500
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 1.81460440158844,
      "learning_rate": 3.340060544904137e-05,
      "loss": 5.9341,
      "step": 43000
    },
    {
      "epoch": 18.261964735516372,
      "grad_norm": 1.0337363481521606,
      "learning_rate": 3.239152371342079e-05,
      "loss": 5.9247,
      "step": 43500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 1.9823477268218994,
      "learning_rate": 3.13824419778002e-05,
      "loss": 5.9255,
      "step": 44000
    },
    {
      "epoch": 18.68178001679261,
      "grad_norm": 0.8637967109680176,
      "learning_rate": 3.0373360242179615e-05,
      "loss": 5.9248,
      "step": 44500
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 0.872508704662323,
      "learning_rate": 2.9364278506559034e-05,
      "loss": 5.9261,
      "step": 45000
    },
    {
      "epoch": 19.10159529806885,
      "grad_norm": 1.7914263010025024,
      "learning_rate": 2.8355196770938446e-05,
      "loss": 5.921,
      "step": 45500
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 1.7649086713790894,
      "learning_rate": 2.734611503531786e-05,
      "loss": 5.9058,
      "step": 46000
    },
    {
      "epoch": 19.521410579345087,
      "grad_norm": 0.9645077586174011,
      "learning_rate": 2.6337033299697278e-05,
      "loss": 5.907,
      "step": 46500
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 1.0561953783035278,
      "learning_rate": 2.532795156407669e-05,
      "loss": 5.9097,
      "step": 47000
    },
    {
      "epoch": 19.941225860621326,
      "grad_norm": 0.970642626285553,
      "learning_rate": 2.4318869828456106e-05,
      "loss": 5.9055,
      "step": 47500
    },
    {
      "epoch": 20.151133501259444,
      "grad_norm": 0.8695760369300842,
      "learning_rate": 2.330978809283552e-05,
      "loss": 5.8992,
      "step": 48000
    },
    {
      "epoch": 20.361041141897566,
      "grad_norm": 1.0435950756072998,
      "learning_rate": 2.2300706357214934e-05,
      "loss": 5.8977,
      "step": 48500
    },
    {
      "epoch": 20.570948782535684,
      "grad_norm": 0.8484736680984497,
      "learning_rate": 2.129162462159435e-05,
      "loss": 5.8909,
      "step": 49000
    },
    {
      "epoch": 20.7808564231738,
      "grad_norm": 2.0572588443756104,
      "learning_rate": 2.0282542885973765e-05,
      "loss": 5.8899,
      "step": 49500
    },
    {
      "epoch": 20.990764063811923,
      "grad_norm": 0.7758699655532837,
      "learning_rate": 1.927346115035318e-05,
      "loss": 5.89,
      "step": 50000
    },
    {
      "epoch": 20.990764063811923,
      "eval_accuracy": 0.15235371632040823,
      "eval_loss": 6.6476731300354,
      "eval_runtime": 1.0305,
      "eval_samples_per_second": 464.818,
      "eval_steps_per_second": 4.852,
      "step": 50000
    },
    {
      "epoch": 21.20067170445004,
      "grad_norm": 1.1873598098754883,
      "learning_rate": 1.8264379414732593e-05,
      "loss": 5.8807,
      "step": 50500
    },
    {
      "epoch": 21.410579345088163,
      "grad_norm": 0.7782743573188782,
      "learning_rate": 1.725529767911201e-05,
      "loss": 5.8834,
      "step": 51000
    },
    {
      "epoch": 21.62048698572628,
      "grad_norm": 0.8238320350646973,
      "learning_rate": 1.6246215943491424e-05,
      "loss": 5.88,
      "step": 51500
    },
    {
      "epoch": 21.8303946263644,
      "grad_norm": 4.082571506500244,
      "learning_rate": 1.5237134207870837e-05,
      "loss": 5.8796,
      "step": 52000
    },
    {
      "epoch": 22.04030226700252,
      "grad_norm": 0.774568498134613,
      "learning_rate": 1.4228052472250252e-05,
      "loss": 5.8765,
      "step": 52500
    },
    {
      "epoch": 22.250209907640638,
      "grad_norm": 3.585805892944336,
      "learning_rate": 1.3218970736629668e-05,
      "loss": 5.8671,
      "step": 53000
    },
    {
      "epoch": 22.460117548278756,
      "grad_norm": 0.7828291654586792,
      "learning_rate": 1.2209889001009082e-05,
      "loss": 5.8694,
      "step": 53500
    },
    {
      "epoch": 22.670025188916878,
      "grad_norm": 1.8001397848129272,
      "learning_rate": 1.1200807265388496e-05,
      "loss": 5.8712,
      "step": 54000
    },
    {
      "epoch": 22.879932829554996,
      "grad_norm": 0.8042154908180237,
      "learning_rate": 1.0191725529767912e-05,
      "loss": 5.8654,
      "step": 54500
    },
    {
      "epoch": 23.089840470193113,
      "grad_norm": 3.4733877182006836,
      "learning_rate": 9.182643794147326e-06,
      "loss": 5.8632,
      "step": 55000
    },
    {
      "epoch": 23.299748110831235,
      "grad_norm": 0.7870926260948181,
      "learning_rate": 8.173562058526741e-06,
      "loss": 5.8613,
      "step": 55500
    },
    {
      "epoch": 23.509655751469353,
      "grad_norm": 3.1393957138061523,
      "learning_rate": 7.164480322906156e-06,
      "loss": 5.8606,
      "step": 56000
    },
    {
      "epoch": 23.719563392107474,
      "grad_norm": 0.7654667496681213,
      "learning_rate": 6.155398587285571e-06,
      "loss": 5.8556,
      "step": 56500
    },
    {
      "epoch": 23.929471032745592,
      "grad_norm": 0.7754602432250977,
      "learning_rate": 5.146316851664986e-06,
      "loss": 5.8596,
      "step": 57000
    },
    {
      "epoch": 24.13937867338371,
      "grad_norm": 0.7847160696983337,
      "learning_rate": 4.1372351160444e-06,
      "loss": 5.8528,
      "step": 57500
    },
    {
      "epoch": 24.349286314021832,
      "grad_norm": 0.8076699376106262,
      "learning_rate": 3.1281533804238145e-06,
      "loss": 5.854,
      "step": 58000
    },
    {
      "epoch": 24.55919395465995,
      "grad_norm": 3.1537649631500244,
      "learning_rate": 2.1190716448032293e-06,
      "loss": 5.851,
      "step": 58500
    },
    {
      "epoch": 24.769101595298068,
      "grad_norm": 0.7634235620498657,
      "learning_rate": 1.109989909182644e-06,
      "loss": 5.8503,
      "step": 59000
    },
    {
      "epoch": 24.97900923593619,
      "grad_norm": 4.646351337432861,
      "learning_rate": 1.0090817356205853e-07,
      "loss": 5.8501,
      "step": 59500
    },
    {
      "epoch": 25.0,
      "step": 59550,
      "total_flos": 1.5151126801503744e+18,
      "train_loss": 6.324693534844668,
      "train_runtime": 24831.9556,
      "train_samples_per_second": 230.186,
      "train_steps_per_second": 2.398
    }
  ],
  "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": 1.5151126801503744e+18,
  "train_batch_size": 96,
  "trial_name": null,
  "trial_params": null
}