{
  "best_metric": 0.31098222732543945,
  "best_model_checkpoint": "./seq_clf/0523013301/checkpoint-24528",
  "epoch": 25.0,
  "eval_steps": 500,
  "global_step": 102200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1223091976516634,
      "grad_norm": 3.0948333740234375,
      "learning_rate": 1.990215264187867e-05,
      "loss": 1.4987,
      "step": 500
    },
    {
      "epoch": 0.2446183953033268,
      "grad_norm": 4.519032955169678,
      "learning_rate": 1.9804305283757338e-05,
      "loss": 0.8512,
      "step": 1000
    },
    {
      "epoch": 0.3669275929549902,
      "grad_norm": 6.118898868560791,
      "learning_rate": 1.970645792563601e-05,
      "loss": 0.7319,
      "step": 1500
    },
    {
      "epoch": 0.4892367906066536,
      "grad_norm": 5.870689392089844,
      "learning_rate": 1.9608610567514678e-05,
      "loss": 0.673,
      "step": 2000
    },
    {
      "epoch": 0.6115459882583171,
      "grad_norm": 5.046342849731445,
      "learning_rate": 1.951076320939335e-05,
      "loss": 0.6408,
      "step": 2500
    },
    {
      "epoch": 0.7338551859099804,
      "grad_norm": 5.137794494628906,
      "learning_rate": 1.9412915851272015e-05,
      "loss": 0.6029,
      "step": 3000
    },
    {
      "epoch": 0.8561643835616438,
      "grad_norm": 5.616779327392578,
      "learning_rate": 1.9315068493150686e-05,
      "loss": 0.5867,
      "step": 3500
    },
    {
      "epoch": 0.9784735812133072,
      "grad_norm": 3.4356954097747803,
      "learning_rate": 1.9217221135029354e-05,
      "loss": 0.5654,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8249802208386364,
      "eval_loss": 0.5318922400474548,
      "eval_runtime": 77.9312,
      "eval_samples_per_second": 373.034,
      "eval_steps_per_second": 11.664,
      "step": 4088
    },
    {
      "epoch": 1.1007827788649707,
      "grad_norm": 5.042200088500977,
      "learning_rate": 1.9119373776908026e-05,
      "loss": 0.5171,
      "step": 4500
    },
    {
      "epoch": 1.223091976516634,
      "grad_norm": 5.630020618438721,
      "learning_rate": 1.9021526418786694e-05,
      "loss": 0.4983,
      "step": 5000
    },
    {
      "epoch": 1.3454011741682974,
      "grad_norm": 3.5620667934417725,
      "learning_rate": 1.8923679060665363e-05,
      "loss": 0.488,
      "step": 5500
    },
    {
      "epoch": 1.467710371819961,
      "grad_norm": 6.017724990844727,
      "learning_rate": 1.8825831702544034e-05,
      "loss": 0.4867,
      "step": 6000
    },
    {
      "epoch": 1.5900195694716244,
      "grad_norm": 4.974002361297607,
      "learning_rate": 1.8727984344422703e-05,
      "loss": 0.4645,
      "step": 6500
    },
    {
      "epoch": 1.7123287671232876,
      "grad_norm": 5.986278057098389,
      "learning_rate": 1.863013698630137e-05,
      "loss": 0.4582,
      "step": 7000
    },
    {
      "epoch": 1.8346379647749511,
      "grad_norm": 5.255044937133789,
      "learning_rate": 1.853228962818004e-05,
      "loss": 0.4609,
      "step": 7500
    },
    {
      "epoch": 1.9569471624266144,
      "grad_norm": 5.023875713348389,
      "learning_rate": 1.843444227005871e-05,
      "loss": 0.4495,
      "step": 8000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8506759313405111,
      "eval_loss": 0.4426937699317932,
      "eval_runtime": 78.0987,
      "eval_samples_per_second": 372.234,
      "eval_steps_per_second": 11.639,
      "step": 8176
    },
    {
      "epoch": 2.079256360078278,
      "grad_norm": 6.947430610656738,
      "learning_rate": 1.833659491193738e-05,
      "loss": 0.4058,
      "step": 8500
    },
    {
      "epoch": 2.2015655577299413,
      "grad_norm": 2.9549953937530518,
      "learning_rate": 1.8238747553816047e-05,
      "loss": 0.3764,
      "step": 9000
    },
    {
      "epoch": 2.323874755381605,
      "grad_norm": 6.063886642456055,
      "learning_rate": 1.814090019569472e-05,
      "loss": 0.3785,
      "step": 9500
    },
    {
      "epoch": 2.446183953033268,
      "grad_norm": 6.911349296569824,
      "learning_rate": 1.8043052837573387e-05,
      "loss": 0.3736,
      "step": 10000
    },
    {
      "epoch": 2.5684931506849313,
      "grad_norm": 5.118221282958984,
      "learning_rate": 1.7945205479452055e-05,
      "loss": 0.3684,
      "step": 10500
    },
    {
      "epoch": 2.690802348336595,
      "grad_norm": 4.369062423706055,
      "learning_rate": 1.7847358121330724e-05,
      "loss": 0.3607,
      "step": 11000
    },
    {
      "epoch": 2.8131115459882583,
      "grad_norm": 6.040197849273682,
      "learning_rate": 1.7749510763209395e-05,
      "loss": 0.3657,
      "step": 11500
    },
    {
      "epoch": 2.935420743639922,
      "grad_norm": 6.751057147979736,
      "learning_rate": 1.7651663405088064e-05,
      "loss": 0.3517,
      "step": 12000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8710398679096006,
      "eval_loss": 0.3752569854259491,
      "eval_runtime": 78.0858,
      "eval_samples_per_second": 372.295,
      "eval_steps_per_second": 11.641,
      "step": 12264
    },
    {
      "epoch": 3.0577299412915853,
      "grad_norm": 5.28397798538208,
      "learning_rate": 1.7553816046966735e-05,
      "loss": 0.3172,
      "step": 12500
    },
    {
      "epoch": 3.1800391389432487,
      "grad_norm": 3.873892068862915,
      "learning_rate": 1.74559686888454e-05,
      "loss": 0.2865,
      "step": 13000
    },
    {
      "epoch": 3.302348336594912,
      "grad_norm": 3.584193468093872,
      "learning_rate": 1.735812133072407e-05,
      "loss": 0.285,
      "step": 13500
    },
    {
      "epoch": 3.4246575342465753,
      "grad_norm": 6.080500602722168,
      "learning_rate": 1.726027397260274e-05,
      "loss": 0.2867,
      "step": 14000
    },
    {
      "epoch": 3.5469667318982387,
      "grad_norm": 5.05246114730835,
      "learning_rate": 1.716242661448141e-05,
      "loss": 0.2793,
      "step": 14500
    },
    {
      "epoch": 3.6692759295499022,
      "grad_norm": 4.842950344085693,
      "learning_rate": 1.706457925636008e-05,
      "loss": 0.2829,
      "step": 15000
    },
    {
      "epoch": 3.7915851272015657,
      "grad_norm": 6.52248477935791,
      "learning_rate": 1.6966731898238748e-05,
      "loss": 0.2776,
      "step": 15500
    },
    {
      "epoch": 3.9138943248532287,
      "grad_norm": 5.814143657684326,
      "learning_rate": 1.686888454011742e-05,
      "loss": 0.268,
      "step": 16000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8836985311822779,
      "eval_loss": 0.33874690532684326,
      "eval_runtime": 78.7296,
      "eval_samples_per_second": 369.251,
      "eval_steps_per_second": 11.546,
      "step": 16352
    },
    {
      "epoch": 4.036203522504892,
      "grad_norm": 4.4718194007873535,
      "learning_rate": 1.6771037181996088e-05,
      "loss": 0.2524,
      "step": 16500
    },
    {
      "epoch": 4.158512720156556,
      "grad_norm": 6.8225531578063965,
      "learning_rate": 1.6673189823874756e-05,
      "loss": 0.2121,
      "step": 17000
    },
    {
      "epoch": 4.280821917808219,
      "grad_norm": 4.292703151702881,
      "learning_rate": 1.6575342465753425e-05,
      "loss": 0.2168,
      "step": 17500
    },
    {
      "epoch": 4.403131115459883,
      "grad_norm": 6.740828514099121,
      "learning_rate": 1.6477495107632096e-05,
      "loss": 0.2079,
      "step": 18000
    },
    {
      "epoch": 4.525440313111546,
      "grad_norm": 7.279580593109131,
      "learning_rate": 1.6379647749510764e-05,
      "loss": 0.2178,
      "step": 18500
    },
    {
      "epoch": 4.64774951076321,
      "grad_norm": 8.839160919189453,
      "learning_rate": 1.6281800391389433e-05,
      "loss": 0.2135,
      "step": 19000
    },
    {
      "epoch": 4.770058708414873,
      "grad_norm": 5.748088836669922,
      "learning_rate": 1.6183953033268104e-05,
      "loss": 0.215,
      "step": 19500
    },
    {
      "epoch": 4.892367906066536,
      "grad_norm": 6.222439765930176,
      "learning_rate": 1.6086105675146773e-05,
      "loss": 0.2114,
      "step": 20000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.895531629458911,
      "eval_loss": 0.3136844038963318,
      "eval_runtime": 78.002,
      "eval_samples_per_second": 372.695,
      "eval_steps_per_second": 11.654,
      "step": 20440
    },
    {
      "epoch": 5.014677103718199,
      "grad_norm": 5.067689895629883,
      "learning_rate": 1.598825831702544e-05,
      "loss": 0.2001,
      "step": 20500
    },
    {
      "epoch": 5.136986301369863,
      "grad_norm": 5.334500789642334,
      "learning_rate": 1.589041095890411e-05,
      "loss": 0.1574,
      "step": 21000
    },
    {
      "epoch": 5.259295499021526,
      "grad_norm": 8.226490020751953,
      "learning_rate": 1.579256360078278e-05,
      "loss": 0.1599,
      "step": 21500
    },
    {
      "epoch": 5.38160469667319,
      "grad_norm": 5.279560565948486,
      "learning_rate": 1.569471624266145e-05,
      "loss": 0.1594,
      "step": 22000
    },
    {
      "epoch": 5.503913894324853,
      "grad_norm": 8.88842487335205,
      "learning_rate": 1.559686888454012e-05,
      "loss": 0.1577,
      "step": 22500
    },
    {
      "epoch": 5.626223091976517,
      "grad_norm": 6.64274787902832,
      "learning_rate": 1.5499021526418785e-05,
      "loss": 0.1566,
      "step": 23000
    },
    {
      "epoch": 5.74853228962818,
      "grad_norm": 8.864998817443848,
      "learning_rate": 1.5401174168297457e-05,
      "loss": 0.163,
      "step": 23500
    },
    {
      "epoch": 5.870841487279844,
      "grad_norm": 4.494806289672852,
      "learning_rate": 1.5303326810176125e-05,
      "loss": 0.159,
      "step": 24000
    },
    {
      "epoch": 5.993150684931507,
      "grad_norm": 8.841043472290039,
      "learning_rate": 1.5205479452054797e-05,
      "loss": 0.1593,
      "step": 24500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9032025042138213,
      "eval_loss": 0.31098222732543945,
      "eval_runtime": 78.1256,
      "eval_samples_per_second": 372.106,
      "eval_steps_per_second": 11.635,
      "step": 24528
    },
    {
      "epoch": 6.1154598825831705,
      "grad_norm": 5.516913890838623,
      "learning_rate": 1.5107632093933464e-05,
      "loss": 0.1186,
      "step": 25000
    },
    {
      "epoch": 6.237769080234834,
      "grad_norm": 17.19109344482422,
      "learning_rate": 1.5009784735812134e-05,
      "loss": 0.1242,
      "step": 25500
    },
    {
      "epoch": 6.3600782778864975,
      "grad_norm": 4.817780494689941,
      "learning_rate": 1.4911937377690804e-05,
      "loss": 0.1264,
      "step": 26000
    },
    {
      "epoch": 6.48238747553816,
      "grad_norm": 9.93541431427002,
      "learning_rate": 1.4814090019569473e-05,
      "loss": 0.1251,
      "step": 26500
    },
    {
      "epoch": 6.604696673189824,
      "grad_norm": 7.402759552001953,
      "learning_rate": 1.4716242661448142e-05,
      "loss": 0.1205,
      "step": 27000
    },
    {
      "epoch": 6.727005870841487,
      "grad_norm": 4.997498989105225,
      "learning_rate": 1.4618395303326812e-05,
      "loss": 0.1221,
      "step": 27500
    },
    {
      "epoch": 6.8493150684931505,
      "grad_norm": 8.2695951461792,
      "learning_rate": 1.4520547945205482e-05,
      "loss": 0.126,
      "step": 28000
    },
    {
      "epoch": 6.971624266144814,
      "grad_norm": 7.226652145385742,
      "learning_rate": 1.4422700587084152e-05,
      "loss": 0.1238,
      "step": 28500
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9069863437790238,
      "eval_loss": 0.3151151239871979,
      "eval_runtime": 77.3038,
      "eval_samples_per_second": 376.062,
      "eval_steps_per_second": 11.759,
      "step": 28616
    },
    {
      "epoch": 7.0939334637964775,
      "grad_norm": 6.771386623382568,
      "learning_rate": 1.4324853228962818e-05,
      "loss": 0.1014,
      "step": 29000
    },
    {
      "epoch": 7.216242661448141,
      "grad_norm": 7.908430099487305,
      "learning_rate": 1.4227005870841488e-05,
      "loss": 0.0917,
      "step": 29500
    },
    {
      "epoch": 7.3385518590998045,
      "grad_norm": 11.553496360778809,
      "learning_rate": 1.4129158512720158e-05,
      "loss": 0.0989,
      "step": 30000
    },
    {
      "epoch": 7.460861056751468,
      "grad_norm": 5.48296594619751,
      "learning_rate": 1.4031311154598828e-05,
      "loss": 0.0945,
      "step": 30500
    },
    {
      "epoch": 7.583170254403131,
      "grad_norm": 4.382979869842529,
      "learning_rate": 1.3933463796477496e-05,
      "loss": 0.0904,
      "step": 31000
    },
    {
      "epoch": 7.705479452054795,
      "grad_norm": 4.479061126708984,
      "learning_rate": 1.3835616438356164e-05,
      "loss": 0.1003,
      "step": 31500
    },
    {
      "epoch": 7.8277886497064575,
      "grad_norm": 2.2252697944641113,
      "learning_rate": 1.3737769080234834e-05,
      "loss": 0.0963,
      "step": 32000
    },
    {
      "epoch": 7.950097847358121,
      "grad_norm": 5.149953842163086,
      "learning_rate": 1.3639921722113504e-05,
      "loss": 0.0969,
      "step": 32500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9127652987513329,
      "eval_loss": 0.3146636486053467,
      "eval_runtime": 77.5909,
      "eval_samples_per_second": 374.67,
      "eval_steps_per_second": 11.715,
      "step": 32704
    },
    {
      "epoch": 8.072407045009784,
      "grad_norm": 10.789752960205078,
      "learning_rate": 1.3542074363992173e-05,
      "loss": 0.0788,
      "step": 33000
    },
    {
      "epoch": 8.194716242661448,
      "grad_norm": 5.515634536743164,
      "learning_rate": 1.3444227005870843e-05,
      "loss": 0.0717,
      "step": 33500
    },
    {
      "epoch": 8.317025440313111,
      "grad_norm": 10.454425811767578,
      "learning_rate": 1.3346379647749513e-05,
      "loss": 0.0719,
      "step": 34000
    },
    {
      "epoch": 8.439334637964775,
      "grad_norm": 7.216288089752197,
      "learning_rate": 1.3248532289628183e-05,
      "loss": 0.0764,
      "step": 34500
    },
    {
      "epoch": 8.561643835616438,
      "grad_norm": 6.058413028717041,
      "learning_rate": 1.3150684931506849e-05,
      "loss": 0.0775,
      "step": 35000
    },
    {
      "epoch": 8.683953033268102,
      "grad_norm": 1.9321287870407104,
      "learning_rate": 1.3052837573385519e-05,
      "loss": 0.0762,
      "step": 35500
    },
    {
      "epoch": 8.806262230919765,
      "grad_norm": 5.581308841705322,
      "learning_rate": 1.2954990215264189e-05,
      "loss": 0.0763,
      "step": 36000
    },
    {
      "epoch": 8.928571428571429,
      "grad_norm": 8.296586036682129,
      "learning_rate": 1.2857142857142859e-05,
      "loss": 0.0826,
      "step": 36500
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9133156754153624,
      "eval_loss": 0.33166706562042236,
      "eval_runtime": 77.4584,
      "eval_samples_per_second": 375.311,
      "eval_steps_per_second": 11.735,
      "step": 36792
    },
    {
      "epoch": 9.050880626223092,
      "grad_norm": 6.40161657333374,
      "learning_rate": 1.2759295499021527e-05,
      "loss": 0.0689,
      "step": 37000
    },
    {
      "epoch": 9.173189823874756,
      "grad_norm": 9.357067108154297,
      "learning_rate": 1.2661448140900197e-05,
      "loss": 0.0581,
      "step": 37500
    },
    {
      "epoch": 9.29549902152642,
      "grad_norm": 4.852538108825684,
      "learning_rate": 1.2563600782778867e-05,
      "loss": 0.0582,
      "step": 38000
    },
    {
      "epoch": 9.417808219178083,
      "grad_norm": 6.204359531402588,
      "learning_rate": 1.2465753424657537e-05,
      "loss": 0.06,
      "step": 38500
    },
    {
      "epoch": 9.540117416829746,
      "grad_norm": 11.658947944641113,
      "learning_rate": 1.2367906066536204e-05,
      "loss": 0.0626,
      "step": 39000
    },
    {
      "epoch": 9.66242661448141,
      "grad_norm": 12.047215461730957,
      "learning_rate": 1.2270058708414874e-05,
      "loss": 0.0626,
      "step": 39500
    },
    {
      "epoch": 9.784735812133073,
      "grad_norm": 11.078405380249023,
      "learning_rate": 1.2172211350293543e-05,
      "loss": 0.0655,
      "step": 40000
    },
    {
      "epoch": 9.907045009784735,
      "grad_norm": 3.5147736072540283,
      "learning_rate": 1.2074363992172213e-05,
      "loss": 0.0663,
      "step": 40500
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9173747033125795,
      "eval_loss": 0.34417206048965454,
      "eval_runtime": 77.3814,
      "eval_samples_per_second": 375.685,
      "eval_steps_per_second": 11.747,
      "step": 40880
    },
    {
      "epoch": 10.029354207436398,
      "grad_norm": 2.839688539505005,
      "learning_rate": 1.1976516634050882e-05,
      "loss": 0.0609,
      "step": 41000
    },
    {
      "epoch": 10.151663405088062,
      "grad_norm": 10.794103622436523,
      "learning_rate": 1.187866927592955e-05,
      "loss": 0.0467,
      "step": 41500
    },
    {
      "epoch": 10.273972602739725,
      "grad_norm": 3.0600426197052,
      "learning_rate": 1.178082191780822e-05,
      "loss": 0.0464,
      "step": 42000
    },
    {
      "epoch": 10.396281800391389,
      "grad_norm": 7.135795593261719,
      "learning_rate": 1.168297455968689e-05,
      "loss": 0.051,
      "step": 42500
    },
    {
      "epoch": 10.518590998043052,
      "grad_norm": 5.585446834564209,
      "learning_rate": 1.1585127201565558e-05,
      "loss": 0.0531,
      "step": 43000
    },
    {
      "epoch": 10.640900195694716,
      "grad_norm": 0.05372029170393944,
      "learning_rate": 1.1487279843444228e-05,
      "loss": 0.0504,
      "step": 43500
    },
    {
      "epoch": 10.76320939334638,
      "grad_norm": 12.072264671325684,
      "learning_rate": 1.1389432485322898e-05,
      "loss": 0.0511,
      "step": 44000
    },
    {
      "epoch": 10.885518590998043,
      "grad_norm": 5.787899971008301,
      "learning_rate": 1.1291585127201568e-05,
      "loss": 0.0519,
      "step": 44500
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.9171339135220666,
      "eval_loss": 0.3852131962776184,
      "eval_runtime": 77.3953,
      "eval_samples_per_second": 375.617,
      "eval_steps_per_second": 11.745,
      "step": 44968
    },
    {
      "epoch": 11.007827788649706,
      "grad_norm": 8.351863861083984,
      "learning_rate": 1.1193737769080235e-05,
      "loss": 0.0508,
      "step": 45000
    },
    {
      "epoch": 11.13013698630137,
      "grad_norm": 6.567861080169678,
      "learning_rate": 1.1095890410958904e-05,
      "loss": 0.0403,
      "step": 45500
    },
    {
      "epoch": 11.252446183953033,
      "grad_norm": 5.368165016174316,
      "learning_rate": 1.0998043052837574e-05,
      "loss": 0.0419,
      "step": 46000
    },
    {
      "epoch": 11.374755381604697,
      "grad_norm": 11.540104866027832,
      "learning_rate": 1.0900195694716244e-05,
      "loss": 0.0466,
      "step": 46500
    },
    {
      "epoch": 11.49706457925636,
      "grad_norm": 2.5703656673431396,
      "learning_rate": 1.0802348336594913e-05,
      "loss": 0.0422,
      "step": 47000
    },
    {
      "epoch": 11.619373776908024,
      "grad_norm": 10.161659240722656,
      "learning_rate": 1.0704500978473583e-05,
      "loss": 0.0492,
      "step": 47500
    },
    {
      "epoch": 11.741682974559687,
      "grad_norm": 9.083320617675781,
      "learning_rate": 1.0606653620352253e-05,
      "loss": 0.0444,
      "step": 48000
    },
    {
      "epoch": 11.86399217221135,
      "grad_norm": 7.005116939544678,
      "learning_rate": 1.0508806262230922e-05,
      "loss": 0.0431,
      "step": 48500
    },
    {
      "epoch": 11.986301369863014,
      "grad_norm": 5.291450023651123,
      "learning_rate": 1.0410958904109589e-05,
      "loss": 0.045,
      "step": 49000
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9189226376801624,
      "eval_loss": 0.3933347463607788,
      "eval_runtime": 77.3468,
      "eval_samples_per_second": 375.852,
      "eval_steps_per_second": 11.752,
      "step": 49056
    },
    {
      "epoch": 12.108610567514678,
      "grad_norm": 3.0380940437316895,
      "learning_rate": 1.0313111545988259e-05,
      "loss": 0.0338,
      "step": 49500
    },
    {
      "epoch": 12.230919765166341,
      "grad_norm": 0.6975650787353516,
      "learning_rate": 1.0215264187866929e-05,
      "loss": 0.0343,
      "step": 50000
    },
    {
      "epoch": 12.353228962818005,
      "grad_norm": 0.18672548234462738,
      "learning_rate": 1.0117416829745599e-05,
      "loss": 0.0349,
      "step": 50500
    },
    {
      "epoch": 12.475538160469668,
      "grad_norm": 6.868373394012451,
      "learning_rate": 1.0019569471624267e-05,
      "loss": 0.0343,
      "step": 51000
    },
    {
      "epoch": 12.597847358121331,
      "grad_norm": 1.4025439023971558,
      "learning_rate": 9.921722113502935e-06,
      "loss": 0.0396,
      "step": 51500
    },
    {
      "epoch": 12.720156555772995,
      "grad_norm": 4.942438125610352,
      "learning_rate": 9.823874755381605e-06,
      "loss": 0.0381,
      "step": 52000
    },
    {
      "epoch": 12.842465753424658,
      "grad_norm": 4.044002056121826,
      "learning_rate": 9.726027397260275e-06,
      "loss": 0.0382,
      "step": 52500
    },
    {
      "epoch": 12.96477495107632,
      "grad_norm": 0.6736146211624146,
      "learning_rate": 9.628180039138944e-06,
      "loss": 0.0368,
      "step": 53000
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.9228096728698704,
      "eval_loss": 0.38157734274864197,
      "eval_runtime": 77.3834,
      "eval_samples_per_second": 375.675,
      "eval_steps_per_second": 11.747,
      "step": 53144
    },
    {
      "epoch": 13.087084148727984,
      "grad_norm": 1.1303954124450684,
      "learning_rate": 9.530332681017614e-06,
      "loss": 0.0306,
      "step": 53500
    },
    {
      "epoch": 13.209393346379647,
      "grad_norm": 1.899757742881775,
      "learning_rate": 9.432485322896282e-06,
      "loss": 0.0302,
      "step": 54000
    },
    {
      "epoch": 13.33170254403131,
      "grad_norm": 10.160910606384277,
      "learning_rate": 9.334637964774952e-06,
      "loss": 0.0284,
      "step": 54500
    },
    {
      "epoch": 13.454011741682974,
      "grad_norm": 6.715288162231445,
      "learning_rate": 9.23679060665362e-06,
      "loss": 0.0351,
      "step": 55000
    },
    {
      "epoch": 13.576320939334638,
      "grad_norm": 9.383013725280762,
      "learning_rate": 9.13894324853229e-06,
      "loss": 0.0342,
      "step": 55500
    },
    {
      "epoch": 13.698630136986301,
      "grad_norm": 0.5024349093437195,
      "learning_rate": 9.04109589041096e-06,
      "loss": 0.0358,
      "step": 56000
    },
    {
      "epoch": 13.820939334637965,
      "grad_norm": 6.495013236999512,
      "learning_rate": 8.943248532289628e-06,
      "loss": 0.0361,
      "step": 56500
    },
    {
      "epoch": 13.943248532289628,
      "grad_norm": 3.360276699066162,
      "learning_rate": 8.845401174168298e-06,
      "loss": 0.0332,
      "step": 57000
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.9229816655773795,
      "eval_loss": 0.4262143075466156,
      "eval_runtime": 77.1924,
      "eval_samples_per_second": 376.604,
      "eval_steps_per_second": 11.776,
      "step": 57232
    },
    {
      "epoch": 14.065557729941291,
      "grad_norm": 2.5697081089019775,
      "learning_rate": 8.747553816046968e-06,
      "loss": 0.0271,
      "step": 57500
    },
    {
      "epoch": 14.187866927592955,
      "grad_norm": 2.5552070140838623,
      "learning_rate": 8.649706457925636e-06,
      "loss": 0.0229,
      "step": 58000
    },
    {
      "epoch": 14.310176125244618,
      "grad_norm": 11.48392105102539,
      "learning_rate": 8.551859099804306e-06,
      "loss": 0.0275,
      "step": 58500
    },
    {
      "epoch": 14.432485322896282,
      "grad_norm": 4.35728645324707,
      "learning_rate": 8.454011741682975e-06,
      "loss": 0.026,
      "step": 59000
    },
    {
      "epoch": 14.554794520547945,
      "grad_norm": 18.579906463623047,
      "learning_rate": 8.356164383561644e-06,
      "loss": 0.0259,
      "step": 59500
    },
    {
      "epoch": 14.677103718199609,
      "grad_norm": 7.566853046417236,
      "learning_rate": 8.258317025440313e-06,
      "loss": 0.0281,
      "step": 60000
    },
    {
      "epoch": 14.799412915851272,
      "grad_norm": 0.4659290611743927,
      "learning_rate": 8.160469667318983e-06,
      "loss": 0.0283,
      "step": 60500
    },
    {
      "epoch": 14.921722113502936,
      "grad_norm": 5.014148235321045,
      "learning_rate": 8.062622309197653e-06,
      "loss": 0.0289,
      "step": 61000
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.9236008393244126,
      "eval_loss": 0.43280109763145447,
      "eval_runtime": 77.9069,
      "eval_samples_per_second": 373.151,
      "eval_steps_per_second": 11.668,
      "step": 61320
    },
    {
      "epoch": 15.0440313111546,
      "grad_norm": 4.811933994293213,
      "learning_rate": 7.964774951076321e-06,
      "loss": 0.027,
      "step": 61500
    },
    {
      "epoch": 15.166340508806263,
      "grad_norm": 9.811365127563477,
      "learning_rate": 7.86692759295499e-06,
      "loss": 0.0214,
      "step": 62000
    },
    {
      "epoch": 15.288649706457926,
      "grad_norm": 1.3940095901489258,
      "learning_rate": 7.76908023483366e-06,
      "loss": 0.0219,
      "step": 62500
    },
    {
      "epoch": 15.41095890410959,
      "grad_norm": 0.06226726993918419,
      "learning_rate": 7.671232876712329e-06,
      "loss": 0.0226,
      "step": 63000
    },
    {
      "epoch": 15.533268101761253,
      "grad_norm": 8.294939994812012,
      "learning_rate": 7.573385518590999e-06,
      "loss": 0.0251,
      "step": 63500
    },
    {
      "epoch": 15.655577299412915,
      "grad_norm": 1.8952441215515137,
      "learning_rate": 7.475538160469667e-06,
      "loss": 0.0232,
      "step": 64000
    },
    {
      "epoch": 15.777886497064578,
      "grad_norm": 5.1830315589904785,
      "learning_rate": 7.377690802348337e-06,
      "loss": 0.0254,
      "step": 64500
    },
    {
      "epoch": 15.900195694716242,
      "grad_norm": 1.0473072528839111,
      "learning_rate": 7.279843444227006e-06,
      "loss": 0.0248,
      "step": 65000
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.9260431357710434,
      "eval_loss": 0.4366961419582367,
      "eval_runtime": 77.3755,
      "eval_samples_per_second": 375.713,
      "eval_steps_per_second": 11.748,
      "step": 65408
    },
    {
      "epoch": 16.022504892367905,
      "grad_norm": 0.5687731504440308,
      "learning_rate": 7.181996086105676e-06,
      "loss": 0.0238,
      "step": 65500
    },
    {
      "epoch": 16.14481409001957,
      "grad_norm": 1.2872580289840698,
      "learning_rate": 7.0841487279843445e-06,
      "loss": 0.0194,
      "step": 66000
    },
    {
      "epoch": 16.267123287671232,
      "grad_norm": 5.059693813323975,
      "learning_rate": 6.9863013698630145e-06,
      "loss": 0.0199,
      "step": 66500
    },
    {
      "epoch": 16.389432485322896,
      "grad_norm": 6.738853931427002,
      "learning_rate": 6.8884540117416836e-06,
      "loss": 0.0205,
      "step": 67000
    },
    {
      "epoch": 16.51174168297456,
      "grad_norm": 11.091848373413086,
      "learning_rate": 6.790606653620353e-06,
      "loss": 0.0215,
      "step": 67500
    },
    {
      "epoch": 16.634050880626223,
      "grad_norm": 0.19167885184288025,
      "learning_rate": 6.692759295499022e-06,
      "loss": 0.022,
      "step": 68000
    },
    {
      "epoch": 16.756360078277886,
      "grad_norm": 5.966000080108643,
      "learning_rate": 6.594911937377692e-06,
      "loss": 0.0203,
      "step": 68500
    },
    {
      "epoch": 16.87866927592955,
      "grad_norm": 3.5066215991973877,
      "learning_rate": 6.49706457925636e-06,
      "loss": 0.0187,
      "step": 69000
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.9272470847236077,
      "eval_loss": 0.44777625799179077,
      "eval_runtime": 77.1492,
      "eval_samples_per_second": 376.815,
      "eval_steps_per_second": 11.782,
      "step": 69496
    },
    {
      "epoch": 17.000978473581213,
      "grad_norm": 0.6858524680137634,
      "learning_rate": 6.39921722113503e-06,
      "loss": 0.0204,
      "step": 69500
    },
    {
      "epoch": 17.123287671232877,
      "grad_norm": 9.703988075256348,
      "learning_rate": 6.301369863013699e-06,
      "loss": 0.0161,
      "step": 70000
    },
    {
      "epoch": 17.24559686888454,
      "grad_norm": 4.51020622253418,
      "learning_rate": 6.203522504892369e-06,
      "loss": 0.016,
      "step": 70500
    },
    {
      "epoch": 17.367906066536204,
      "grad_norm": 0.09826533496379852,
      "learning_rate": 6.105675146771037e-06,
      "loss": 0.0177,
      "step": 71000
    },
    {
      "epoch": 17.490215264187867,
      "grad_norm": 11.620870590209961,
      "learning_rate": 6.007827788649707e-06,
      "loss": 0.02,
      "step": 71500
    },
    {
      "epoch": 17.61252446183953,
      "grad_norm": 8.661576271057129,
      "learning_rate": 5.909980430528376e-06,
      "loss": 0.017,
      "step": 72000
    },
    {
      "epoch": 17.734833659491194,
      "grad_norm": 0.7512950897216797,
      "learning_rate": 5.812133072407045e-06,
      "loss": 0.0186,
      "step": 72500
    },
    {
      "epoch": 17.857142857142858,
      "grad_norm": 0.44471457600593567,
      "learning_rate": 5.7142857142857145e-06,
      "loss": 0.0228,
      "step": 73000
    },
    {
      "epoch": 17.97945205479452,
      "grad_norm": 1.0926498174667358,
      "learning_rate": 5.6164383561643845e-06,
      "loss": 0.0178,
      "step": 73500
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.9264559182690654,
      "eval_loss": 0.46360349655151367,
      "eval_runtime": 77.3263,
      "eval_samples_per_second": 375.952,
      "eval_steps_per_second": 11.755,
      "step": 73584
    },
    {
      "epoch": 18.101761252446185,
      "grad_norm": 8.134523391723633,
      "learning_rate": 5.518590998043053e-06,
      "loss": 0.0153,
      "step": 74000
    },
    {
      "epoch": 18.224070450097848,
      "grad_norm": 0.08053633570671082,
      "learning_rate": 5.420743639921723e-06,
      "loss": 0.016,
      "step": 74500
    },
    {
      "epoch": 18.34637964774951,
      "grad_norm": 0.44435006380081177,
      "learning_rate": 5.322896281800392e-06,
      "loss": 0.0146,
      "step": 75000
    },
    {
      "epoch": 18.468688845401175,
      "grad_norm": 2.0229480266571045,
      "learning_rate": 5.225048923679062e-06,
      "loss": 0.017,
      "step": 75500
    },
    {
      "epoch": 18.59099804305284,
      "grad_norm": 4.502400875091553,
      "learning_rate": 5.12720156555773e-06,
      "loss": 0.017,
      "step": 76000
    },
    {
      "epoch": 18.713307240704502,
      "grad_norm": 0.14978571236133575,
      "learning_rate": 5.0293542074364e-06,
      "loss": 0.0154,
      "step": 76500
    },
    {
      "epoch": 18.835616438356166,
      "grad_norm": 9.524068832397461,
      "learning_rate": 4.931506849315069e-06,
      "loss": 0.0165,
      "step": 77000
    },
    {
      "epoch": 18.95792563600783,
      "grad_norm": 0.9707121253013611,
      "learning_rate": 4.833659491193738e-06,
      "loss": 0.0154,
      "step": 77500
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.9266623095180765,
      "eval_loss": 0.47839030623435974,
      "eval_runtime": 77.6682,
      "eval_samples_per_second": 374.297,
      "eval_steps_per_second": 11.704,
      "step": 77672
    },
    {
      "epoch": 19.080234833659492,
      "grad_norm": 11.346772193908691,
      "learning_rate": 4.735812133072407e-06,
      "loss": 0.0136,
      "step": 78000
    },
    {
      "epoch": 19.202544031311156,
      "grad_norm": 0.7873362302780151,
      "learning_rate": 4.637964774951076e-06,
      "loss": 0.0131,
      "step": 78500
    },
    {
      "epoch": 19.32485322896282,
      "grad_norm": 1.0734236240386963,
      "learning_rate": 4.5401174168297455e-06,
      "loss": 0.0156,
      "step": 79000
    },
    {
      "epoch": 19.447162426614483,
      "grad_norm": 0.031163902953267097,
      "learning_rate": 4.442270058708415e-06,
      "loss": 0.0148,
      "step": 79500
    },
    {
      "epoch": 19.569471624266146,
      "grad_norm": 3.0132200717926025,
      "learning_rate": 4.3444227005870845e-06,
      "loss": 0.0142,
      "step": 80000
    },
    {
      "epoch": 19.69178082191781,
      "grad_norm": 5.072915554046631,
      "learning_rate": 4.246575342465754e-06,
      "loss": 0.0133,
      "step": 80500
    },
    {
      "epoch": 19.81409001956947,
      "grad_norm": 0.12334894388914108,
      "learning_rate": 4.148727984344423e-06,
      "loss": 0.0153,
      "step": 81000
    },
    {
      "epoch": 19.936399217221137,
      "grad_norm": 3.8586535453796387,
      "learning_rate": 4.050880626223092e-06,
      "loss": 0.0138,
      "step": 81500
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.9280726497196519,
      "eval_loss": 0.4880678355693817,
      "eval_runtime": 77.1562,
      "eval_samples_per_second": 376.781,
      "eval_steps_per_second": 11.781,
      "step": 81760
    },
    {
      "epoch": 20.058708414872797,
      "grad_norm": 5.422935485839844,
      "learning_rate": 3.953033268101762e-06,
      "loss": 0.0123,
      "step": 82000
    },
    {
      "epoch": 20.18101761252446,
      "grad_norm": 0.02908429130911827,
      "learning_rate": 3.855185909980431e-06,
      "loss": 0.0127,
      "step": 82500
    },
    {
      "epoch": 20.303326810176124,
      "grad_norm": 0.9683161973953247,
      "learning_rate": 3.7573385518591e-06,
      "loss": 0.0106,
      "step": 83000
    },
    {
      "epoch": 20.425636007827787,
      "grad_norm": 2.011960029602051,
      "learning_rate": 3.659491193737769e-06,
      "loss": 0.0129,
      "step": 83500
    },
    {
      "epoch": 20.54794520547945,
      "grad_norm": 0.011290821246802807,
      "learning_rate": 3.5616438356164386e-06,
      "loss": 0.0098,
      "step": 84000
    },
    {
      "epoch": 20.670254403131114,
      "grad_norm": 3.6643381118774414,
      "learning_rate": 3.4637964774951077e-06,
      "loss": 0.0123,
      "step": 84500
    },
    {
      "epoch": 20.792563600782778,
      "grad_norm": 0.28946495056152344,
      "learning_rate": 3.365949119373777e-06,
      "loss": 0.0119,
      "step": 85000
    },
    {
      "epoch": 20.91487279843444,
      "grad_norm": 0.8526332974433899,
      "learning_rate": 3.2681017612524463e-06,
      "loss": 0.0127,
      "step": 85500
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.9281758453441574,
      "eval_loss": 0.49737828969955444,
      "eval_runtime": 77.6066,
      "eval_samples_per_second": 374.594,
      "eval_steps_per_second": 11.713,
      "step": 85848
    },
    {
      "epoch": 21.037181996086105,
      "grad_norm": 0.038626112043857574,
      "learning_rate": 3.1702544031311154e-06,
      "loss": 0.0133,
      "step": 86000
    },
    {
      "epoch": 21.159491193737768,
      "grad_norm": 1.5674916505813599,
      "learning_rate": 3.072407045009785e-06,
      "loss": 0.0079,
      "step": 86500
    },
    {
      "epoch": 21.28180039138943,
      "grad_norm": 0.023863431066274643,
      "learning_rate": 2.974559686888454e-06,
      "loss": 0.0089,
      "step": 87000
    },
    {
      "epoch": 21.404109589041095,
      "grad_norm": 0.027295144274830818,
      "learning_rate": 2.876712328767123e-06,
      "loss": 0.0099,
      "step": 87500
    },
    {
      "epoch": 21.52641878669276,
      "grad_norm": 1.8621610403060913,
      "learning_rate": 2.7788649706457927e-06,
      "loss": 0.01,
      "step": 88000
    },
    {
      "epoch": 21.648727984344422,
      "grad_norm": 0.09760987758636475,
      "learning_rate": 2.681017612524462e-06,
      "loss": 0.0115,
      "step": 88500
    },
    {
      "epoch": 21.771037181996086,
      "grad_norm": 0.0476505346596241,
      "learning_rate": 2.5831702544031313e-06,
      "loss": 0.0096,
      "step": 89000
    },
    {
      "epoch": 21.89334637964775,
      "grad_norm": 0.08127936720848083,
      "learning_rate": 2.4853228962818004e-06,
      "loss": 0.0116,
      "step": 89500
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.9293797942967218,
      "eval_loss": 0.4935356378555298,
      "eval_runtime": 79.4825,
      "eval_samples_per_second": 365.754,
      "eval_steps_per_second": 11.436,
      "step": 89936
    },
    {
      "epoch": 22.015655577299412,
      "grad_norm": 1.2286592721939087,
      "learning_rate": 2.3874755381604695e-06,
      "loss": 0.0086,
      "step": 90000
    },
    {
      "epoch": 22.137964774951076,
      "grad_norm": 1.9946578741073608,
      "learning_rate": 2.289628180039139e-06,
      "loss": 0.008,
      "step": 90500
    },
    {
      "epoch": 22.26027397260274,
      "grad_norm": 0.024408530443906784,
      "learning_rate": 2.191780821917808e-06,
      "loss": 0.0115,
      "step": 91000
    },
    {
      "epoch": 22.382583170254403,
      "grad_norm": 0.011225187219679356,
      "learning_rate": 2.0939334637964777e-06,
      "loss": 0.0085,
      "step": 91500
    },
    {
      "epoch": 22.504892367906066,
      "grad_norm": 1.2969368696212769,
      "learning_rate": 1.996086105675147e-06,
      "loss": 0.0089,
      "step": 92000
    },
    {
      "epoch": 22.62720156555773,
      "grad_norm": 6.824276447296143,
      "learning_rate": 1.8982387475538161e-06,
      "loss": 0.0088,
      "step": 92500
    },
    {
      "epoch": 22.749510763209393,
      "grad_norm": 0.052455250173807144,
      "learning_rate": 1.8003913894324854e-06,
      "loss": 0.0088,
      "step": 93000
    },
    {
      "epoch": 22.871819960861057,
      "grad_norm": 0.04021551460027695,
      "learning_rate": 1.7025440313111545e-06,
      "loss": 0.009,
      "step": 93500
    },
    {
      "epoch": 22.99412915851272,
      "grad_norm": 0.06902284175157547,
      "learning_rate": 1.6046966731898239e-06,
      "loss": 0.0106,
      "step": 94000
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.93065254033229,
      "eval_loss": 0.4948473870754242,
      "eval_runtime": 79.6762,
      "eval_samples_per_second": 364.864,
      "eval_steps_per_second": 11.409,
      "step": 94024
    },
    {
      "epoch": 23.116438356164384,
      "grad_norm": 0.5471883416175842,
      "learning_rate": 1.5068493150684932e-06,
      "loss": 0.0068,
      "step": 94500
    },
    {
      "epoch": 23.238747553816047,
      "grad_norm": 0.16019482910633087,
      "learning_rate": 1.4090019569471625e-06,
      "loss": 0.0089,
      "step": 95000
    },
    {
      "epoch": 23.36105675146771,
      "grad_norm": 6.4080963134765625,
      "learning_rate": 1.3111545988258318e-06,
      "loss": 0.0077,
      "step": 95500
    },
    {
      "epoch": 23.483365949119374,
      "grad_norm": 0.13792355358600616,
      "learning_rate": 1.213307240704501e-06,
      "loss": 0.0076,
      "step": 96000
    },
    {
      "epoch": 23.605675146771038,
      "grad_norm": 2.563711404800415,
      "learning_rate": 1.1154598825831702e-06,
      "loss": 0.0075,
      "step": 96500
    },
    {
      "epoch": 23.7279843444227,
      "grad_norm": 0.04809940978884697,
      "learning_rate": 1.0176125244618395e-06,
      "loss": 0.0067,
      "step": 97000
    },
    {
      "epoch": 23.850293542074365,
      "grad_norm": 0.057318106293678284,
      "learning_rate": 9.197651663405089e-07,
      "loss": 0.0062,
      "step": 97500
    },
    {
      "epoch": 23.972602739726028,
      "grad_norm": 4.935258865356445,
      "learning_rate": 8.219178082191781e-07,
      "loss": 0.0085,
      "step": 98000
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.9311685184548175,
      "eval_loss": 0.49324169754981995,
      "eval_runtime": 79.011,
      "eval_samples_per_second": 367.936,
      "eval_steps_per_second": 11.505,
      "step": 98112
    },
    {
      "epoch": 24.09491193737769,
      "grad_norm": 6.268829345703125,
      "learning_rate": 7.240704500978474e-07,
      "loss": 0.0055,
      "step": 98500
    },
    {
      "epoch": 24.217221135029355,
      "grad_norm": 0.00456130551174283,
      "learning_rate": 6.262230919765167e-07,
      "loss": 0.0054,
      "step": 99000
    },
    {
      "epoch": 24.33953033268102,
      "grad_norm": 0.014587494544684887,
      "learning_rate": 5.283757338551859e-07,
      "loss": 0.0067,
      "step": 99500
    },
    {
      "epoch": 24.461839530332682,
      "grad_norm": 0.7037560343742371,
      "learning_rate": 4.305283757338552e-07,
      "loss": 0.0059,
      "step": 100000
    },
    {
      "epoch": 24.584148727984346,
      "grad_norm": 0.006079969462007284,
      "learning_rate": 3.326810176125245e-07,
      "loss": 0.0074,
      "step": 100500
    },
    {
      "epoch": 24.70645792563601,
      "grad_norm": 1.3056970834732056,
      "learning_rate": 2.3483365949119375e-07,
      "loss": 0.0075,
      "step": 101000
    },
    {
      "epoch": 24.828767123287673,
      "grad_norm": 8.87458324432373,
      "learning_rate": 1.36986301369863e-07,
      "loss": 0.0065,
      "step": 101500
    },
    {
      "epoch": 24.951076320939336,
      "grad_norm": 14.917941093444824,
      "learning_rate": 3.9138943248532294e-08,
      "loss": 0.0072,
      "step": 102000
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.9311341199133156,
      "eval_loss": 0.49332907795906067,
      "eval_runtime": 79.9831,
      "eval_samples_per_second": 363.464,
      "eval_steps_per_second": 11.365,
      "step": 102200
    }
  ],
  "logging_steps": 500,
  "max_steps": 102200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.33360446554112e+17,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}