{
  "best_metric": 0.29125073552131653,
  "best_model_checkpoint": "deit-cvc-drop-aug/checkpoint-1800",
  "epoch": 14.979591836734693,
  "eval_steps": 100,
  "global_step": 5505,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.27,
      "learning_rate": 9.991860321746833e-05,
      "loss": 0.5453,
      "step": 100
    },
    {
      "epoch": 0.27,
      "eval_accuracy": 0.7695840775708089,
      "eval_f1": 0.7750933997509339,
      "eval_loss": 0.48243534564971924,
      "eval_precision": 0.7776111944027986,
      "eval_recall": 0.7725918570009931,
      "eval_runtime": 15.4841,
      "eval_samples_per_second": 253.098,
      "eval_steps_per_second": 15.823,
      "step": 100
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.967467788732156e-05,
      "loss": 0.4324,
      "step": 200
    },
    {
      "epoch": 0.54,
      "eval_accuracy": 0.7685634090329165,
      "eval_f1": 0.7637405574368324,
      "eval_loss": 0.4796455502510071,
      "eval_precision": 0.8032876712328767,
      "eval_recall": 0.7279046673286991,
      "eval_runtime": 15.2008,
      "eval_samples_per_second": 257.815,
      "eval_steps_per_second": 16.118,
      "step": 200
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.926901819904178e-05,
      "loss": 0.4042,
      "step": 300
    },
    {
      "epoch": 0.82,
      "eval_accuracy": 0.8321000255167135,
      "eval_f1": 0.8546819787985865,
      "eval_loss": 0.3790486454963684,
      "eval_precision": 0.7696897374701671,
      "eval_recall": 0.9607745779543198,
      "eval_runtime": 15.77,
      "eval_samples_per_second": 248.509,
      "eval_steps_per_second": 15.536,
      "step": 300
    },
    {
      "epoch": 1.09,
      "learning_rate": 9.870294492836612e-05,
      "loss": 0.3849,
      "step": 400
    },
    {
      "epoch": 1.09,
      "eval_accuracy": 0.8101556519520285,
      "eval_f1": 0.8161146811665843,
      "eval_loss": 0.4099680185317993,
      "eval_precision": 0.8125,
      "eval_recall": 0.8197616683217478,
      "eval_runtime": 13.9661,
      "eval_samples_per_second": 280.608,
      "eval_steps_per_second": 17.542,
      "step": 400
    },
    {
      "epoch": 1.36,
      "learning_rate": 9.797830113701101e-05,
      "loss": 0.3621,
      "step": 500
    },
    {
      "epoch": 1.36,
      "eval_accuracy": 0.8387343710130135,
      "eval_f1": 0.8510838831291235,
      "eval_loss": 0.36887437105178833,
      "eval_precision": 0.809865470852018,
      "eval_recall": 0.8967229394240318,
      "eval_runtime": 14.7142,
      "eval_samples_per_second": 266.341,
      "eval_steps_per_second": 16.651,
      "step": 500
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.709744617190038e-05,
      "loss": 0.3457,
      "step": 600
    },
    {
      "epoch": 1.63,
      "eval_accuracy": 0.8458790507782598,
      "eval_f1": 0.8642086330935251,
      "eval_loss": 0.3312939703464508,
      "eval_precision": 0.7896466721446179,
      "eval_recall": 0.9543197616683218,
      "eval_runtime": 15.9456,
      "eval_samples_per_second": 245.773,
      "eval_steps_per_second": 15.365,
      "step": 600
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.606324798343567e-05,
      "loss": 0.3443,
      "step": 700
    },
    {
      "epoch": 1.9,
      "eval_accuracy": 0.840520540954325,
      "eval_f1": 0.8599596683844947,
      "eval_loss": 0.3424055874347687,
      "eval_precision": 0.7835851367905268,
      "eval_recall": 0.9528301886792453,
      "eval_runtime": 15.7574,
      "eval_samples_per_second": 248.708,
      "eval_steps_per_second": 15.548,
      "step": 700
    },
    {
      "epoch": 2.18,
      "learning_rate": 9.487907378781853e-05,
      "loss": 0.3287,
      "step": 800
    },
    {
      "epoch": 2.18,
      "eval_accuracy": 0.8453687165093136,
      "eval_f1": 0.8560570071258907,
      "eval_loss": 0.33078569173812866,
      "eval_precision": 0.8205828779599271,
      "eval_recall": 0.8947368421052632,
      "eval_runtime": 14.597,
      "eval_samples_per_second": 268.479,
      "eval_steps_per_second": 16.784,
      "step": 800
    },
    {
      "epoch": 2.45,
      "learning_rate": 9.35487791038282e-05,
      "loss": 0.3224,
      "step": 900
    },
    {
      "epoch": 2.45,
      "eval_accuracy": 0.7655014034192396,
      "eval_f1": 0.7437970448843045,
      "eval_loss": 0.454572468996048,
      "eval_precision": 0.8480610298792117,
      "eval_recall": 0.6623634558093346,
      "eval_runtime": 15.6304,
      "eval_samples_per_second": 250.73,
      "eval_steps_per_second": 15.675,
      "step": 900
    },
    {
      "epoch": 2.72,
      "learning_rate": 9.207669519974851e-05,
      "loss": 0.3096,
      "step": 1000
    },
    {
      "epoch": 2.72,
      "eval_accuracy": 0.8438377137024751,
      "eval_f1": 0.8521024649589173,
      "eval_loss": 0.34022626280784607,
      "eval_precision": 0.8300376647834274,
      "eval_recall": 0.8753723932472691,
      "eval_runtime": 15.4283,
      "eval_samples_per_second": 254.014,
      "eval_steps_per_second": 15.88,
      "step": 1000
    },
    {
      "epoch": 2.99,
      "learning_rate": 9.046761499131578e-05,
      "loss": 0.3095,
      "step": 1100
    },
    {
      "epoch": 2.99,
      "eval_accuracy": 0.8384792038785405,
      "eval_f1": 0.8524131499183958,
      "eval_loss": 0.36910393834114075,
      "eval_precision": 0.8035164835164835,
      "eval_recall": 0.9076464746772592,
      "eval_runtime": 15.4854,
      "eval_samples_per_second": 253.077,
      "eval_steps_per_second": 15.821,
      "step": 1100
    },
    {
      "epoch": 3.27,
      "learning_rate": 8.872677743660209e-05,
      "loss": 0.2901,
      "step": 1200
    },
    {
      "epoch": 3.27,
      "eval_accuracy": 0.8328655269201327,
      "eval_f1": 0.8467119120056167,
      "eval_loss": 0.36432939767837524,
      "eval_precision": 0.8007968127490039,
      "eval_recall": 0.8982125124131083,
      "eval_runtime": 14.6332,
      "eval_samples_per_second": 267.816,
      "eval_steps_per_second": 16.743,
      "step": 1200
    },
    {
      "epoch": 3.54,
      "learning_rate": 8.685985047864204e-05,
      "loss": 0.2939,
      "step": 1300
    },
    {
      "epoch": 3.54,
      "eval_accuracy": 0.8601684103087522,
      "eval_f1": 0.8760180995475113,
      "eval_loss": 0.30208107829093933,
      "eval_precision": 0.8046550290939318,
      "eval_recall": 0.9612711022840119,
      "eval_runtime": 14.6512,
      "eval_samples_per_second": 267.487,
      "eval_steps_per_second": 16.722,
      "step": 1300
    },
    {
      "epoch": 3.81,
      "learning_rate": 8.487291259133956e-05,
      "loss": 0.2946,
      "step": 1400
    },
    {
      "epoch": 3.81,
      "eval_accuracy": 0.8300586884409288,
      "eval_f1": 0.8342458934793429,
      "eval_loss": 0.36171799898147583,
      "eval_precision": 0.8363273453093812,
      "eval_recall": 0.8321747765640516,
      "eval_runtime": 15.3842,
      "eval_samples_per_second": 254.741,
      "eval_steps_per_second": 15.925,
      "step": 1400
    },
    {
      "epoch": 4.08,
      "learning_rate": 8.277243298873936e-05,
      "loss": 0.2856,
      "step": 1500
    },
    {
      "epoch": 4.08,
      "eval_accuracy": 0.8127073232967594,
      "eval_f1": 0.8116016427104723,
      "eval_loss": 0.48843976855278015,
      "eval_precision": 0.8400637619553666,
      "eval_recall": 0.7850049652432969,
      "eval_runtime": 14.4792,
      "eval_samples_per_second": 270.663,
      "eval_steps_per_second": 16.921,
      "step": 1500
    },
    {
      "epoch": 4.35,
      "learning_rate": 8.056525056209841e-05,
      "loss": 0.2683,
      "step": 1600
    },
    {
      "epoch": 4.35,
      "eval_accuracy": 0.834141362592498,
      "eval_f1": 0.8380667663178873,
      "eval_loss": 0.3540255129337311,
      "eval_precision": 0.841,
      "eval_recall": 0.8351539225422046,
      "eval_runtime": 15.1561,
      "eval_samples_per_second": 258.575,
      "eval_steps_per_second": 16.165,
      "step": 1600
    },
    {
      "epoch": 4.63,
      "learning_rate": 7.82585516133363e-05,
      "loss": 0.2724,
      "step": 1700
    },
    {
      "epoch": 4.63,
      "eval_accuracy": 0.8581270732329676,
      "eval_f1": 0.8664745437079732,
      "eval_loss": 0.30780917406082153,
      "eval_precision": 0.8390697674418605,
      "eval_recall": 0.8957298907646475,
      "eval_runtime": 14.446,
      "eval_samples_per_second": 271.286,
      "eval_steps_per_second": 16.96,
      "step": 1700
    },
    {
      "epoch": 4.9,
      "learning_rate": 7.585984645736096e-05,
      "loss": 0.2685,
      "step": 1800
    },
    {
      "epoch": 4.9,
      "eval_accuracy": 0.862720081653483,
      "eval_f1": 0.8703614457831326,
      "eval_loss": 0.29125073552131653,
      "eval_precision": 0.8455056179775281,
      "eval_recall": 0.8967229394240318,
      "eval_runtime": 15.5954,
      "eval_samples_per_second": 251.291,
      "eval_steps_per_second": 15.71,
      "step": 1800
    },
    {
      "epoch": 5.17,
      "learning_rate": 7.33769449694499e-05,
      "loss": 0.2449,
      "step": 1900
    },
    {
      "epoch": 5.17,
      "eval_accuracy": 0.8443480479714213,
      "eval_f1": 0.8490099009900989,
      "eval_loss": 0.38658004999160767,
      "eval_precision": 0.8464955577492597,
      "eval_recall": 0.8515392254220456,
      "eval_runtime": 16.0186,
      "eval_samples_per_second": 244.653,
      "eval_steps_per_second": 15.295,
      "step": 1900
    },
    {
      "epoch": 5.44,
      "learning_rate": 7.081793115730153e-05,
      "loss": 0.2468,
      "step": 2000
    },
    {
      "epoch": 5.44,
      "eval_accuracy": 0.8588925746363868,
      "eval_f1": 0.8670353450348642,
      "eval_loss": 0.30722111463546753,
      "eval_precision": 0.8405594405594405,
      "eval_recall": 0.8952333664349553,
      "eval_runtime": 15.614,
      "eval_samples_per_second": 250.993,
      "eval_steps_per_second": 15.691,
      "step": 2000
    },
    {
      "epoch": 5.71,
      "learning_rate": 6.819113684054634e-05,
      "loss": 0.2557,
      "step": 2100
    },
    {
      "epoch": 5.71,
      "eval_accuracy": 0.8257208471548865,
      "eval_f1": 0.8232858990944373,
      "eval_loss": 0.3735339641571045,
      "eval_precision": 0.8595353862776878,
      "eval_recall": 0.7899702085402185,
      "eval_runtime": 15.378,
      "eval_samples_per_second": 254.845,
      "eval_steps_per_second": 15.932,
      "step": 2100
    },
    {
      "epoch": 5.99,
      "learning_rate": 6.550511452341459e-05,
      "loss": 0.25,
      "step": 2200
    },
    {
      "epoch": 5.99,
      "eval_accuracy": 0.8553202347537637,
      "eval_f1": 0.8561278863232682,
      "eval_loss": 0.3116500973701477,
      "eval_precision": 0.8754540736896731,
      "eval_recall": 0.8376365441906654,
      "eval_runtime": 15.151,
      "eval_samples_per_second": 258.663,
      "eval_steps_per_second": 16.171,
      "step": 2200
    },
    {
      "epoch": 6.26,
      "learning_rate": 6.276860954888322e-05,
      "loss": 0.2256,
      "step": 2300
    },
    {
      "epoch": 6.26,
      "eval_accuracy": 0.8573615718295483,
      "eval_f1": 0.8652687394552905,
      "eval_loss": 0.3264216184616089,
      "eval_precision": 0.8407494145199064,
      "eval_recall": 0.891261171797418,
      "eval_runtime": 13.8247,
      "eval_samples_per_second": 283.478,
      "eval_steps_per_second": 17.722,
      "step": 2300
    },
    {
      "epoch": 6.53,
      "learning_rate": 5.999053162496453e-05,
      "loss": 0.234,
      "step": 2400
    },
    {
      "epoch": 6.53,
      "eval_accuracy": 0.8295483541719827,
      "eval_f1": 0.8203335126412049,
      "eval_loss": 0.3617473244667053,
      "eval_precision": 0.8949530516431925,
      "eval_recall": 0.7571996027805362,
      "eval_runtime": 14.3944,
      "eval_samples_per_second": 272.259,
      "eval_steps_per_second": 17.02,
      "step": 2400
    },
    {
      "epoch": 6.8,
      "learning_rate": 5.717992581584373e-05,
      "loss": 0.2259,
      "step": 2500
    },
    {
      "epoch": 6.8,
      "eval_accuracy": 0.8609339117121715,
      "eval_f1": 0.8679428156045553,
      "eval_loss": 0.32836005091667175,
      "eval_precision": 0.8476100331282537,
      "eval_recall": 0.8892750744786495,
      "eval_runtime": 15.9578,
      "eval_samples_per_second": 245.585,
      "eval_steps_per_second": 15.353,
      "step": 2500
    },
    {
      "epoch": 7.07,
      "learning_rate": 5.434594309231389e-05,
      "loss": 0.2261,
      "step": 2600
    },
    {
      "epoch": 7.07,
      "eval_accuracy": 0.8443480479714213,
      "eval_f1": 0.8375066595631327,
      "eval_loss": 0.34862977266311646,
      "eval_precision": 0.903448275862069,
      "eval_recall": 0.7805362462760675,
      "eval_runtime": 14.7094,
      "eval_samples_per_second": 266.428,
      "eval_steps_per_second": 16.656,
      "step": 2600
    },
    {
      "epoch": 7.35,
      "learning_rate": 5.1497810537392844e-05,
      "loss": 0.2087,
      "step": 2700
    },
    {
      "epoch": 7.35,
      "eval_accuracy": 0.836948201071702,
      "eval_f1": 0.8365310821181888,
      "eval_loss": 0.3970935046672821,
      "eval_precision": 0.862796833773087,
      "eval_recall": 0.8118172790466733,
      "eval_runtime": 15.0606,
      "eval_samples_per_second": 260.215,
      "eval_steps_per_second": 16.268,
      "step": 2700
    },
    {
      "epoch": 7.62,
      "learning_rate": 4.8644801304128374e-05,
      "loss": 0.2035,
      "step": 2800
    },
    {
      "epoch": 7.62,
      "eval_accuracy": 0.878285276856341,
      "eval_f1": 0.8830595734248591,
      "eval_loss": 0.31056222319602966,
      "eval_precision": 0.8721549636803874,
      "eval_recall": 0.894240317775571,
      "eval_runtime": 15.0478,
      "eval_samples_per_second": 260.437,
      "eval_steps_per_second": 16.281,
      "step": 2800
    },
    {
      "epoch": 7.89,
      "learning_rate": 4.57962044234053e-05,
      "loss": 0.2116,
      "step": 2900
    },
    {
      "epoch": 7.89,
      "eval_accuracy": 0.8451135493748405,
      "eval_f1": 0.8428682371214082,
      "eval_loss": 0.3734387159347534,
      "eval_precision": 0.8804759329367225,
      "eval_recall": 0.8083416087388282,
      "eval_runtime": 15.0966,
      "eval_samples_per_second": 259.594,
      "eval_steps_per_second": 16.229,
      "step": 2900
    },
    {
      "epoch": 8.16,
      "learning_rate": 4.2961294560056445e-05,
      "loss": 0.1956,
      "step": 3000
    },
    {
      "epoch": 8.16,
      "eval_accuracy": 0.85914774177086,
      "eval_f1": 0.863298662704309,
      "eval_loss": 0.3442569077014923,
      "eval_precision": 0.8611660079051383,
      "eval_recall": 0.865441906653426,
      "eval_runtime": 15.4752,
      "eval_samples_per_second": 253.245,
      "eval_steps_per_second": 15.832,
      "step": 3000
    },
    {
      "epoch": 8.44,
      "learning_rate": 4.01493018157476e-05,
      "loss": 0.1826,
      "step": 3100
    },
    {
      "epoch": 8.44,
      "eval_accuracy": 0.8423067108956367,
      "eval_f1": 0.8373684210526314,
      "eval_loss": 0.3795164227485657,
      "eval_precision": 0.8908174692049272,
      "eval_recall": 0.7899702085402185,
      "eval_runtime": 15.3206,
      "eval_samples_per_second": 255.8,
      "eval_steps_per_second": 15.992,
      "step": 3100
    },
    {
      "epoch": 8.71,
      "learning_rate": 3.7369381676954284e-05,
      "loss": 0.1918,
      "step": 3200
    },
    {
      "epoch": 8.71,
      "eval_accuracy": 0.8581270732329676,
      "eval_f1": 0.8569222851260936,
      "eval_loss": 0.33620110154151917,
      "eval_precision": 0.8894230769230769,
      "eval_recall": 0.8267130089374379,
      "eval_runtime": 14.574,
      "eval_samples_per_second": 268.904,
      "eval_steps_per_second": 16.811,
      "step": 3200
    },
    {
      "epoch": 8.98,
      "learning_rate": 3.463058520587625e-05,
      "loss": 0.1886,
      "step": 3300
    },
    {
      "epoch": 8.98,
      "eval_accuracy": 0.8639959173258485,
      "eval_f1": 0.8692666176109884,
      "eval_loss": 0.3259018063545227,
      "eval_precision": 0.8589432864760058,
      "eval_recall": 0.8798411122144985,
      "eval_runtime": 14.5325,
      "eval_samples_per_second": 269.671,
      "eval_steps_per_second": 16.859,
      "step": 3300
    },
    {
      "epoch": 9.25,
      "learning_rate": 3.194182957134365e-05,
      "loss": 0.1716,
      "step": 3400
    },
    {
      "epoch": 9.25,
      "eval_accuracy": 0.846389385047206,
      "eval_f1": 0.8481331987891019,
      "eval_loss": 0.4269343912601471,
      "eval_precision": 0.862051282051282,
      "eval_recall": 0.8346573982125124,
      "eval_runtime": 15.0706,
      "eval_samples_per_second": 260.043,
      "eval_steps_per_second": 16.257,
      "step": 3400
    },
    {
      "epoch": 9.52,
      "learning_rate": 2.9311869015663125e-05,
      "loss": 0.1654,
      "step": 3500
    },
    {
      "epoch": 9.52,
      "eval_accuracy": 0.859658076039806,
      "eval_f1": 0.858974358974359,
      "eval_loss": 0.40663468837738037,
      "eval_precision": 0.8881230116648993,
      "eval_recall": 0.8316782522343595,
      "eval_runtime": 13.6261,
      "eval_samples_per_second": 287.609,
      "eval_steps_per_second": 17.98,
      "step": 3500
    },
    {
      "epoch": 9.8,
      "learning_rate": 2.674926635193134e-05,
      "loss": 0.1625,
      "step": 3600
    },
    {
      "epoch": 9.8,
      "eval_accuracy": 0.8512375606021945,
      "eval_f1": 0.8488462535649469,
      "eval_loss": 0.3926689624786377,
      "eval_precision": 0.8882257189365166,
      "eval_recall": 0.8128103277060575,
      "eval_runtime": 15.0764,
      "eval_samples_per_second": 259.943,
      "eval_steps_per_second": 16.251,
      "step": 3600
    },
    {
      "epoch": 10.07,
      "learning_rate": 2.42623650846177e-05,
      "loss": 0.1659,
      "step": 3700
    },
    {
      "epoch": 10.07,
      "eval_accuracy": 0.8548099004848175,
      "eval_f1": 0.8529335745670716,
      "eval_loss": 0.37967267632484436,
      "eval_precision": 0.889487870619946,
      "eval_recall": 0.8192651439920556,
      "eval_runtime": 14.8402,
      "eval_samples_per_second": 264.08,
      "eval_steps_per_second": 16.509,
      "step": 3700
    },
    {
      "epoch": 10.34,
      "learning_rate": 2.1859262244187556e-05,
      "loss": 0.1519,
      "step": 3800
    },
    {
      "epoch": 10.34,
      "eval_accuracy": 0.8512375606021945,
      "eval_f1": 0.8502440277421012,
      "eval_loss": 0.4088890254497528,
      "eval_precision": 0.8807876530069185,
      "eval_recall": 0.8217477656405164,
      "eval_runtime": 15.4244,
      "eval_samples_per_second": 254.079,
      "eval_steps_per_second": 15.884,
      "step": 3800
    },
    {
      "epoch": 10.61,
      "learning_rate": 1.9547782024213047e-05,
      "loss": 0.1484,
      "step": 3900
    },
    {
      "epoch": 10.61,
      "eval_accuracy": 0.8545547333503445,
      "eval_f1": 0.8533950617283951,
      "eval_loss": 0.3864934742450714,
      "eval_precision": 0.8852721451440768,
      "eval_recall": 0.823733862959285,
      "eval_runtime": 14.9541,
      "eval_samples_per_second": 262.069,
      "eval_steps_per_second": 16.383,
      "step": 3900
    },
    {
      "epoch": 10.88,
      "learning_rate": 1.7335450306805827e-05,
      "loss": 0.1427,
      "step": 4000
    },
    {
      "epoch": 10.88,
      "eval_accuracy": 0.846389385047206,
      "eval_f1": 0.8427377220480669,
      "eval_loss": 0.4346730411052704,
      "eval_precision": 0.8891951488423374,
      "eval_recall": 0.8008937437934459,
      "eval_runtime": 15.9395,
      "eval_samples_per_second": 245.867,
      "eval_steps_per_second": 15.371,
      "step": 4000
    },
    {
      "epoch": 11.16,
      "learning_rate": 1.522947015931348e-05,
      "loss": 0.1375,
      "step": 4100
    },
    {
      "epoch": 11.16,
      "eval_accuracy": 0.8548099004848175,
      "eval_f1": 0.8532370389476398,
      "eval_loss": 0.4687642753124237,
      "eval_precision": 0.8878153515834675,
      "eval_recall": 0.8212512413108243,
      "eval_runtime": 14.957,
      "eval_samples_per_second": 262.017,
      "eval_steps_per_second": 16.38,
      "step": 4100
    },
    {
      "epoch": 11.43,
      "learning_rate": 1.3236698382059287e-05,
      "loss": 0.1276,
      "step": 4200
    },
    {
      "epoch": 11.43,
      "eval_accuracy": 0.846899719316152,
      "eval_f1": 0.8426023084994753,
      "eval_loss": 0.4686568081378937,
      "eval_precision": 0.8932146829810901,
      "eval_recall": 0.7974180734856008,
      "eval_runtime": 13.9054,
      "eval_samples_per_second": 281.833,
      "eval_steps_per_second": 17.619,
      "step": 4200
    },
    {
      "epoch": 11.7,
      "learning_rate": 1.1363623183482775e-05,
      "loss": 0.1275,
      "step": 4300
    },
    {
      "epoch": 11.7,
      "eval_accuracy": 0.8486858892574637,
      "eval_f1": 0.8447237496726892,
      "eval_loss": 0.4493071436882019,
      "eval_precision": 0.8936288088642659,
      "eval_recall": 0.8008937437934459,
      "eval_runtime": 14.5625,
      "eval_samples_per_second": 269.116,
      "eval_steps_per_second": 16.824,
      "step": 4300
    },
    {
      "epoch": 11.97,
      "learning_rate": 9.616343055368083e-06,
      "loss": 0.1349,
      "step": 4400
    },
    {
      "epoch": 11.97,
      "eval_accuracy": 0.8423067108956367,
      "eval_f1": 0.8360742705570292,
      "eval_loss": 0.4618338346481323,
      "eval_precision": 0.89749430523918,
      "eval_recall": 0.7825223435948362,
      "eval_runtime": 14.9603,
      "eval_samples_per_second": 261.96,
      "eval_steps_per_second": 16.377,
      "step": 4400
    },
    {
      "epoch": 12.24,
      "learning_rate": 8.000546916939644e-06,
      "loss": 0.1217,
      "step": 4500
    },
    {
      "epoch": 12.24,
      "eval_accuracy": 0.849706557795356,
      "eval_f1": 0.8450407787424362,
      "eval_loss": 0.4635533094406128,
      "eval_precision": 0.8987129266927812,
      "eval_recall": 0.7974180734856008,
      "eval_runtime": 15.9792,
      "eval_samples_per_second": 245.256,
      "eval_steps_per_second": 15.332,
      "step": 4500
    },
    {
      "epoch": 12.52,
      "learning_rate": 6.521495592473259e-06,
      "loss": 0.1211,
      "step": 4600
    },
    {
      "epoch": 12.52,
      "eval_accuracy": 0.8555754018882368,
      "eval_f1": 0.855316973415133,
      "eval_loss": 0.45266029238700867,
      "eval_precision": 0.8814541622760801,
      "eval_recall": 0.8306852035749752,
      "eval_runtime": 15.4568,
      "eval_samples_per_second": 253.545,
      "eval_steps_per_second": 15.851,
      "step": 4600
    },
    {
      "epoch": 12.79,
      "learning_rate": 5.184004682729348e-06,
      "loss": 0.1164,
      "step": 4700
    },
    {
      "epoch": 12.79,
      "eval_accuracy": 0.8545547333503445,
      "eval_f1": 0.851639770952629,
      "eval_loss": 0.46692270040512085,
      "eval_precision": 0.8949671772428884,
      "eval_recall": 0.8123138033763655,
      "eval_runtime": 16.339,
      "eval_samples_per_second": 239.856,
      "eval_steps_per_second": 14.995,
      "step": 4700
    },
    {
      "epoch": 13.06,
      "learning_rate": 3.992428885976652e-06,
      "loss": 0.1119,
      "step": 4800
    },
    {
      "epoch": 13.06,
      "eval_accuracy": 0.8517478948711406,
      "eval_f1": 0.8495987574424021,
      "eval_loss": 0.461725115776062,
      "eval_precision": 0.8875067604110329,
      "eval_recall": 0.8147964250248262,
      "eval_runtime": 15.7126,
      "eval_samples_per_second": 249.418,
      "eval_steps_per_second": 15.593,
      "step": 4800
    },
    {
      "epoch": 13.33,
      "learning_rate": 2.9506478196551055e-06,
      "loss": 0.11,
      "step": 4900
    },
    {
      "epoch": 13.33,
      "eval_accuracy": 0.8507272263332483,
      "eval_f1": 0.848012470771629,
      "eval_loss": 0.4718102216720581,
      "eval_precision": 0.8893732970027248,
      "eval_recall": 0.8103277060575969,
      "eval_runtime": 15.1226,
      "eval_samples_per_second": 259.149,
      "eval_steps_per_second": 16.201,
      "step": 4900
    },
    {
      "epoch": 13.61,
      "learning_rate": 2.062053388840768e-06,
      "loss": 0.1138,
      "step": 5000
    },
    {
      "epoch": 13.61,
      "eval_accuracy": 0.8479203878540444,
      "eval_f1": 0.8437336130047195,
      "eval_loss": 0.48920777440071106,
      "eval_precision": 0.8938888888888888,
      "eval_recall": 0.7989076464746773,
      "eval_runtime": 15.5429,
      "eval_samples_per_second": 252.14,
      "eval_steps_per_second": 15.763,
      "step": 5000
    },
    {
      "epoch": 13.88,
      "learning_rate": 1.329538742639358e-06,
      "loss": 0.1058,
      "step": 5100
    },
    {
      "epoch": 13.88,
      "eval_accuracy": 0.849961724929829,
      "eval_f1": 0.8474312402698495,
      "eval_loss": 0.47247427701950073,
      "eval_precision": 0.8875,
      "eval_recall": 0.810824230387289,
      "eval_runtime": 16.0011,
      "eval_samples_per_second": 244.92,
      "eval_steps_per_second": 15.311,
      "step": 5100
    },
    {
      "epoch": 14.15,
      "learning_rate": 7.554888544652305e-07,
      "loss": 0.1042,
      "step": 5200
    },
    {
      "epoch": 14.15,
      "eval_accuracy": 0.849706557795356,
      "eval_f1": 0.8464946572843368,
      "eval_loss": 0.47878143191337585,
      "eval_precision": 0.8908392759188152,
      "eval_recall": 0.8063555114200596,
      "eval_runtime": 15.3441,
      "eval_samples_per_second": 255.408,
      "eval_steps_per_second": 15.967,
      "step": 5200
    },
    {
      "epoch": 14.42,
      "learning_rate": 3.4177275687476974e-07,
      "loss": 0.107,
      "step": 5300
    },
    {
      "epoch": 14.42,
      "eval_accuracy": 0.849706557795356,
      "eval_f1": 0.8466545170528509,
      "eval_loss": 0.4759483337402344,
      "eval_precision": 0.8899835796387521,
      "eval_recall": 0.8073485600794439,
      "eval_runtime": 15.151,
      "eval_samples_per_second": 258.663,
      "eval_steps_per_second": 16.171,
      "step": 5300
    },
    {
      "epoch": 14.69,
      "learning_rate": 8.973745623699903e-08,
      "loss": 0.1047,
      "step": 5400
    },
    {
      "epoch": 14.69,
      "eval_accuracy": 0.8489410563919367,
      "eval_f1": 0.8458333333333334,
      "eval_loss": 0.4766782820224762,
      "eval_precision": 0.8893756845564075,
      "eval_recall": 0.8063555114200596,
      "eval_runtime": 15.3016,
      "eval_samples_per_second": 256.118,
      "eval_steps_per_second": 16.011,
      "step": 5400
    },
    {
      "epoch": 14.97,
      "learning_rate": 2.0354705417280351e-10,
      "loss": 0.1085,
      "step": 5500
    },
    {
      "epoch": 14.97,
      "eval_accuracy": 0.8489410563919367,
      "eval_f1": 0.8458333333333334,
      "eval_loss": 0.47687891125679016,
      "eval_precision": 0.8893756845564075,
      "eval_recall": 0.8063555114200596,
      "eval_runtime": 15.647,
      "eval_samples_per_second": 250.463,
      "eval_steps_per_second": 15.658,
      "step": 5500
    },
    {
      "epoch": 14.98,
      "step": 5505,
      "total_flos": 2.046979800964418e+20,
      "train_loss": 0.2219778089443193,
      "train_runtime": 55065.7639,
      "train_samples_per_second": 9.607,
      "train_steps_per_second": 0.1
    },
    {
      "epoch": 14.98,
      "eval_accuracy": 0.725,
      "eval_f1": 0.7441860465116279,
      "eval_loss": 0.5065506100654602,
      "eval_precision": 0.6956521739130435,
      "eval_recall": 0.8,
      "eval_runtime": 1.8887,
      "eval_samples_per_second": 21.179,
      "eval_steps_per_second": 1.588,
      "step": 5505
    }
  ],
  "logging_steps": 100,
  "max_steps": 5505,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 100,
  "total_flos": 2.046979800964418e+20,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}