{
  "best_global_step": 50000,
  "best_metric": 63.090843200683594,
  "best_model_checkpoint": "tinybert_base_train_kd/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": 1038.087890625,
      "learning_rate": 4.9900000000000005e-06,
      "loss": 6028.7805,
      "step": 500
    },
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 1000.9192504882812,
      "learning_rate": 9.990000000000001e-06,
      "loss": 3688.4922,
      "step": 1000
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 1443.9014892578125,
      "learning_rate": 1.499e-05,
      "loss": 2203.5858,
      "step": 1500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 2327.7587890625,
      "learning_rate": 1.999e-05,
      "loss": 1477.7162,
      "step": 2000
    },
    {
      "epoch": 1.0495382031905962,
      "grad_norm": 1928.0552978515625,
      "learning_rate": 2.4990000000000003e-05,
      "loss": 1145.1917,
      "step": 2500
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 1427.1600341796875,
      "learning_rate": 2.9990000000000003e-05,
      "loss": 941.4575,
      "step": 3000
    },
    {
      "epoch": 1.4693534844668346,
      "grad_norm": 832.8273315429688,
      "learning_rate": 3.499e-05,
      "loss": 812.8064,
      "step": 3500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 880.94140625,
      "learning_rate": 3.999e-05,
      "loss": 722.3781,
      "step": 4000
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 458.8153381347656,
      "learning_rate": 4.499e-05,
      "loss": 657.1395,
      "step": 4500
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 192.89340209960938,
      "learning_rate": 4.999e-05,
      "loss": 608.8174,
      "step": 5000
    },
    {
      "epoch": 2.3089840470193117,
      "grad_norm": 173.0674285888672,
      "learning_rate": 5.499000000000001e-05,
      "loss": 575.3464,
      "step": 5500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 119.65222930908203,
      "learning_rate": 5.999e-05,
      "loss": 550.9334,
      "step": 6000
    },
    {
      "epoch": 2.72879932829555,
      "grad_norm": 163.03219604492188,
      "learning_rate": 6.499000000000001e-05,
      "loss": 532.3316,
      "step": 6500
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 103.94256591796875,
      "learning_rate": 6.999e-05,
      "loss": 516.1778,
      "step": 7000
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 137.8463134765625,
      "learning_rate": 7.499e-05,
      "loss": 502.3519,
      "step": 7500
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 93.18783569335938,
      "learning_rate": 7.999000000000001e-05,
      "loss": 493.4963,
      "step": 8000
    },
    {
      "epoch": 3.568429890848027,
      "grad_norm": 81.264404296875,
      "learning_rate": 8.499e-05,
      "loss": 486.5324,
      "step": 8500
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 96.80731201171875,
      "learning_rate": 8.999000000000001e-05,
      "loss": 477.9452,
      "step": 9000
    },
    {
      "epoch": 3.988245172124265,
      "grad_norm": 99.205322265625,
      "learning_rate": 9.499e-05,
      "loss": 463.8162,
      "step": 9500
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 118.66422271728516,
      "learning_rate": 9.999000000000001e-05,
      "loss": 447.3158,
      "step": 10000
    },
    {
      "epoch": 4.198152812762385,
      "eval_accuracy": 0.16583655638721168,
      "eval_loss": 428.842041015625,
      "eval_runtime": 3.6172,
      "eval_samples_per_second": 132.423,
      "eval_steps_per_second": 1.382,
      "step": 10000
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 144.82168579101562,
      "learning_rate": 9.899293642785066e-05,
      "loss": 430.4443,
      "step": 10500
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 166.46974182128906,
      "learning_rate": 9.798385469223008e-05,
      "loss": 416.3207,
      "step": 11000
    },
    {
      "epoch": 4.827875734676742,
      "grad_norm": 232.3704071044922,
      "learning_rate": 9.69747729566095e-05,
      "loss": 403.2591,
      "step": 11500
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 254.5363311767578,
      "learning_rate": 9.59656912209889e-05,
      "loss": 390.4413,
      "step": 12000
    },
    {
      "epoch": 5.247691015952981,
      "grad_norm": 270.5792236328125,
      "learning_rate": 9.495660948536832e-05,
      "loss": 377.3261,
      "step": 12500
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 244.77581787109375,
      "learning_rate": 9.394752774974774e-05,
      "loss": 365.2654,
      "step": 13000
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 267.1288757324219,
      "learning_rate": 9.293844601412714e-05,
      "loss": 352.7486,
      "step": 13500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 315.1538391113281,
      "learning_rate": 9.192936427850656e-05,
      "loss": 338.1035,
      "step": 14000
    },
    {
      "epoch": 6.087321578505458,
      "grad_norm": 344.386962890625,
      "learning_rate": 9.092028254288598e-05,
      "loss": 323.3387,
      "step": 14500
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 345.59320068359375,
      "learning_rate": 8.991120080726539e-05,
      "loss": 308.7323,
      "step": 15000
    },
    {
      "epoch": 6.507136859781696,
      "grad_norm": 373.200439453125,
      "learning_rate": 8.890211907164481e-05,
      "loss": 293.0822,
      "step": 15500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 387.7672119140625,
      "learning_rate": 8.789303733602423e-05,
      "loss": 276.3089,
      "step": 16000
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 380.2518615722656,
      "learning_rate": 8.688395560040363e-05,
      "loss": 257.2038,
      "step": 16500
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 306.2325134277344,
      "learning_rate": 8.587487386478305e-05,
      "loss": 221.4098,
      "step": 17000
    },
    {
      "epoch": 7.346767422334173,
      "grad_norm": 244.85455322265625,
      "learning_rate": 8.486579212916247e-05,
      "loss": 200.0532,
      "step": 17500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 243.97576904296875,
      "learning_rate": 8.385671039354188e-05,
      "loss": 186.9741,
      "step": 18000
    },
    {
      "epoch": 7.766582703610411,
      "grad_norm": 201.4752655029297,
      "learning_rate": 8.28476286579213e-05,
      "loss": 176.1845,
      "step": 18500
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 180.40164184570312,
      "learning_rate": 8.183854692230071e-05,
      "loss": 167.2202,
      "step": 19000
    },
    {
      "epoch": 8.18639798488665,
      "grad_norm": 169.40943908691406,
      "learning_rate": 8.082946518668012e-05,
      "loss": 158.6278,
      "step": 19500
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 142.1228790283203,
      "learning_rate": 7.982038345105954e-05,
      "loss": 151.7919,
      "step": 20000
    },
    {
      "epoch": 8.39630562552477,
      "eval_accuracy": 0.48157916888801594,
      "eval_loss": 135.78590393066406,
      "eval_runtime": 3.5122,
      "eval_samples_per_second": 136.38,
      "eval_steps_per_second": 1.424,
      "step": 20000
    },
    {
      "epoch": 8.606213266162888,
      "grad_norm": 155.93728637695312,
      "learning_rate": 7.881130171543896e-05,
      "loss": 145.7355,
      "step": 20500
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 139.52723693847656,
      "learning_rate": 7.780221997981836e-05,
      "loss": 140.6628,
      "step": 21000
    },
    {
      "epoch": 9.026028547439127,
      "grad_norm": 146.34320068359375,
      "learning_rate": 7.679313824419778e-05,
      "loss": 135.8976,
      "step": 21500
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 129.9792938232422,
      "learning_rate": 7.57840565085772e-05,
      "loss": 131.474,
      "step": 22000
    },
    {
      "epoch": 9.445843828715365,
      "grad_norm": 122.61577606201172,
      "learning_rate": 7.477497477295662e-05,
      "loss": 127.8668,
      "step": 22500
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 130.75668334960938,
      "learning_rate": 7.376589303733603e-05,
      "loss": 124.6147,
      "step": 23000
    },
    {
      "epoch": 9.865659109991604,
      "grad_norm": 130.1267547607422,
      "learning_rate": 7.275681130171544e-05,
      "loss": 121.6354,
      "step": 23500
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 138.63009643554688,
      "learning_rate": 7.174772956609486e-05,
      "loss": 118.5567,
      "step": 24000
    },
    {
      "epoch": 10.285474391267842,
      "grad_norm": 128.6032257080078,
      "learning_rate": 7.073864783047427e-05,
      "loss": 115.905,
      "step": 24500
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 117.67427825927734,
      "learning_rate": 6.972956609485369e-05,
      "loss": 113.5025,
      "step": 25000
    },
    {
      "epoch": 10.705289672544081,
      "grad_norm": 132.61322021484375,
      "learning_rate": 6.872048435923311e-05,
      "loss": 111.2961,
      "step": 25500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 124.02651977539062,
      "learning_rate": 6.771140262361251e-05,
      "loss": 109.3261,
      "step": 26000
    },
    {
      "epoch": 11.125104953820319,
      "grad_norm": 116.80554962158203,
      "learning_rate": 6.670232088799193e-05,
      "loss": 106.9513,
      "step": 26500
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 120.98764038085938,
      "learning_rate": 6.569323915237135e-05,
      "loss": 105.1095,
      "step": 27000
    },
    {
      "epoch": 11.544920235096557,
      "grad_norm": 109.36351776123047,
      "learning_rate": 6.468415741675076e-05,
      "loss": 103.5172,
      "step": 27500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 115.52029418945312,
      "learning_rate": 6.367507568113018e-05,
      "loss": 101.8046,
      "step": 28000
    },
    {
      "epoch": 11.964735516372796,
      "grad_norm": 110.40377044677734,
      "learning_rate": 6.26659939455096e-05,
      "loss": 100.335,
      "step": 28500
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 106.47163391113281,
      "learning_rate": 6.1656912209889e-05,
      "loss": 98.6005,
      "step": 29000
    },
    {
      "epoch": 12.384550797649034,
      "grad_norm": 114.08397674560547,
      "learning_rate": 6.064783047426842e-05,
      "loss": 97.2174,
      "step": 29500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 108.48992156982422,
      "learning_rate": 5.963874873864783e-05,
      "loss": 95.9257,
      "step": 30000
    },
    {
      "epoch": 12.594458438287154,
      "eval_accuracy": 0.5308223969555362,
      "eval_loss": 84.8065414428711,
      "eval_runtime": 3.6726,
      "eval_samples_per_second": 130.426,
      "eval_steps_per_second": 1.361,
      "step": 30000
    },
    {
      "epoch": 12.804366078925273,
      "grad_norm": 116.85979461669922,
      "learning_rate": 5.8629667003027243e-05,
      "loss": 94.7173,
      "step": 30500
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 101.37419891357422,
      "learning_rate": 5.762058526740667e-05,
      "loss": 93.6031,
      "step": 31000
    },
    {
      "epoch": 13.224181360201511,
      "grad_norm": 101.53105926513672,
      "learning_rate": 5.661150353178608e-05,
      "loss": 92.2381,
      "step": 31500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 131.7721405029297,
      "learning_rate": 5.5602421796165494e-05,
      "loss": 91.2291,
      "step": 32000
    },
    {
      "epoch": 13.64399664147775,
      "grad_norm": 147.26918029785156,
      "learning_rate": 5.4593340060544906e-05,
      "loss": 90.265,
      "step": 32500
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 107.17750549316406,
      "learning_rate": 5.358425832492432e-05,
      "loss": 89.2605,
      "step": 33000
    },
    {
      "epoch": 14.063811922753988,
      "grad_norm": 135.6392364501953,
      "learning_rate": 5.257517658930373e-05,
      "loss": 88.1679,
      "step": 33500
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 131.0419158935547,
      "learning_rate": 5.1566094853683156e-05,
      "loss": 87.1749,
      "step": 34000
    },
    {
      "epoch": 14.483627204030226,
      "grad_norm": 110.07307434082031,
      "learning_rate": 5.055701311806257e-05,
      "loss": 86.2937,
      "step": 34500
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 107.72781372070312,
      "learning_rate": 4.954793138244198e-05,
      "loss": 85.5873,
      "step": 35000
    },
    {
      "epoch": 14.903442485306465,
      "grad_norm": 91.38671112060547,
      "learning_rate": 4.853884964682139e-05,
      "loss": 84.7775,
      "step": 35500
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 107.55229187011719,
      "learning_rate": 4.752976791120081e-05,
      "loss": 83.9663,
      "step": 36000
    },
    {
      "epoch": 15.323257766582703,
      "grad_norm": 93.78266906738281,
      "learning_rate": 4.6520686175580225e-05,
      "loss": 83.2307,
      "step": 36500
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 94.07606506347656,
      "learning_rate": 4.551160443995964e-05,
      "loss": 82.4276,
      "step": 37000
    },
    {
      "epoch": 15.743073047858942,
      "grad_norm": 89.64346313476562,
      "learning_rate": 4.4502522704339056e-05,
      "loss": 81.7749,
      "step": 37500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 90.450927734375,
      "learning_rate": 4.349344096871847e-05,
      "loss": 81.2509,
      "step": 38000
    },
    {
      "epoch": 16.162888329135182,
      "grad_norm": 104.07128143310547,
      "learning_rate": 4.248435923309788e-05,
      "loss": 80.4306,
      "step": 38500
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 97.70896911621094,
      "learning_rate": 4.14752774974773e-05,
      "loss": 79.7886,
      "step": 39000
    },
    {
      "epoch": 16.582703610411418,
      "grad_norm": 107.2480697631836,
      "learning_rate": 4.046619576185671e-05,
      "loss": 79.2786,
      "step": 39500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 94.23933410644531,
      "learning_rate": 3.9457114026236124e-05,
      "loss": 78.7736,
      "step": 40000
    },
    {
      "epoch": 16.79261125104954,
      "eval_accuracy": 0.5467926432656723,
      "eval_loss": 70.87545013427734,
      "eval_runtime": 2.7701,
      "eval_samples_per_second": 172.917,
      "eval_steps_per_second": 1.805,
      "step": 40000
    },
    {
      "epoch": 17.002518891687657,
      "grad_norm": 95.26321411132812,
      "learning_rate": 3.844803229061554e-05,
      "loss": 78.2565,
      "step": 40500
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 97.84280395507812,
      "learning_rate": 3.7438950554994956e-05,
      "loss": 77.6301,
      "step": 41000
    },
    {
      "epoch": 17.422334172963897,
      "grad_norm": 118.63774871826172,
      "learning_rate": 3.642986881937437e-05,
      "loss": 77.0761,
      "step": 41500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 120.7638168334961,
      "learning_rate": 3.542078708375379e-05,
      "loss": 76.71,
      "step": 42000
    },
    {
      "epoch": 17.842149454240133,
      "grad_norm": 96.10567474365234,
      "learning_rate": 3.44117053481332e-05,
      "loss": 76.231,
      "step": 42500
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 109.37013244628906,
      "learning_rate": 3.340262361251261e-05,
      "loss": 75.7584,
      "step": 43000
    },
    {
      "epoch": 18.261964735516372,
      "grad_norm": 81.04769134521484,
      "learning_rate": 3.239354187689203e-05,
      "loss": 75.1399,
      "step": 43500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 107.45811462402344,
      "learning_rate": 3.138446014127144e-05,
      "loss": 74.9177,
      "step": 44000
    },
    {
      "epoch": 18.68178001679261,
      "grad_norm": 91.4131088256836,
      "learning_rate": 3.0375378405650862e-05,
      "loss": 74.4926,
      "step": 44500
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 103.88982391357422,
      "learning_rate": 2.9366296670030274e-05,
      "loss": 74.1645,
      "step": 45000
    },
    {
      "epoch": 19.10159529806885,
      "grad_norm": 101.31675720214844,
      "learning_rate": 2.8357214934409686e-05,
      "loss": 73.6817,
      "step": 45500
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 97.94842529296875,
      "learning_rate": 2.7348133198789106e-05,
      "loss": 73.2026,
      "step": 46000
    },
    {
      "epoch": 19.521410579345087,
      "grad_norm": 85.70989990234375,
      "learning_rate": 2.6339051463168518e-05,
      "loss": 72.9578,
      "step": 46500
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 93.49505615234375,
      "learning_rate": 2.532996972754793e-05,
      "loss": 72.7161,
      "step": 47000
    },
    {
      "epoch": 19.941225860621326,
      "grad_norm": 89.21786499023438,
      "learning_rate": 2.4320887991927346e-05,
      "loss": 72.3488,
      "step": 47500
    },
    {
      "epoch": 20.151133501259444,
      "grad_norm": 86.77904510498047,
      "learning_rate": 2.331180625630676e-05,
      "loss": 71.9474,
      "step": 48000
    },
    {
      "epoch": 20.361041141897566,
      "grad_norm": 104.81689453125,
      "learning_rate": 2.2302724520686177e-05,
      "loss": 71.6556,
      "step": 48500
    },
    {
      "epoch": 20.570948782535684,
      "grad_norm": 85.92950439453125,
      "learning_rate": 2.129364278506559e-05,
      "loss": 71.3194,
      "step": 49000
    },
    {
      "epoch": 20.7808564231738,
      "grad_norm": 86.256103515625,
      "learning_rate": 2.0284561049445005e-05,
      "loss": 71.1533,
      "step": 49500
    },
    {
      "epoch": 20.990764063811923,
      "grad_norm": 92.44305419921875,
      "learning_rate": 1.927547931382442e-05,
      "loss": 70.9704,
      "step": 50000
    },
    {
      "epoch": 20.990764063811923,
      "eval_accuracy": 0.5510415790506631,
      "eval_loss": 63.090843200683594,
      "eval_runtime": 2.8803,
      "eval_samples_per_second": 166.303,
      "eval_steps_per_second": 1.736,
      "step": 50000
    },
    {
      "epoch": 21.20067170445004,
      "grad_norm": 80.83171844482422,
      "learning_rate": 1.8266397578203836e-05,
      "loss": 70.5263,
      "step": 50500
    },
    {
      "epoch": 21.410579345088163,
      "grad_norm": 86.68183898925781,
      "learning_rate": 1.725731584258325e-05,
      "loss": 70.3711,
      "step": 51000
    },
    {
      "epoch": 21.62048698572628,
      "grad_norm": 75.10340118408203,
      "learning_rate": 1.6248234106962664e-05,
      "loss": 70.0563,
      "step": 51500
    },
    {
      "epoch": 21.8303946263644,
      "grad_norm": 80.75882720947266,
      "learning_rate": 1.523915237134208e-05,
      "loss": 69.9151,
      "step": 52000
    },
    {
      "epoch": 22.04030226700252,
      "grad_norm": 80.31729888916016,
      "learning_rate": 1.4230070635721492e-05,
      "loss": 69.7056,
      "step": 52500
    },
    {
      "epoch": 22.250209907640638,
      "grad_norm": 79.79283905029297,
      "learning_rate": 1.3220988900100908e-05,
      "loss": 69.3878,
      "step": 53000
    },
    {
      "epoch": 22.460117548278756,
      "grad_norm": 88.36482238769531,
      "learning_rate": 1.2211907164480324e-05,
      "loss": 69.2763,
      "step": 53500
    },
    {
      "epoch": 22.670025188916878,
      "grad_norm": 82.19640350341797,
      "learning_rate": 1.1202825428859738e-05,
      "loss": 69.1046,
      "step": 54000
    },
    {
      "epoch": 22.879932829554996,
      "grad_norm": 77.26073455810547,
      "learning_rate": 1.0193743693239152e-05,
      "loss": 68.8851,
      "step": 54500
    },
    {
      "epoch": 23.089840470193113,
      "grad_norm": 89.86582946777344,
      "learning_rate": 9.184661957618567e-06,
      "loss": 68.7541,
      "step": 55000
    },
    {
      "epoch": 23.299748110831235,
      "grad_norm": 70.37703704833984,
      "learning_rate": 8.175580221997981e-06,
      "loss": 68.5219,
      "step": 55500
    },
    {
      "epoch": 23.509655751469353,
      "grad_norm": 71.59619140625,
      "learning_rate": 7.166498486377397e-06,
      "loss": 68.3648,
      "step": 56000
    },
    {
      "epoch": 23.719563392107474,
      "grad_norm": 69.95408630371094,
      "learning_rate": 6.157416750756812e-06,
      "loss": 68.3101,
      "step": 56500
    },
    {
      "epoch": 23.929471032745592,
      "grad_norm": 71.7118148803711,
      "learning_rate": 5.148335015136227e-06,
      "loss": 68.1746,
      "step": 57000
    },
    {
      "epoch": 24.13937867338371,
      "grad_norm": 67.84393310546875,
      "learning_rate": 4.139253279515641e-06,
      "loss": 67.9893,
      "step": 57500
    },
    {
      "epoch": 24.349286314021832,
      "grad_norm": 70.03734588623047,
      "learning_rate": 3.130171543895056e-06,
      "loss": 67.9278,
      "step": 58000
    },
    {
      "epoch": 24.55919395465995,
      "grad_norm": 67.93866729736328,
      "learning_rate": 2.1210898082744702e-06,
      "loss": 67.8202,
      "step": 58500
    },
    {
      "epoch": 24.769101595298068,
      "grad_norm": 61.62685775756836,
      "learning_rate": 1.112008072653885e-06,
      "loss": 67.7249,
      "step": 59000
    },
    {
      "epoch": 24.97900923593619,
      "grad_norm": 63.67091751098633,
      "learning_rate": 1.0292633703329971e-07,
      "loss": 67.6809,
      "step": 59500
    },
    {
      "epoch": 25.0,
      "step": 59550,
      "total_flos": 7.577162542158336e+17,
      "train_loss": 301.8068682845954,
      "train_runtime": 49685.0337,
      "train_samples_per_second": 115.044,
      "train_steps_per_second": 1.199
    }
  ],
  "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": 7.577162542158336e+17,
  "train_batch_size": 96,
  "trial_name": null,
  "trial_params": null
}