{
  "best_global_step": 20000,
  "best_metric": 6.545135974884033,
  "best_model_checkpoint": "distilbert_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": 3.1452674865722656,
      "learning_rate": 5e-06,
      "loss": 12.261,
      "step": 500
    },
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 3.8694403171539307,
      "learning_rate": 1e-05,
      "loss": 10.3584,
      "step": 1000
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 6.9578471183776855,
      "learning_rate": 1.5e-05,
      "loss": 8.9096,
      "step": 1500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 8.066250801086426,
      "learning_rate": 2e-05,
      "loss": 8.186,
      "step": 2000
    },
    {
      "epoch": 1.0495382031905962,
      "grad_norm": 7.291591644287109,
      "learning_rate": 2.5e-05,
      "loss": 7.8507,
      "step": 2500
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 8.559544563293457,
      "learning_rate": 3e-05,
      "loss": 7.6129,
      "step": 3000
    },
    {
      "epoch": 1.4693534844668346,
      "grad_norm": 10.504677772521973,
      "learning_rate": 3.5e-05,
      "loss": 7.4428,
      "step": 3500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 5.783501625061035,
      "learning_rate": 4e-05,
      "loss": 7.2844,
      "step": 4000
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 4.191773414611816,
      "learning_rate": 4.5e-05,
      "loss": 7.1556,
      "step": 4500
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 3.775890588760376,
      "learning_rate": 5e-05,
      "loss": 7.0265,
      "step": 5000
    },
    {
      "epoch": 2.3089840470193117,
      "grad_norm": 4.755401134490967,
      "learning_rate": 5.500000000000001e-05,
      "loss": 6.9304,
      "step": 5500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 5.325192928314209,
      "learning_rate": 6e-05,
      "loss": 6.899,
      "step": 6000
    },
    {
      "epoch": 2.72879932829555,
      "grad_norm": 5.6083197593688965,
      "learning_rate": 6.500000000000001e-05,
      "loss": 6.861,
      "step": 6500
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 3.9431545734405518,
      "learning_rate": 7e-05,
      "loss": 6.8246,
      "step": 7000
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 2.4364116191864014,
      "learning_rate": 7.500000000000001e-05,
      "loss": 6.7309,
      "step": 7500
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 4.049054145812988,
      "learning_rate": 8e-05,
      "loss": 6.6998,
      "step": 8000
    },
    {
      "epoch": 3.568429890848027,
      "grad_norm": 2.9931015968322754,
      "learning_rate": 8.5e-05,
      "loss": 6.7119,
      "step": 8500
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 2.900836229324341,
      "learning_rate": 9e-05,
      "loss": 6.7069,
      "step": 9000
    },
    {
      "epoch": 3.988245172124265,
      "grad_norm": 2.292576551437378,
      "learning_rate": 9.5e-05,
      "loss": 6.6867,
      "step": 9500
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 3.8952012062072754,
      "learning_rate": 0.0001,
      "loss": 6.5819,
      "step": 10000
    },
    {
      "epoch": 4.198152812762385,
      "eval_accuracy": 0.1500876485155402,
      "eval_loss": 6.729597091674805,
      "eval_runtime": 0.8324,
      "eval_samples_per_second": 575.458,
      "eval_steps_per_second": 6.007,
      "step": 10000
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 4.258574485778809,
      "learning_rate": 9.899091826437942e-05,
      "loss": 6.5948,
      "step": 10500
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 3.889688491821289,
      "learning_rate": 9.798183652875883e-05,
      "loss": 6.5882,
      "step": 11000
    },
    {
      "epoch": 4.827875734676742,
      "grad_norm": 2.8931455612182617,
      "learning_rate": 9.697275479313825e-05,
      "loss": 6.5881,
      "step": 11500
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 1.7556513547897339,
      "learning_rate": 9.596367305751767e-05,
      "loss": 6.5572,
      "step": 12000
    },
    {
      "epoch": 5.247691015952981,
      "grad_norm": 2.560131072998047,
      "learning_rate": 9.495459132189707e-05,
      "loss": 6.465,
      "step": 12500
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 2.8700766563415527,
      "learning_rate": 9.394550958627649e-05,
      "loss": 6.4725,
      "step": 13000
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 2.16776967048645,
      "learning_rate": 9.293642785065591e-05,
      "loss": 6.4749,
      "step": 13500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 3.95516037940979,
      "learning_rate": 9.192734611503532e-05,
      "loss": 6.4696,
      "step": 14000
    },
    {
      "epoch": 6.087321578505458,
      "grad_norm": 2.1974260807037354,
      "learning_rate": 9.091826437941474e-05,
      "loss": 6.4287,
      "step": 14500
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 2.972895622253418,
      "learning_rate": 8.990918264379415e-05,
      "loss": 6.3753,
      "step": 15000
    },
    {
      "epoch": 6.507136859781696,
      "grad_norm": 1.6085268259048462,
      "learning_rate": 8.890010090817356e-05,
      "loss": 6.3764,
      "step": 15500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 3.43790340423584,
      "learning_rate": 8.789101917255298e-05,
      "loss": 6.3761,
      "step": 16000
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 2.285003662109375,
      "learning_rate": 8.68819374369324e-05,
      "loss": 6.3822,
      "step": 16500
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 2.4064154624938965,
      "learning_rate": 8.58728557013118e-05,
      "loss": 6.3183,
      "step": 17000
    },
    {
      "epoch": 7.346767422334173,
      "grad_norm": 1.8181557655334473,
      "learning_rate": 8.486377396569122e-05,
      "loss": 6.295,
      "step": 17500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 1.9758304357528687,
      "learning_rate": 8.385469223007064e-05,
      "loss": 6.3043,
      "step": 18000
    },
    {
      "epoch": 7.766582703610411,
      "grad_norm": 3.297147274017334,
      "learning_rate": 8.284561049445006e-05,
      "loss": 6.3055,
      "step": 18500
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 2.5799708366394043,
      "learning_rate": 8.183652875882947e-05,
      "loss": 6.3099,
      "step": 19000
    },
    {
      "epoch": 8.18639798488665,
      "grad_norm": 1.6020463705062866,
      "learning_rate": 8.082744702320889e-05,
      "loss": 6.2404,
      "step": 19500
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 2.8870275020599365,
      "learning_rate": 7.98183652875883e-05,
      "loss": 6.2427,
      "step": 20000
    },
    {
      "epoch": 8.39630562552477,
      "eval_accuracy": 0.153287129268498,
      "eval_loss": 6.545135974884033,
      "eval_runtime": 0.8345,
      "eval_samples_per_second": 574.016,
      "eval_steps_per_second": 5.992,
      "step": 20000
    },
    {
      "epoch": 8.606213266162888,
      "grad_norm": 2.7224345207214355,
      "learning_rate": 7.880928355196771e-05,
      "loss": 6.2408,
      "step": 20500
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 3.342298984527588,
      "learning_rate": 7.780020181634713e-05,
      "loss": 6.2508,
      "step": 21000
    },
    {
      "epoch": 9.026028547439127,
      "grad_norm": 2.7848408222198486,
      "learning_rate": 7.679112008072655e-05,
      "loss": 6.2423,
      "step": 21500
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 2.56316876411438,
      "learning_rate": 7.578203834510595e-05,
      "loss": 6.1933,
      "step": 22000
    },
    {
      "epoch": 9.445843828715365,
      "grad_norm": 2.9379024505615234,
      "learning_rate": 7.477295660948537e-05,
      "loss": 6.1899,
      "step": 22500
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 3.125236988067627,
      "learning_rate": 7.376387487386479e-05,
      "loss": 6.1954,
      "step": 23000
    },
    {
      "epoch": 9.865659109991604,
      "grad_norm": 1.8371341228485107,
      "learning_rate": 7.27547931382442e-05,
      "loss": 6.1935,
      "step": 23500
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 2.501499891281128,
      "learning_rate": 7.174571140262362e-05,
      "loss": 6.1719,
      "step": 24000
    },
    {
      "epoch": 10.285474391267842,
      "grad_norm": 3.183001756668091,
      "learning_rate": 7.073662966700304e-05,
      "loss": 6.1509,
      "step": 24500
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 2.322767496109009,
      "learning_rate": 6.972754793138244e-05,
      "loss": 6.147,
      "step": 25000
    },
    {
      "epoch": 10.705289672544081,
      "grad_norm": 1.6661608219146729,
      "learning_rate": 6.871846619576186e-05,
      "loss": 6.1462,
      "step": 25500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 2.1413402557373047,
      "learning_rate": 6.770938446014128e-05,
      "loss": 6.1538,
      "step": 26000
    },
    {
      "epoch": 11.125104953820319,
      "grad_norm": 2.5893688201904297,
      "learning_rate": 6.670030272452068e-05,
      "loss": 6.1272,
      "step": 26500
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 3.3126704692840576,
      "learning_rate": 6.56912209889001e-05,
      "loss": 6.1133,
      "step": 27000
    },
    {
      "epoch": 11.544920235096557,
      "grad_norm": 2.832763910293579,
      "learning_rate": 6.468213925327952e-05,
      "loss": 6.1124,
      "step": 27500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 2.2899630069732666,
      "learning_rate": 6.367305751765893e-05,
      "loss": 6.1087,
      "step": 28000
    },
    {
      "epoch": 11.964735516372796,
      "grad_norm": 2.9168951511383057,
      "learning_rate": 6.266397578203835e-05,
      "loss": 6.1099,
      "step": 28500
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 2.782327175140381,
      "learning_rate": 6.165489404641777e-05,
      "loss": 6.0873,
      "step": 29000
    },
    {
      "epoch": 12.384550797649034,
      "grad_norm": 2.4683403968811035,
      "learning_rate": 6.064581231079718e-05,
      "loss": 6.0806,
      "step": 29500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 2.769644021987915,
      "learning_rate": 5.963673057517659e-05,
      "loss": 6.0849,
      "step": 30000
    },
    {
      "epoch": 12.594458438287154,
      "eval_accuracy": 0.15440579791252762,
      "eval_loss": 6.642980575561523,
      "eval_runtime": 0.8114,
      "eval_samples_per_second": 590.303,
      "eval_steps_per_second": 6.162,
      "step": 30000
    },
    {
      "epoch": 12.804366078925273,
      "grad_norm": 3.0742828845977783,
      "learning_rate": 5.8627648839556e-05,
      "loss": 6.0772,
      "step": 30500
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 3.4945228099823,
      "learning_rate": 5.7618567103935416e-05,
      "loss": 6.08,
      "step": 31000
    },
    {
      "epoch": 13.224181360201511,
      "grad_norm": 2.476950168609619,
      "learning_rate": 5.660948536831484e-05,
      "loss": 6.0543,
      "step": 31500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 3.351191997528076,
      "learning_rate": 5.5600403632694254e-05,
      "loss": 6.0539,
      "step": 32000
    },
    {
      "epoch": 13.64399664147775,
      "grad_norm": 2.638902187347412,
      "learning_rate": 5.4591321897073666e-05,
      "loss": 6.0555,
      "step": 32500
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 2.204129695892334,
      "learning_rate": 5.358224016145308e-05,
      "loss": 6.0547,
      "step": 33000
    },
    {
      "epoch": 14.063811922753988,
      "grad_norm": 1.0944510698318481,
      "learning_rate": 5.257315842583249e-05,
      "loss": 6.0422,
      "step": 33500
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 3.2853310108184814,
      "learning_rate": 5.15640766902119e-05,
      "loss": 6.0308,
      "step": 34000
    },
    {
      "epoch": 14.483627204030226,
      "grad_norm": 2.579535484313965,
      "learning_rate": 5.055499495459133e-05,
      "loss": 6.0299,
      "step": 34500
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 1.7781260013580322,
      "learning_rate": 4.954591321897074e-05,
      "loss": 6.0272,
      "step": 35000
    },
    {
      "epoch": 14.903442485306465,
      "grad_norm": 1.3856323957443237,
      "learning_rate": 4.853683148335015e-05,
      "loss": 6.0275,
      "step": 35500
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 2.3442111015319824,
      "learning_rate": 4.7527749747729566e-05,
      "loss": 6.0157,
      "step": 36000
    },
    {
      "epoch": 15.323257766582703,
      "grad_norm": 1.6177752017974854,
      "learning_rate": 4.6518668012108985e-05,
      "loss": 6.0139,
      "step": 36500
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 3.992405414581299,
      "learning_rate": 4.55095862764884e-05,
      "loss": 6.0051,
      "step": 37000
    },
    {
      "epoch": 15.743073047858942,
      "grad_norm": 2.8688621520996094,
      "learning_rate": 4.450050454086781e-05,
      "loss": 6.007,
      "step": 37500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 2.171865224838257,
      "learning_rate": 4.349142280524723e-05,
      "loss": 6.003,
      "step": 38000
    },
    {
      "epoch": 16.162888329135182,
      "grad_norm": 1.6749943494796753,
      "learning_rate": 4.248234106962664e-05,
      "loss": 5.9944,
      "step": 38500
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 1.6332727670669556,
      "learning_rate": 4.147325933400605e-05,
      "loss": 5.9856,
      "step": 39000
    },
    {
      "epoch": 16.582703610411418,
      "grad_norm": 2.5190377235412598,
      "learning_rate": 4.046417759838547e-05,
      "loss": 5.9882,
      "step": 39500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 0.7850143313407898,
      "learning_rate": 3.9455095862764884e-05,
      "loss": 5.981,
      "step": 40000
    },
    {
      "epoch": 16.79261125104954,
      "eval_accuracy": 0.15291017158237075,
      "eval_loss": 6.637546062469482,
      "eval_runtime": 0.81,
      "eval_samples_per_second": 591.327,
      "eval_steps_per_second": 6.173,
      "step": 40000
    },
    {
      "epoch": 17.002518891687657,
      "grad_norm": 1.6559672355651855,
      "learning_rate": 3.84460141271443e-05,
      "loss": 5.9858,
      "step": 40500
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 1.9162330627441406,
      "learning_rate": 3.7436932391523715e-05,
      "loss": 5.9709,
      "step": 41000
    },
    {
      "epoch": 17.422334172963897,
      "grad_norm": 1.2056223154067993,
      "learning_rate": 3.642785065590313e-05,
      "loss": 5.9673,
      "step": 41500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 2.0800726413726807,
      "learning_rate": 3.541876892028255e-05,
      "loss": 5.968,
      "step": 42000
    },
    {
      "epoch": 17.842149454240133,
      "grad_norm": 2.444692611694336,
      "learning_rate": 3.440968718466196e-05,
      "loss": 5.9631,
      "step": 42500
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 1.0705031156539917,
      "learning_rate": 3.340060544904137e-05,
      "loss": 5.9624,
      "step": 43000
    },
    {
      "epoch": 18.261964735516372,
      "grad_norm": 0.7841247320175171,
      "learning_rate": 3.239152371342079e-05,
      "loss": 5.9492,
      "step": 43500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 1.7728303670883179,
      "learning_rate": 3.13824419778002e-05,
      "loss": 5.9518,
      "step": 44000
    },
    {
      "epoch": 18.68178001679261,
      "grad_norm": 2.2180380821228027,
      "learning_rate": 3.0373360242179615e-05,
      "loss": 5.9502,
      "step": 44500
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 3.176880359649658,
      "learning_rate": 2.9364278506559034e-05,
      "loss": 5.9516,
      "step": 45000
    },
    {
      "epoch": 19.10159529806885,
      "grad_norm": 0.8827639818191528,
      "learning_rate": 2.8355196770938446e-05,
      "loss": 5.9464,
      "step": 45500
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 0.8466761708259583,
      "learning_rate": 2.734611503531786e-05,
      "loss": 5.9331,
      "step": 46000
    },
    {
      "epoch": 19.521410579345087,
      "grad_norm": 0.9533854126930237,
      "learning_rate": 2.6337033299697278e-05,
      "loss": 5.9362,
      "step": 46500
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 2.770329475402832,
      "learning_rate": 2.532795156407669e-05,
      "loss": 5.9359,
      "step": 47000
    },
    {
      "epoch": 19.941225860621326,
      "grad_norm": 2.179800033569336,
      "learning_rate": 2.4318869828456106e-05,
      "loss": 5.9333,
      "step": 47500
    },
    {
      "epoch": 20.151133501259444,
      "grad_norm": 1.3078432083129883,
      "learning_rate": 2.330978809283552e-05,
      "loss": 5.9278,
      "step": 48000
    },
    {
      "epoch": 20.361041141897566,
      "grad_norm": 1.246522307395935,
      "learning_rate": 2.2300706357214934e-05,
      "loss": 5.9247,
      "step": 48500
    },
    {
      "epoch": 20.570948782535684,
      "grad_norm": 1.245393991470337,
      "learning_rate": 2.129162462159435e-05,
      "loss": 5.9197,
      "step": 49000
    },
    {
      "epoch": 20.7808564231738,
      "grad_norm": 1.014298677444458,
      "learning_rate": 2.0282542885973765e-05,
      "loss": 5.9173,
      "step": 49500
    },
    {
      "epoch": 20.990764063811923,
      "grad_norm": 2.5100197792053223,
      "learning_rate": 1.927346115035318e-05,
      "loss": 5.9158,
      "step": 50000
    },
    {
      "epoch": 20.990764063811923,
      "eval_accuracy": 0.1510359716264334,
      "eval_loss": 6.70709228515625,
      "eval_runtime": 0.8142,
      "eval_samples_per_second": 588.285,
      "eval_steps_per_second": 6.141,
      "step": 50000
    },
    {
      "epoch": 21.20067170445004,
      "grad_norm": 3.72275710105896,
      "learning_rate": 1.8264379414732593e-05,
      "loss": 5.9088,
      "step": 50500
    },
    {
      "epoch": 21.410579345088163,
      "grad_norm": 2.1459269523620605,
      "learning_rate": 1.725529767911201e-05,
      "loss": 5.9123,
      "step": 51000
    },
    {
      "epoch": 21.62048698572628,
      "grad_norm": 3.2029337882995605,
      "learning_rate": 1.6246215943491424e-05,
      "loss": 5.9093,
      "step": 51500
    },
    {
      "epoch": 21.8303946263644,
      "grad_norm": 3.592463493347168,
      "learning_rate": 1.5237134207870837e-05,
      "loss": 5.9086,
      "step": 52000
    },
    {
      "epoch": 22.04030226700252,
      "grad_norm": 0.7952311038970947,
      "learning_rate": 1.4228052472250252e-05,
      "loss": 5.9067,
      "step": 52500
    },
    {
      "epoch": 22.250209907640638,
      "grad_norm": 3.0370988845825195,
      "learning_rate": 1.3218970736629668e-05,
      "loss": 5.8962,
      "step": 53000
    },
    {
      "epoch": 22.460117548278756,
      "grad_norm": 1.8484214544296265,
      "learning_rate": 1.2209889001009082e-05,
      "loss": 5.9004,
      "step": 53500
    },
    {
      "epoch": 22.670025188916878,
      "grad_norm": 0.8853672742843628,
      "learning_rate": 1.1200807265388496e-05,
      "loss": 5.9016,
      "step": 54000
    },
    {
      "epoch": 22.879932829554996,
      "grad_norm": 1.6122455596923828,
      "learning_rate": 1.0191725529767912e-05,
      "loss": 5.899,
      "step": 54500
    },
    {
      "epoch": 23.089840470193113,
      "grad_norm": 0.7871816754341125,
      "learning_rate": 9.182643794147326e-06,
      "loss": 5.8927,
      "step": 55000
    },
    {
      "epoch": 23.299748110831235,
      "grad_norm": 0.8111832141876221,
      "learning_rate": 8.173562058526741e-06,
      "loss": 5.8922,
      "step": 55500
    },
    {
      "epoch": 23.509655751469353,
      "grad_norm": 3.7575721740722656,
      "learning_rate": 7.164480322906156e-06,
      "loss": 5.8922,
      "step": 56000
    },
    {
      "epoch": 23.719563392107474,
      "grad_norm": 0.7564641833305359,
      "learning_rate": 6.155398587285571e-06,
      "loss": 5.8864,
      "step": 56500
    },
    {
      "epoch": 23.929471032745592,
      "grad_norm": 0.7905524969100952,
      "learning_rate": 5.146316851664986e-06,
      "loss": 5.89,
      "step": 57000
    },
    {
      "epoch": 24.13937867338371,
      "grad_norm": 2.1481614112854004,
      "learning_rate": 4.1372351160444e-06,
      "loss": 5.8844,
      "step": 57500
    },
    {
      "epoch": 24.349286314021832,
      "grad_norm": 0.798080325126648,
      "learning_rate": 3.1281533804238145e-06,
      "loss": 5.8862,
      "step": 58000
    },
    {
      "epoch": 24.55919395465995,
      "grad_norm": 0.7877392172813416,
      "learning_rate": 2.1190716448032293e-06,
      "loss": 5.8835,
      "step": 58500
    },
    {
      "epoch": 24.769101595298068,
      "grad_norm": 0.7649791240692139,
      "learning_rate": 1.109989909182644e-06,
      "loss": 5.8822,
      "step": 59000
    },
    {
      "epoch": 24.97900923593619,
      "grad_norm": 1.1933664083480835,
      "learning_rate": 1.0090817356205853e-07,
      "loss": 5.8834,
      "step": 59500
    },
    {
      "epoch": 25.0,
      "step": 59550,
      "total_flos": 7.68356804521728e+17,
      "train_loss": 6.33944412602986,
      "train_runtime": 16958.998,
      "train_samples_per_second": 337.047,
      "train_steps_per_second": 3.511
    }
  ],
  "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": 7.68356804521728e+17,
  "train_batch_size": 96,
  "trial_name": null,
  "trial_params": null
}