{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.997632202052092,
  "eval_steps": 100,
  "global_step": 1899,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01578531965272297,
      "grad_norm": 0.7275460362434387,
      "learning_rate": 4.0000000000000003e-07,
      "loss": 0.6962,
      "step": 10
    },
    {
      "epoch": 0.03157063930544594,
      "grad_norm": 0.5963730216026306,
      "learning_rate": 8.000000000000001e-07,
      "loss": 0.6989,
      "step": 20
    },
    {
      "epoch": 0.0473559589581689,
      "grad_norm": 1.253847360610962,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 0.6935,
      "step": 30
    },
    {
      "epoch": 0.06314127861089187,
      "grad_norm": 1.5691624879837036,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 0.6947,
      "step": 40
    },
    {
      "epoch": 0.07892659826361484,
      "grad_norm": 1.1631684303283691,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.6889,
      "step": 50
    },
    {
      "epoch": 0.0947119179163378,
      "grad_norm": 1.152625322341919,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 0.6776,
      "step": 60
    },
    {
      "epoch": 0.11049723756906077,
      "grad_norm": 1.720926284790039,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 0.6468,
      "step": 70
    },
    {
      "epoch": 0.12628255722178375,
      "grad_norm": 3.781614065170288,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 0.6133,
      "step": 80
    },
    {
      "epoch": 0.1420678768745067,
      "grad_norm": 2.526336669921875,
      "learning_rate": 3.6000000000000003e-06,
      "loss": 0.5555,
      "step": 90
    },
    {
      "epoch": 0.15785319652722968,
      "grad_norm": 2.09712290763855,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.5515,
      "step": 100
    },
    {
      "epoch": 0.15785319652722968,
      "eval_accuracy": 0.7578864353312302,
      "eval_accuracy_label_test": 0.5070365902694008,
      "eval_accuracy_label_train": 0.9992263056092844,
      "eval_f1": 0.7415473366691369,
      "eval_loss": 0.5193965435028076,
      "eval_precision": 0.8351793628382634,
      "eval_recall": 0.7578864353312302,
      "eval_runtime": 14.5015,
      "eval_samples_per_second": 349.758,
      "eval_steps_per_second": 21.86,
      "step": 100
    },
    {
      "epoch": 0.17363851617995266,
      "grad_norm": 1.363462209701538,
      "learning_rate": 4.4e-06,
      "loss": 0.5304,
      "step": 110
    },
    {
      "epoch": 0.1894238358326756,
      "grad_norm": 1.779582142829895,
      "learning_rate": 4.800000000000001e-06,
      "loss": 0.5147,
      "step": 120
    },
    {
      "epoch": 0.2052091554853986,
      "grad_norm": 3.7067956924438477,
      "learning_rate": 5.2e-06,
      "loss": 0.4892,
      "step": 130
    },
    {
      "epoch": 0.22099447513812154,
      "grad_norm": 19.64886474609375,
      "learning_rate": 5.600000000000001e-06,
      "loss": 0.4963,
      "step": 140
    },
    {
      "epoch": 0.23677979479084452,
      "grad_norm": 12.231091499328613,
      "learning_rate": 6e-06,
      "loss": 0.451,
      "step": 150
    },
    {
      "epoch": 0.2525651144435675,
      "grad_norm": 5.452483654022217,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 0.3406,
      "step": 160
    },
    {
      "epoch": 0.26835043409629045,
      "grad_norm": 4.669571876525879,
      "learning_rate": 6.800000000000001e-06,
      "loss": 0.3745,
      "step": 170
    },
    {
      "epoch": 0.2841357537490134,
      "grad_norm": 2.206510305404663,
      "learning_rate": 7.2000000000000005e-06,
      "loss": 0.3036,
      "step": 180
    },
    {
      "epoch": 0.2999210734017364,
      "grad_norm": 6.932628631591797,
      "learning_rate": 7.600000000000001e-06,
      "loss": 0.2632,
      "step": 190
    },
    {
      "epoch": 0.31570639305445936,
      "grad_norm": 3.4731557369232178,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.2205,
      "step": 200
    },
    {
      "epoch": 0.31570639305445936,
      "eval_accuracy": 0.9300078864353313,
      "eval_accuracy_label_test": 0.9883393646964214,
      "eval_accuracy_label_train": 0.8738878143133463,
      "eval_f1": 0.9298468332917054,
      "eval_loss": 0.25367653369903564,
      "eval_precision": 0.9361225540545063,
      "eval_recall": 0.9300078864353313,
      "eval_runtime": 14.577,
      "eval_samples_per_second": 347.944,
      "eval_steps_per_second": 21.747,
      "step": 200
    },
    {
      "epoch": 0.3314917127071823,
      "grad_norm": 3.005352735519409,
      "learning_rate": 8.400000000000001e-06,
      "loss": 0.217,
      "step": 210
    },
    {
      "epoch": 0.3472770323599053,
      "grad_norm": 5.314383506774902,
      "learning_rate": 8.8e-06,
      "loss": 0.1349,
      "step": 220
    },
    {
      "epoch": 0.36306235201262826,
      "grad_norm": 1.332632303237915,
      "learning_rate": 9.200000000000002e-06,
      "loss": 0.092,
      "step": 230
    },
    {
      "epoch": 0.3788476716653512,
      "grad_norm": 14.597169876098633,
      "learning_rate": 9.600000000000001e-06,
      "loss": 0.0767,
      "step": 240
    },
    {
      "epoch": 0.39463299131807417,
      "grad_norm": 0.3883499503135681,
      "learning_rate": 1e-05,
      "loss": 0.1029,
      "step": 250
    },
    {
      "epoch": 0.4104183109707972,
      "grad_norm": 18.310523986816406,
      "learning_rate": 1.04e-05,
      "loss": 0.0856,
      "step": 260
    },
    {
      "epoch": 0.4262036306235201,
      "grad_norm": 6.817958354949951,
      "learning_rate": 1.0800000000000002e-05,
      "loss": 0.0514,
      "step": 270
    },
    {
      "epoch": 0.4419889502762431,
      "grad_norm": 10.453371047973633,
      "learning_rate": 1.1200000000000001e-05,
      "loss": 0.0925,
      "step": 280
    },
    {
      "epoch": 0.4577742699289661,
      "grad_norm": 0.29950886964797974,
      "learning_rate": 1.16e-05,
      "loss": 0.0727,
      "step": 290
    },
    {
      "epoch": 0.47355958958168903,
      "grad_norm": 3.3916432857513428,
      "learning_rate": 1.2e-05,
      "loss": 0.1106,
      "step": 300
    },
    {
      "epoch": 0.47355958958168903,
      "eval_accuracy": 0.9128548895899053,
      "eval_accuracy_label_test": 0.995979091274628,
      "eval_accuracy_label_train": 0.8328820116054159,
      "eval_f1": 0.9123898201159903,
      "eval_loss": 0.3449595868587494,
      "eval_precision": 0.9248260970850004,
      "eval_recall": 0.9128548895899053,
      "eval_runtime": 14.6481,
      "eval_samples_per_second": 346.258,
      "eval_steps_per_second": 21.641,
      "step": 300
    },
    {
      "epoch": 0.489344909234412,
      "grad_norm": 12.914605140686035,
      "learning_rate": 1.2400000000000002e-05,
      "loss": 0.061,
      "step": 310
    },
    {
      "epoch": 0.505130228887135,
      "grad_norm": 0.24762243032455444,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 0.0929,
      "step": 320
    },
    {
      "epoch": 0.5209155485398579,
      "grad_norm": 0.3279378116130829,
      "learning_rate": 1.3200000000000002e-05,
      "loss": 0.0511,
      "step": 330
    },
    {
      "epoch": 0.5367008681925809,
      "grad_norm": 7.465598106384277,
      "learning_rate": 1.3600000000000002e-05,
      "loss": 0.0727,
      "step": 340
    },
    {
      "epoch": 0.5524861878453039,
      "grad_norm": 0.9897856712341309,
      "learning_rate": 1.4e-05,
      "loss": 0.095,
      "step": 350
    },
    {
      "epoch": 0.5682715074980268,
      "grad_norm": 19.12044334411621,
      "learning_rate": 1.4400000000000001e-05,
      "loss": 0.0386,
      "step": 360
    },
    {
      "epoch": 0.5840568271507498,
      "grad_norm": 3.458350419998169,
      "learning_rate": 1.48e-05,
      "loss": 0.0467,
      "step": 370
    },
    {
      "epoch": 0.5998421468034728,
      "grad_norm": 0.10713621973991394,
      "learning_rate": 1.5200000000000002e-05,
      "loss": 0.0316,
      "step": 380
    },
    {
      "epoch": 0.6156274664561957,
      "grad_norm": 0.3854784071445465,
      "learning_rate": 1.5600000000000003e-05,
      "loss": 0.0396,
      "step": 390
    },
    {
      "epoch": 0.6314127861089187,
      "grad_norm": 6.402571201324463,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.0384,
      "step": 400
    },
    {
      "epoch": 0.6314127861089187,
      "eval_accuracy": 0.9682570977917981,
      "eval_accuracy_label_test": 0.983514274225975,
      "eval_accuracy_label_train": 0.9535783365570599,
      "eval_f1": 0.9682589499365617,
      "eval_loss": 0.14084377884864807,
      "eval_precision": 0.968730671942094,
      "eval_recall": 0.9682570977917981,
      "eval_runtime": 14.7066,
      "eval_samples_per_second": 344.878,
      "eval_steps_per_second": 21.555,
      "step": 400
    },
    {
      "epoch": 0.6471981057616417,
      "grad_norm": 7.416443347930908,
      "learning_rate": 1.64e-05,
      "loss": 0.0354,
      "step": 410
    },
    {
      "epoch": 0.6629834254143646,
      "grad_norm": 25.271848678588867,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 0.1069,
      "step": 420
    },
    {
      "epoch": 0.6787687450670876,
      "grad_norm": 0.10532895475625992,
      "learning_rate": 1.72e-05,
      "loss": 0.0411,
      "step": 430
    },
    {
      "epoch": 0.6945540647198106,
      "grad_norm": 0.05838713422417641,
      "learning_rate": 1.76e-05,
      "loss": 0.0593,
      "step": 440
    },
    {
      "epoch": 0.7103393843725335,
      "grad_norm": 0.06444909423589706,
      "learning_rate": 1.8e-05,
      "loss": 0.0073,
      "step": 450
    },
    {
      "epoch": 0.7261247040252565,
      "grad_norm": 6.492961883544922,
      "learning_rate": 1.8400000000000003e-05,
      "loss": 0.0539,
      "step": 460
    },
    {
      "epoch": 0.7419100236779794,
      "grad_norm": 0.08374536037445068,
      "learning_rate": 1.88e-05,
      "loss": 0.0757,
      "step": 470
    },
    {
      "epoch": 0.7576953433307024,
      "grad_norm": 0.05994931235909462,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 0.0201,
      "step": 480
    },
    {
      "epoch": 0.7734806629834254,
      "grad_norm": 17.886022567749023,
      "learning_rate": 1.9600000000000002e-05,
      "loss": 0.0536,
      "step": 490
    },
    {
      "epoch": 0.7892659826361483,
      "grad_norm": 2.434602737426758,
      "learning_rate": 2e-05,
      "loss": 0.0631,
      "step": 500
    },
    {
      "epoch": 0.7892659826361483,
      "eval_accuracy": 0.9631309148264984,
      "eval_accuracy_label_test": 0.989545637314033,
      "eval_accuracy_label_train": 0.9377176015473888,
      "eval_f1": 0.9631237556597715,
      "eval_loss": 0.15165044367313385,
      "eval_precision": 0.964482811637024,
      "eval_recall": 0.9631309148264984,
      "eval_runtime": 14.5963,
      "eval_samples_per_second": 347.485,
      "eval_steps_per_second": 21.718,
      "step": 500
    },
    {
      "epoch": 0.8050513022888713,
      "grad_norm": 2.8019778728485107,
      "learning_rate": 1.9857040743388135e-05,
      "loss": 0.0191,
      "step": 510
    },
    {
      "epoch": 0.8208366219415943,
      "grad_norm": 2.233394145965576,
      "learning_rate": 1.971408148677627e-05,
      "loss": 0.0317,
      "step": 520
    },
    {
      "epoch": 0.8366219415943172,
      "grad_norm": 0.1284877210855484,
      "learning_rate": 1.9571122230164405e-05,
      "loss": 0.0467,
      "step": 530
    },
    {
      "epoch": 0.8524072612470402,
      "grad_norm": 0.0791168138384819,
      "learning_rate": 1.942816297355254e-05,
      "loss": 0.0455,
      "step": 540
    },
    {
      "epoch": 0.8681925808997633,
      "grad_norm": 0.35995563864707947,
      "learning_rate": 1.928520371694067e-05,
      "loss": 0.0253,
      "step": 550
    },
    {
      "epoch": 0.8839779005524862,
      "grad_norm": 2.419013738632202,
      "learning_rate": 1.9142244460328808e-05,
      "loss": 0.0551,
      "step": 560
    },
    {
      "epoch": 0.8997632202052092,
      "grad_norm": 12.438931465148926,
      "learning_rate": 1.899928520371694e-05,
      "loss": 0.0563,
      "step": 570
    },
    {
      "epoch": 0.9155485398579322,
      "grad_norm": 0.049261171370744705,
      "learning_rate": 1.8856325947105075e-05,
      "loss": 0.0291,
      "step": 580
    },
    {
      "epoch": 0.9313338595106551,
      "grad_norm": 0.04032573848962784,
      "learning_rate": 1.871336669049321e-05,
      "loss": 0.0026,
      "step": 590
    },
    {
      "epoch": 0.9471191791633781,
      "grad_norm": 1.215957522392273,
      "learning_rate": 1.8570407433881345e-05,
      "loss": 0.0276,
      "step": 600
    },
    {
      "epoch": 0.9471191791633781,
      "eval_accuracy": 0.9386829652996845,
      "eval_accuracy_label_test": 0.9947728186570165,
      "eval_accuracy_label_train": 0.8847195357833656,
      "eval_f1": 0.9385557615622165,
      "eval_loss": 0.36488327383995056,
      "eval_precision": 0.9444060841035706,
      "eval_recall": 0.9386829652996845,
      "eval_runtime": 14.5956,
      "eval_samples_per_second": 347.502,
      "eval_steps_per_second": 21.719,
      "step": 600
    },
    {
      "epoch": 0.9629044988161011,
      "grad_norm": 0.0634990781545639,
      "learning_rate": 1.842744817726948e-05,
      "loss": 0.0267,
      "step": 610
    },
    {
      "epoch": 0.978689818468824,
      "grad_norm": 0.029669882729649544,
      "learning_rate": 1.8284488920657615e-05,
      "loss": 0.0426,
      "step": 620
    },
    {
      "epoch": 0.994475138121547,
      "grad_norm": 0.029255308210849762,
      "learning_rate": 1.8141529664045748e-05,
      "loss": 0.0556,
      "step": 630
    },
    {
      "epoch": 1.01026045777427,
      "grad_norm": 0.0757206529378891,
      "learning_rate": 1.799857040743388e-05,
      "loss": 0.0458,
      "step": 640
    },
    {
      "epoch": 1.0260457774269929,
      "grad_norm": 0.1398855596780777,
      "learning_rate": 1.7855611150822018e-05,
      "loss": 0.0641,
      "step": 650
    },
    {
      "epoch": 1.0418310970797158,
      "grad_norm": 7.250219821929932,
      "learning_rate": 1.771265189421015e-05,
      "loss": 0.0204,
      "step": 660
    },
    {
      "epoch": 1.057616416732439,
      "grad_norm": 0.022936342284083366,
      "learning_rate": 1.7569692637598285e-05,
      "loss": 0.002,
      "step": 670
    },
    {
      "epoch": 1.0734017363851618,
      "grad_norm": 4.3541741371154785,
      "learning_rate": 1.742673338098642e-05,
      "loss": 0.0148,
      "step": 680
    },
    {
      "epoch": 1.0891870560378847,
      "grad_norm": 0.024669496342539787,
      "learning_rate": 1.7283774124374555e-05,
      "loss": 0.062,
      "step": 690
    },
    {
      "epoch": 1.1049723756906078,
      "grad_norm": 0.023665498942136765,
      "learning_rate": 1.7140814867762688e-05,
      "loss": 0.0245,
      "step": 700
    },
    {
      "epoch": 1.1049723756906078,
      "eval_accuracy": 0.9702287066246057,
      "eval_accuracy_label_test": 0.9726578206674709,
      "eval_accuracy_label_train": 0.9678916827852998,
      "eval_f1": 0.9702301478266859,
      "eval_loss": 0.13387098908424377,
      "eval_precision": 0.9702485636354361,
      "eval_recall": 0.9702287066246057,
      "eval_runtime": 14.562,
      "eval_samples_per_second": 348.304,
      "eval_steps_per_second": 21.769,
      "step": 700
    },
    {
      "epoch": 1.1207576953433307,
      "grad_norm": 20.071300506591797,
      "learning_rate": 1.6997855611150825e-05,
      "loss": 0.0867,
      "step": 710
    },
    {
      "epoch": 1.1365430149960536,
      "grad_norm": 0.016291845589876175,
      "learning_rate": 1.6854896354538958e-05,
      "loss": 0.0167,
      "step": 720
    },
    {
      "epoch": 1.1523283346487767,
      "grad_norm": 0.02355334907770157,
      "learning_rate": 1.671193709792709e-05,
      "loss": 0.003,
      "step": 730
    },
    {
      "epoch": 1.1681136543014996,
      "grad_norm": 0.014460445381700993,
      "learning_rate": 1.6568977841315225e-05,
      "loss": 0.0137,
      "step": 740
    },
    {
      "epoch": 1.1838989739542225,
      "grad_norm": 9.110311508178711,
      "learning_rate": 1.642601858470336e-05,
      "loss": 0.0592,
      "step": 750
    },
    {
      "epoch": 1.1996842936069456,
      "grad_norm": 0.03517633676528931,
      "learning_rate": 1.6283059328091495e-05,
      "loss": 0.0356,
      "step": 760
    },
    {
      "epoch": 1.2154696132596685,
      "grad_norm": 1.65547776222229,
      "learning_rate": 1.6140100071479628e-05,
      "loss": 0.0206,
      "step": 770
    },
    {
      "epoch": 1.2312549329123914,
      "grad_norm": 0.044892311096191406,
      "learning_rate": 1.5997140814867765e-05,
      "loss": 0.0138,
      "step": 780
    },
    {
      "epoch": 1.2470402525651145,
      "grad_norm": 16.877887725830078,
      "learning_rate": 1.5854181558255898e-05,
      "loss": 0.0201,
      "step": 790
    },
    {
      "epoch": 1.2628255722178374,
      "grad_norm": 13.245466232299805,
      "learning_rate": 1.5711222301644035e-05,
      "loss": 0.0519,
      "step": 800
    },
    {
      "epoch": 1.2628255722178374,
      "eval_accuracy": 0.9185725552050473,
      "eval_accuracy_label_test": 0.9991958182549256,
      "eval_accuracy_label_train": 0.8410058027079303,
      "eval_f1": 0.9181684151069668,
      "eval_loss": 0.4945215582847595,
      "eval_precision": 0.9299426884244839,
      "eval_recall": 0.9185725552050473,
      "eval_runtime": 14.6038,
      "eval_samples_per_second": 347.306,
      "eval_steps_per_second": 21.707,
      "step": 800
    },
    {
      "epoch": 1.2786108918705603,
      "grad_norm": 12.54496955871582,
      "learning_rate": 1.5568263045032168e-05,
      "loss": 0.0505,
      "step": 810
    },
    {
      "epoch": 1.2943962115232832,
      "grad_norm": 0.01659429259598255,
      "learning_rate": 1.54253037884203e-05,
      "loss": 0.0246,
      "step": 820
    },
    {
      "epoch": 1.3101815311760063,
      "grad_norm": 0.03832576796412468,
      "learning_rate": 1.5282344531808435e-05,
      "loss": 0.0206,
      "step": 830
    },
    {
      "epoch": 1.3259668508287292,
      "grad_norm": 4.976266384124756,
      "learning_rate": 1.513938527519657e-05,
      "loss": 0.011,
      "step": 840
    },
    {
      "epoch": 1.3417521704814521,
      "grad_norm": 16.507583618164062,
      "learning_rate": 1.4996426018584705e-05,
      "loss": 0.0971,
      "step": 850
    },
    {
      "epoch": 1.3575374901341752,
      "grad_norm": 9.206993103027344,
      "learning_rate": 1.485346676197284e-05,
      "loss": 0.0314,
      "step": 860
    },
    {
      "epoch": 1.3733228097868981,
      "grad_norm": 0.040578652173280716,
      "learning_rate": 1.4710507505360973e-05,
      "loss": 0.0053,
      "step": 870
    },
    {
      "epoch": 1.389108129439621,
      "grad_norm": 15.86640739440918,
      "learning_rate": 1.4567548248749108e-05,
      "loss": 0.0268,
      "step": 880
    },
    {
      "epoch": 1.4048934490923441,
      "grad_norm": 1.9645894765853882,
      "learning_rate": 1.4424588992137243e-05,
      "loss": 0.0348,
      "step": 890
    },
    {
      "epoch": 1.420678768745067,
      "grad_norm": 0.01832072250545025,
      "learning_rate": 1.4281629735525378e-05,
      "loss": 0.02,
      "step": 900
    },
    {
      "epoch": 1.420678768745067,
      "eval_accuracy": 0.9548501577287066,
      "eval_accuracy_label_test": 0.995979091274628,
      "eval_accuracy_label_train": 0.9152804642166344,
      "eval_f1": 0.9548094220396917,
      "eval_loss": 0.26368606090545654,
      "eval_precision": 0.958024003205365,
      "eval_recall": 0.9548501577287066,
      "eval_runtime": 14.5839,
      "eval_samples_per_second": 347.781,
      "eval_steps_per_second": 21.736,
      "step": 900
    },
    {
      "epoch": 1.43646408839779,
      "grad_norm": 2.4122228622436523,
      "learning_rate": 1.413867047891351e-05,
      "loss": 0.0431,
      "step": 910
    },
    {
      "epoch": 1.452249408050513,
      "grad_norm": 0.04305430129170418,
      "learning_rate": 1.3995711222301644e-05,
      "loss": 0.0382,
      "step": 920
    },
    {
      "epoch": 1.468034727703236,
      "grad_norm": 0.045583341270685196,
      "learning_rate": 1.385275196568978e-05,
      "loss": 0.0345,
      "step": 930
    },
    {
      "epoch": 1.4838200473559588,
      "grad_norm": 0.08313434571027756,
      "learning_rate": 1.3709792709077914e-05,
      "loss": 0.0229,
      "step": 940
    },
    {
      "epoch": 1.499605367008682,
      "grad_norm": 0.040190890431404114,
      "learning_rate": 1.3566833452466048e-05,
      "loss": 0.0022,
      "step": 950
    },
    {
      "epoch": 1.5153906866614049,
      "grad_norm": 0.12089771777391434,
      "learning_rate": 1.3423874195854183e-05,
      "loss": 0.0126,
      "step": 960
    },
    {
      "epoch": 1.5311760063141278,
      "grad_norm": 0.0291321761906147,
      "learning_rate": 1.3280914939242318e-05,
      "loss": 0.0246,
      "step": 970
    },
    {
      "epoch": 1.5469613259668509,
      "grad_norm": 0.06409675627946854,
      "learning_rate": 1.3137955682630453e-05,
      "loss": 0.0611,
      "step": 980
    },
    {
      "epoch": 1.5627466456195738,
      "grad_norm": 9.994110107421875,
      "learning_rate": 1.2994996426018586e-05,
      "loss": 0.0102,
      "step": 990
    },
    {
      "epoch": 1.5785319652722967,
      "grad_norm": 10.403185844421387,
      "learning_rate": 1.285203716940672e-05,
      "loss": 0.0325,
      "step": 1000
    },
    {
      "epoch": 1.5785319652722967,
      "eval_accuracy": 0.9708201892744479,
      "eval_accuracy_label_test": 0.9851226377161239,
      "eval_accuracy_label_train": 0.9570599613152805,
      "eval_f1": 0.970822203818026,
      "eval_loss": 0.11654461175203323,
      "eval_precision": 0.9712375667689825,
      "eval_recall": 0.9708201892744479,
      "eval_runtime": 14.5989,
      "eval_samples_per_second": 347.423,
      "eval_steps_per_second": 21.714,
      "step": 1000
    },
    {
      "epoch": 1.5943172849250198,
      "grad_norm": 0.05680805817246437,
      "learning_rate": 1.2709077912794854e-05,
      "loss": 0.0028,
      "step": 1010
    },
    {
      "epoch": 1.6101026045777427,
      "grad_norm": 0.017462020739912987,
      "learning_rate": 1.2566118656182988e-05,
      "loss": 0.0357,
      "step": 1020
    },
    {
      "epoch": 1.6258879242304656,
      "grad_norm": 5.752472877502441,
      "learning_rate": 1.2423159399571123e-05,
      "loss": 0.0457,
      "step": 1030
    },
    {
      "epoch": 1.6416732438831887,
      "grad_norm": 0.14555960893630981,
      "learning_rate": 1.2280200142959258e-05,
      "loss": 0.0062,
      "step": 1040
    },
    {
      "epoch": 1.6574585635359116,
      "grad_norm": 0.019668666645884514,
      "learning_rate": 1.2137240886347393e-05,
      "loss": 0.0046,
      "step": 1050
    },
    {
      "epoch": 1.6732438831886345,
      "grad_norm": 0.01673435978591442,
      "learning_rate": 1.1994281629735528e-05,
      "loss": 0.0535,
      "step": 1060
    },
    {
      "epoch": 1.6890292028413576,
      "grad_norm": 4.553021430969238,
      "learning_rate": 1.1851322373123661e-05,
      "loss": 0.0767,
      "step": 1070
    },
    {
      "epoch": 1.7048145224940805,
      "grad_norm": 0.33790552616119385,
      "learning_rate": 1.1708363116511796e-05,
      "loss": 0.0433,
      "step": 1080
    },
    {
      "epoch": 1.7205998421468034,
      "grad_norm": 0.08579400181770325,
      "learning_rate": 1.156540385989993e-05,
      "loss": 0.0051,
      "step": 1090
    },
    {
      "epoch": 1.7363851617995265,
      "grad_norm": 0.24498964846134186,
      "learning_rate": 1.1422444603288063e-05,
      "loss": 0.016,
      "step": 1100
    },
    {
      "epoch": 1.7363851617995265,
      "eval_accuracy": 0.9692429022082019,
      "eval_accuracy_label_test": 0.9529553679131484,
      "eval_accuracy_label_train": 0.9849129593810445,
      "eval_f1": 0.9692264691880251,
      "eval_loss": 0.10065295547246933,
      "eval_precision": 0.9696695972871102,
      "eval_recall": 0.9692429022082019,
      "eval_runtime": 14.644,
      "eval_samples_per_second": 346.353,
      "eval_steps_per_second": 21.647,
      "step": 1100
    },
    {
      "epoch": 1.7521704814522494,
      "grad_norm": 3.6310765743255615,
      "learning_rate": 1.1279485346676198e-05,
      "loss": 0.0676,
      "step": 1110
    },
    {
      "epoch": 1.7679558011049723,
      "grad_norm": 0.26799193024635315,
      "learning_rate": 1.1136526090064333e-05,
      "loss": 0.011,
      "step": 1120
    },
    {
      "epoch": 1.7837411207576954,
      "grad_norm": 0.050716836005449295,
      "learning_rate": 1.0993566833452468e-05,
      "loss": 0.0474,
      "step": 1130
    },
    {
      "epoch": 1.7995264404104183,
      "grad_norm": 0.03631124272942543,
      "learning_rate": 1.0850607576840601e-05,
      "loss": 0.0013,
      "step": 1140
    },
    {
      "epoch": 1.8153117600631412,
      "grad_norm": 0.016972020268440247,
      "learning_rate": 1.0707648320228736e-05,
      "loss": 0.0098,
      "step": 1150
    },
    {
      "epoch": 1.8310970797158643,
      "grad_norm": 0.01586087793111801,
      "learning_rate": 1.0564689063616871e-05,
      "loss": 0.0013,
      "step": 1160
    },
    {
      "epoch": 1.8468823993685872,
      "grad_norm": 0.01447396818548441,
      "learning_rate": 1.0421729807005006e-05,
      "loss": 0.0148,
      "step": 1170
    },
    {
      "epoch": 1.8626677190213101,
      "grad_norm": 0.08018597215414047,
      "learning_rate": 1.0278770550393137e-05,
      "loss": 0.0067,
      "step": 1180
    },
    {
      "epoch": 1.8784530386740332,
      "grad_norm": 0.020933715626597404,
      "learning_rate": 1.0135811293781272e-05,
      "loss": 0.0177,
      "step": 1190
    },
    {
      "epoch": 1.8942383583267561,
      "grad_norm": 0.006128642242401838,
      "learning_rate": 9.992852037169407e-06,
      "loss": 0.0068,
      "step": 1200
    },
    {
      "epoch": 1.8942383583267561,
      "eval_accuracy": 0.9690457413249212,
      "eval_accuracy_label_test": 0.9871330920788098,
      "eval_accuracy_label_train": 0.9516441005802708,
      "eval_f1": 0.9690463008445059,
      "eval_loss": 0.16786406934261322,
      "eval_precision": 0.9696986578163762,
      "eval_recall": 0.9690457413249212,
      "eval_runtime": 14.5325,
      "eval_samples_per_second": 349.011,
      "eval_steps_per_second": 21.813,
      "step": 1200
    },
    {
      "epoch": 1.910023677979479,
      "grad_norm": 0.009371940977871418,
      "learning_rate": 9.849892780557542e-06,
      "loss": 0.024,
      "step": 1210
    },
    {
      "epoch": 1.9258089976322021,
      "grad_norm": 0.014078453183174133,
      "learning_rate": 9.706933523945676e-06,
      "loss": 0.0005,
      "step": 1220
    },
    {
      "epoch": 1.941594317284925,
      "grad_norm": 0.009515127167105675,
      "learning_rate": 9.56397426733381e-06,
      "loss": 0.0046,
      "step": 1230
    },
    {
      "epoch": 1.957379636937648,
      "grad_norm": 0.014615291729569435,
      "learning_rate": 9.421015010721944e-06,
      "loss": 0.0261,
      "step": 1240
    },
    {
      "epoch": 1.973164956590371,
      "grad_norm": 0.06883493065834045,
      "learning_rate": 9.278055754110079e-06,
      "loss": 0.0233,
      "step": 1250
    },
    {
      "epoch": 1.988950276243094,
      "grad_norm": 0.013247163034975529,
      "learning_rate": 9.135096497498214e-06,
      "loss": 0.0168,
      "step": 1260
    },
    {
      "epoch": 2.004735595895817,
      "grad_norm": 0.012575499713420868,
      "learning_rate": 8.992137240886349e-06,
      "loss": 0.0124,
      "step": 1270
    },
    {
      "epoch": 2.02052091554854,
      "grad_norm": 0.017276756465435028,
      "learning_rate": 8.849177984274482e-06,
      "loss": 0.0184,
      "step": 1280
    },
    {
      "epoch": 2.0363062352012626,
      "grad_norm": 0.03682788833975792,
      "learning_rate": 8.706218727662616e-06,
      "loss": 0.0092,
      "step": 1290
    },
    {
      "epoch": 2.0520915548539858,
      "grad_norm": 0.35351473093032837,
      "learning_rate": 8.56325947105075e-06,
      "loss": 0.0042,
      "step": 1300
    },
    {
      "epoch": 2.0520915548539858,
      "eval_accuracy": 0.9733832807570978,
      "eval_accuracy_label_test": 0.9722557297949337,
      "eval_accuracy_label_train": 0.9744680851063829,
      "eval_f1": 0.9733829671319272,
      "eval_loss": 0.11822589486837387,
      "eval_precision": 0.973383334857871,
      "eval_recall": 0.9733832807570978,
      "eval_runtime": 14.5763,
      "eval_samples_per_second": 347.961,
      "eval_steps_per_second": 21.748,
      "step": 1300
    },
    {
      "epoch": 2.067876874506709,
      "grad_norm": 0.012184061110019684,
      "learning_rate": 8.420300214438886e-06,
      "loss": 0.0441,
      "step": 1310
    },
    {
      "epoch": 2.0836621941594315,
      "grad_norm": 0.0278099924325943,
      "learning_rate": 8.27734095782702e-06,
      "loss": 0.008,
      "step": 1320
    },
    {
      "epoch": 2.0994475138121547,
      "grad_norm": 0.9140133261680603,
      "learning_rate": 8.134381701215154e-06,
      "loss": 0.0017,
      "step": 1330
    },
    {
      "epoch": 2.115232833464878,
      "grad_norm": 1.3029474020004272,
      "learning_rate": 7.991422444603289e-06,
      "loss": 0.0007,
      "step": 1340
    },
    {
      "epoch": 2.1310181531176005,
      "grad_norm": 0.010257584042847157,
      "learning_rate": 7.848463187991422e-06,
      "loss": 0.0158,
      "step": 1350
    },
    {
      "epoch": 2.1468034727703236,
      "grad_norm": 0.010051756165921688,
      "learning_rate": 7.705503931379557e-06,
      "loss": 0.0034,
      "step": 1360
    },
    {
      "epoch": 2.1625887924230467,
      "grad_norm": 0.10394269973039627,
      "learning_rate": 7.562544674767692e-06,
      "loss": 0.0404,
      "step": 1370
    },
    {
      "epoch": 2.1783741120757694,
      "grad_norm": 0.011765426024794579,
      "learning_rate": 7.419585418155826e-06,
      "loss": 0.0004,
      "step": 1380
    },
    {
      "epoch": 2.1941594317284925,
      "grad_norm": 1.1091564893722534,
      "learning_rate": 7.276626161543961e-06,
      "loss": 0.0006,
      "step": 1390
    },
    {
      "epoch": 2.2099447513812156,
      "grad_norm": 0.010108958929777145,
      "learning_rate": 7.133666904932095e-06,
      "loss": 0.0005,
      "step": 1400
    },
    {
      "epoch": 2.2099447513812156,
      "eval_accuracy": 0.9729889589905363,
      "eval_accuracy_label_test": 0.9798954563731403,
      "eval_accuracy_label_train": 0.9663442940038685,
      "eval_f1": 0.9729913291480798,
      "eval_loss": 0.1431707739830017,
      "eval_precision": 0.9730972588569067,
      "eval_recall": 0.9729889589905363,
      "eval_runtime": 14.631,
      "eval_samples_per_second": 346.662,
      "eval_steps_per_second": 21.666,
      "step": 1400
    },
    {
      "epoch": 2.2257300710339383,
      "grad_norm": 0.02260042168200016,
      "learning_rate": 6.99070764832023e-06,
      "loss": 0.0014,
      "step": 1410
    },
    {
      "epoch": 2.2415153906866614,
      "grad_norm": 0.0077365124598145485,
      "learning_rate": 6.847748391708363e-06,
      "loss": 0.0004,
      "step": 1420
    },
    {
      "epoch": 2.2573007103393845,
      "grad_norm": 0.008837452158331871,
      "learning_rate": 6.704789135096498e-06,
      "loss": 0.0153,
      "step": 1430
    },
    {
      "epoch": 2.273086029992107,
      "grad_norm": 0.021829022094607353,
      "learning_rate": 6.561829878484632e-06,
      "loss": 0.0003,
      "step": 1440
    },
    {
      "epoch": 2.2888713496448303,
      "grad_norm": 0.014965805225074291,
      "learning_rate": 6.418870621872767e-06,
      "loss": 0.0246,
      "step": 1450
    },
    {
      "epoch": 2.3046566692975534,
      "grad_norm": 0.013634726405143738,
      "learning_rate": 6.275911365260901e-06,
      "loss": 0.0339,
      "step": 1460
    },
    {
      "epoch": 2.320441988950276,
      "grad_norm": 6.8833794593811035,
      "learning_rate": 6.1329521086490355e-06,
      "loss": 0.0059,
      "step": 1470
    },
    {
      "epoch": 2.336227308602999,
      "grad_norm": 0.04814034327864647,
      "learning_rate": 5.98999285203717e-06,
      "loss": 0.0122,
      "step": 1480
    },
    {
      "epoch": 2.3520126282557223,
      "grad_norm": 0.026584748178720474,
      "learning_rate": 5.847033595425305e-06,
      "loss": 0.0022,
      "step": 1490
    },
    {
      "epoch": 2.367797947908445,
      "grad_norm": 0.015504554845392704,
      "learning_rate": 5.704074338813439e-06,
      "loss": 0.0182,
      "step": 1500
    },
    {
      "epoch": 2.367797947908445,
      "eval_accuracy": 0.9718059936908517,
      "eval_accuracy_label_test": 0.9871330920788098,
      "eval_accuracy_label_train": 0.9570599613152805,
      "eval_f1": 0.9718076387635299,
      "eval_loss": 0.14604002237319946,
      "eval_precision": 0.972280875829533,
      "eval_recall": 0.9718059936908517,
      "eval_runtime": 14.6031,
      "eval_samples_per_second": 347.324,
      "eval_steps_per_second": 21.708,
      "step": 1500
    },
    {
      "epoch": 2.383583267561168,
      "grad_norm": 0.07147639244794846,
      "learning_rate": 5.561115082201572e-06,
      "loss": 0.0005,
      "step": 1510
    },
    {
      "epoch": 2.3993685872138912,
      "grad_norm": 0.024004925042390823,
      "learning_rate": 5.418155825589707e-06,
      "loss": 0.0439,
      "step": 1520
    },
    {
      "epoch": 2.415153906866614,
      "grad_norm": 0.07405655831098557,
      "learning_rate": 5.275196568977842e-06,
      "loss": 0.0004,
      "step": 1530
    },
    {
      "epoch": 2.430939226519337,
      "grad_norm": 0.01881037838757038,
      "learning_rate": 5.132237312365976e-06,
      "loss": 0.0004,
      "step": 1540
    },
    {
      "epoch": 2.44672454617206,
      "grad_norm": 0.05772541090846062,
      "learning_rate": 4.98927805575411e-06,
      "loss": 0.0059,
      "step": 1550
    },
    {
      "epoch": 2.462509865824783,
      "grad_norm": 0.014262210577726364,
      "learning_rate": 4.846318799142245e-06,
      "loss": 0.0004,
      "step": 1560
    },
    {
      "epoch": 2.478295185477506,
      "grad_norm": 0.005270448978990316,
      "learning_rate": 4.703359542530379e-06,
      "loss": 0.0003,
      "step": 1570
    },
    {
      "epoch": 2.494080505130229,
      "grad_norm": 0.008271850645542145,
      "learning_rate": 4.560400285918514e-06,
      "loss": 0.0231,
      "step": 1580
    },
    {
      "epoch": 2.5098658247829517,
      "grad_norm": 0.013198798522353172,
      "learning_rate": 4.417441029306648e-06,
      "loss": 0.0034,
      "step": 1590
    },
    {
      "epoch": 2.525651144435675,
      "grad_norm": 0.07948605716228485,
      "learning_rate": 4.274481772694782e-06,
      "loss": 0.0004,
      "step": 1600
    },
    {
      "epoch": 2.525651144435675,
      "eval_accuracy": 0.973186119873817,
      "eval_accuracy_label_test": 0.9843184559710495,
      "eval_accuracy_label_train": 0.9624758220502901,
      "eval_f1": 0.973188538207199,
      "eval_loss": 0.13827118277549744,
      "eval_precision": 0.973445480376393,
      "eval_recall": 0.973186119873817,
      "eval_runtime": 14.6108,
      "eval_samples_per_second": 347.14,
      "eval_steps_per_second": 21.696,
      "step": 1600
    },
    {
      "epoch": 2.541436464088398,
      "grad_norm": 0.24930787086486816,
      "learning_rate": 4.131522516082916e-06,
      "loss": 0.0221,
      "step": 1610
    },
    {
      "epoch": 2.5572217837411206,
      "grad_norm": 0.011009340174496174,
      "learning_rate": 3.988563259471051e-06,
      "loss": 0.0003,
      "step": 1620
    },
    {
      "epoch": 2.5730071033938438,
      "grad_norm": 0.02251257933676243,
      "learning_rate": 3.845604002859185e-06,
      "loss": 0.0003,
      "step": 1630
    },
    {
      "epoch": 2.5887924230465664,
      "grad_norm": 0.004396820440888405,
      "learning_rate": 3.70264474624732e-06,
      "loss": 0.0332,
      "step": 1640
    },
    {
      "epoch": 2.6045777426992895,
      "grad_norm": 1.1856316328048706,
      "learning_rate": 3.5596854896354545e-06,
      "loss": 0.0006,
      "step": 1650
    },
    {
      "epoch": 2.6203630623520127,
      "grad_norm": 0.019586117938160896,
      "learning_rate": 3.4167262330235886e-06,
      "loss": 0.0132,
      "step": 1660
    },
    {
      "epoch": 2.636148382004736,
      "grad_norm": 0.010670648887753487,
      "learning_rate": 3.273766976411723e-06,
      "loss": 0.0007,
      "step": 1670
    },
    {
      "epoch": 2.6519337016574585,
      "grad_norm": 0.37919822335243225,
      "learning_rate": 3.130807719799857e-06,
      "loss": 0.0003,
      "step": 1680
    },
    {
      "epoch": 2.6677190213101816,
      "grad_norm": 0.2330523133277893,
      "learning_rate": 2.987848463187992e-06,
      "loss": 0.0004,
      "step": 1690
    },
    {
      "epoch": 2.6835043409629042,
      "grad_norm": 0.007127601653337479,
      "learning_rate": 2.8448892065761256e-06,
      "loss": 0.0003,
      "step": 1700
    },
    {
      "epoch": 2.6835043409629042,
      "eval_accuracy": 0.9743690851735016,
      "eval_accuracy_label_test": 0.9831121833534379,
      "eval_accuracy_label_train": 0.9659574468085106,
      "eval_f1": 0.9743714686586191,
      "eval_loss": 0.13812877237796783,
      "eval_precision": 0.9745341434373056,
      "eval_recall": 0.9743690851735016,
      "eval_runtime": 14.5972,
      "eval_samples_per_second": 347.464,
      "eval_steps_per_second": 21.717,
      "step": 1700
    },
    {
      "epoch": 2.6992896606156274,
      "grad_norm": 0.005769920535385609,
      "learning_rate": 2.7019299499642602e-06,
      "loss": 0.0005,
      "step": 1710
    },
    {
      "epoch": 2.7150749802683505,
      "grad_norm": 0.005288603249937296,
      "learning_rate": 2.5589706933523952e-06,
      "loss": 0.0006,
      "step": 1720
    },
    {
      "epoch": 2.7308602999210736,
      "grad_norm": 0.005266103427857161,
      "learning_rate": 2.416011436740529e-06,
      "loss": 0.0032,
      "step": 1730
    },
    {
      "epoch": 2.7466456195737963,
      "grad_norm": 0.005252339411526918,
      "learning_rate": 2.2730521801286635e-06,
      "loss": 0.0006,
      "step": 1740
    },
    {
      "epoch": 2.7624309392265194,
      "grad_norm": 0.004924137610942125,
      "learning_rate": 2.1300929235167977e-06,
      "loss": 0.0002,
      "step": 1750
    },
    {
      "epoch": 2.778216258879242,
      "grad_norm": 0.003662313334643841,
      "learning_rate": 1.9871336669049322e-06,
      "loss": 0.0017,
      "step": 1760
    },
    {
      "epoch": 2.794001578531965,
      "grad_norm": 0.00455325935035944,
      "learning_rate": 1.8441744102930666e-06,
      "loss": 0.0002,
      "step": 1770
    },
    {
      "epoch": 2.8097868981846883,
      "grad_norm": 0.0084315724670887,
      "learning_rate": 1.701215153681201e-06,
      "loss": 0.0007,
      "step": 1780
    },
    {
      "epoch": 2.8255722178374114,
      "grad_norm": 0.021251995116472244,
      "learning_rate": 1.5582558970693353e-06,
      "loss": 0.0002,
      "step": 1790
    },
    {
      "epoch": 2.841357537490134,
      "grad_norm": 0.005972001701593399,
      "learning_rate": 1.4152966404574697e-06,
      "loss": 0.0002,
      "step": 1800
    },
    {
      "epoch": 2.841357537490134,
      "eval_accuracy": 0.972397476340694,
      "eval_accuracy_label_test": 0.9863289103337354,
      "eval_accuracy_label_train": 0.9589941972920696,
      "eval_f1": 0.9723994850059051,
      "eval_loss": 0.1599443554878235,
      "eval_precision": 0.9727934306989875,
      "eval_recall": 0.972397476340694,
      "eval_runtime": 14.9428,
      "eval_samples_per_second": 339.427,
      "eval_steps_per_second": 21.214,
      "step": 1800
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.005644650664180517,
      "learning_rate": 1.272337383845604e-06,
      "loss": 0.0002,
      "step": 1810
    },
    {
      "epoch": 2.87292817679558,
      "grad_norm": 0.010099658742547035,
      "learning_rate": 1.1293781272337384e-06,
      "loss": 0.0003,
      "step": 1820
    },
    {
      "epoch": 2.888713496448303,
      "grad_norm": 0.0057546221651136875,
      "learning_rate": 9.864188706218728e-07,
      "loss": 0.0002,
      "step": 1830
    },
    {
      "epoch": 2.904498816101026,
      "grad_norm": 0.010815597139298916,
      "learning_rate": 8.434596140100073e-07,
      "loss": 0.0104,
      "step": 1840
    },
    {
      "epoch": 2.9202841357537492,
      "grad_norm": 0.0059271338395774364,
      "learning_rate": 7.005003573981415e-07,
      "loss": 0.0002,
      "step": 1850
    },
    {
      "epoch": 2.936069455406472,
      "grad_norm": 0.004583253525197506,
      "learning_rate": 5.57541100786276e-07,
      "loss": 0.0033,
      "step": 1860
    },
    {
      "epoch": 2.951854775059195,
      "grad_norm": 0.003865251550450921,
      "learning_rate": 4.145818441744103e-07,
      "loss": 0.0102,
      "step": 1870
    },
    {
      "epoch": 2.9676400947119177,
      "grad_norm": 0.006317495368421078,
      "learning_rate": 2.716225875625447e-07,
      "loss": 0.0002,
      "step": 1880
    },
    {
      "epoch": 2.983425414364641,
      "grad_norm": 0.004333311691880226,
      "learning_rate": 1.2866333095067907e-07,
      "loss": 0.0003,
      "step": 1890
    },
    {
      "epoch": 2.997632202052092,
      "step": 1899,
      "total_flos": 7993457212661760.0,
      "train_loss": 0.08041724508642717,
      "train_runtime": 817.1898,
      "train_samples_per_second": 74.406,
      "train_steps_per_second": 2.324
    },
    {
      "epoch": 2.997632202052092,
      "eval_accuracy": 0.9704258675078864,
      "eval_accuracy_label_test": 0.9879372738238842,
      "eval_accuracy_label_train": 0.9535783365570599,
      "eval_f1": 0.9704266952348471,
      "eval_loss": 0.1764293909072876,
      "eval_precision": 0.9710387924576857,
      "eval_recall": 0.9704258675078864,
      "eval_runtime": 14.6294,
      "eval_samples_per_second": 346.699,
      "eval_steps_per_second": 21.669,
      "step": 1899
    }
  ],
  "logging_steps": 10,
  "max_steps": 1899,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7993457212661760.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}