{
  "best_metric": 0.986134647974118,
  "best_model_checkpoint": "beit-base-patch16-224-pt22k-ft22k-finetuned-lora-medmnistv2/checkpoint-4864",
  "epoch": 9.990749306197966,
  "eval_steps": 500,
  "global_step": 5400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "grad_norm": 3.66642165184021,
      "learning_rate": 0.004990740740740741,
      "loss": 1.7617,
      "step": 10
    },
    {
      "epoch": 0.04,
      "grad_norm": 5.545452117919922,
      "learning_rate": 0.004982407407407408,
      "loss": 1.1989,
      "step": 20
    },
    {
      "epoch": 0.06,
      "grad_norm": 5.292081356048584,
      "learning_rate": 0.004973148148148148,
      "loss": 1.0735,
      "step": 30
    },
    {
      "epoch": 0.07,
      "grad_norm": 3.0206360816955566,
      "learning_rate": 0.004963888888888889,
      "loss": 1.1936,
      "step": 40
    },
    {
      "epoch": 0.09,
      "grad_norm": 3.947781562805176,
      "learning_rate": 0.004954629629629629,
      "loss": 1.0373,
      "step": 50
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.839845895767212,
      "learning_rate": 0.004945370370370371,
      "loss": 0.9221,
      "step": 60
    },
    {
      "epoch": 0.13,
      "grad_norm": 2.843251943588257,
      "learning_rate": 0.004936111111111112,
      "loss": 0.8724,
      "step": 70
    },
    {
      "epoch": 0.15,
      "grad_norm": 3.0122690200805664,
      "learning_rate": 0.0049268518518518515,
      "loss": 0.9201,
      "step": 80
    },
    {
      "epoch": 0.17,
      "grad_norm": 3.403461456298828,
      "learning_rate": 0.004917592592592593,
      "loss": 0.9221,
      "step": 90
    },
    {
      "epoch": 0.19,
      "grad_norm": 2.4073173999786377,
      "learning_rate": 0.004908333333333333,
      "loss": 0.8766,
      "step": 100
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.793009877204895,
      "learning_rate": 0.004899074074074074,
      "loss": 0.8742,
      "step": 110
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.8483041524887085,
      "learning_rate": 0.004889814814814815,
      "loss": 0.8097,
      "step": 120
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.997031807899475,
      "learning_rate": 0.0048805555555555555,
      "loss": 0.888,
      "step": 130
    },
    {
      "epoch": 0.26,
      "grad_norm": 3.958829641342163,
      "learning_rate": 0.004871296296296296,
      "loss": 0.8276,
      "step": 140
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.8835649490356445,
      "learning_rate": 0.004862037037037037,
      "loss": 0.8469,
      "step": 150
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.2343573570251465,
      "learning_rate": 0.004852777777777778,
      "loss": 0.8354,
      "step": 160
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.8973902463912964,
      "learning_rate": 0.004843518518518519,
      "loss": 0.7974,
      "step": 170
    },
    {
      "epoch": 0.33,
      "grad_norm": 2.303403854370117,
      "learning_rate": 0.0048342592592592595,
      "loss": 0.7531,
      "step": 180
    },
    {
      "epoch": 0.35,
      "grad_norm": 2.833897352218628,
      "learning_rate": 0.004825,
      "loss": 0.7917,
      "step": 190
    },
    {
      "epoch": 0.37,
      "grad_norm": 2.833991527557373,
      "learning_rate": 0.004815740740740741,
      "loss": 0.9165,
      "step": 200
    },
    {
      "epoch": 0.39,
      "grad_norm": 2.109811305999756,
      "learning_rate": 0.004806481481481482,
      "loss": 0.7347,
      "step": 210
    },
    {
      "epoch": 0.41,
      "grad_norm": 4.833127498626709,
      "learning_rate": 0.004797222222222223,
      "loss": 0.8852,
      "step": 220
    },
    {
      "epoch": 0.43,
      "grad_norm": 6.393899440765381,
      "learning_rate": 0.004787962962962963,
      "loss": 0.7416,
      "step": 230
    },
    {
      "epoch": 0.44,
      "grad_norm": 2.243924856185913,
      "learning_rate": 0.004778703703703704,
      "loss": 0.7616,
      "step": 240
    },
    {
      "epoch": 0.46,
      "grad_norm": 2.55934739112854,
      "learning_rate": 0.004769444444444444,
      "loss": 0.6881,
      "step": 250
    },
    {
      "epoch": 0.48,
      "grad_norm": 2.707200765609741,
      "learning_rate": 0.004760185185185185,
      "loss": 0.7597,
      "step": 260
    },
    {
      "epoch": 0.5,
      "grad_norm": 3.238736867904663,
      "learning_rate": 0.004750925925925926,
      "loss": 0.6598,
      "step": 270
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.6335750818252563,
      "learning_rate": 0.004741666666666667,
      "loss": 0.6988,
      "step": 280
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.6415876150131226,
      "learning_rate": 0.004732407407407407,
      "loss": 0.7615,
      "step": 290
    },
    {
      "epoch": 0.56,
      "grad_norm": 2.09512996673584,
      "learning_rate": 0.004723148148148148,
      "loss": 0.7345,
      "step": 300
    },
    {
      "epoch": 0.57,
      "grad_norm": 2.1141433715820312,
      "learning_rate": 0.004713888888888889,
      "loss": 0.7503,
      "step": 310
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.1782424449920654,
      "learning_rate": 0.00470462962962963,
      "loss": 0.6997,
      "step": 320
    },
    {
      "epoch": 0.61,
      "grad_norm": 2.7585694789886475,
      "learning_rate": 0.004695370370370371,
      "loss": 0.7703,
      "step": 330
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.3906846046447754,
      "learning_rate": 0.004686111111111111,
      "loss": 0.6595,
      "step": 340
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.9824657440185547,
      "learning_rate": 0.004676851851851852,
      "loss": 0.612,
      "step": 350
    },
    {
      "epoch": 0.67,
      "grad_norm": 3.896270990371704,
      "learning_rate": 0.004667592592592593,
      "loss": 0.8571,
      "step": 360
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.860967993736267,
      "learning_rate": 0.004658333333333333,
      "loss": 0.774,
      "step": 370
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.4278061389923096,
      "learning_rate": 0.004649074074074074,
      "loss": 0.7693,
      "step": 380
    },
    {
      "epoch": 0.72,
      "grad_norm": 4.371230125427246,
      "learning_rate": 0.004639814814814815,
      "loss": 0.6989,
      "step": 390
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.6411677598953247,
      "learning_rate": 0.004630555555555555,
      "loss": 0.8566,
      "step": 400
    },
    {
      "epoch": 0.76,
      "grad_norm": 3.412655830383301,
      "learning_rate": 0.004621296296296296,
      "loss": 0.6902,
      "step": 410
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.9510608911514282,
      "learning_rate": 0.004612037037037038,
      "loss": 0.8284,
      "step": 420
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.8800582885742188,
      "learning_rate": 0.004602777777777778,
      "loss": 0.6096,
      "step": 430
    },
    {
      "epoch": 0.81,
      "grad_norm": 2.7876181602478027,
      "learning_rate": 0.0045935185185185185,
      "loss": 0.7293,
      "step": 440
    },
    {
      "epoch": 0.83,
      "grad_norm": 2.0592384338378906,
      "learning_rate": 0.004584259259259259,
      "loss": 0.756,
      "step": 450
    },
    {
      "epoch": 0.85,
      "grad_norm": 3.5372824668884277,
      "learning_rate": 0.004575,
      "loss": 0.6865,
      "step": 460
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.7686892747879028,
      "learning_rate": 0.004565740740740741,
      "loss": 0.7477,
      "step": 470
    },
    {
      "epoch": 0.89,
      "grad_norm": 2.8619141578674316,
      "learning_rate": 0.004556481481481482,
      "loss": 0.6861,
      "step": 480
    },
    {
      "epoch": 0.91,
      "grad_norm": 2.4887492656707764,
      "learning_rate": 0.0045472222222222225,
      "loss": 0.7623,
      "step": 490
    },
    {
      "epoch": 0.93,
      "grad_norm": 1.9150561094284058,
      "learning_rate": 0.004537962962962963,
      "loss": 0.7462,
      "step": 500
    },
    {
      "epoch": 0.94,
      "grad_norm": 2.1251327991485596,
      "learning_rate": 0.004528703703703704,
      "loss": 0.673,
      "step": 510
    },
    {
      "epoch": 0.96,
      "grad_norm": 2.338467597961426,
      "learning_rate": 0.004519444444444444,
      "loss": 0.6898,
      "step": 520
    },
    {
      "epoch": 0.98,
      "grad_norm": 2.199749708175659,
      "learning_rate": 0.004510185185185186,
      "loss": 0.6776,
      "step": 530
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.5489593744277954,
      "learning_rate": 0.0045009259259259264,
      "loss": 0.6786,
      "step": 540
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9339084886766292,
      "eval_f1": 0.938537564429046,
      "eval_loss": 0.17764028906822205,
      "eval_precision": 0.9507286953627705,
      "eval_recall": 0.934120062770795,
      "eval_runtime": 54.7942,
      "eval_samples_per_second": 118.462,
      "eval_steps_per_second": 7.41,
      "step": 540
    },
    {
      "epoch": 1.02,
      "grad_norm": 2.148123264312744,
      "learning_rate": 0.004491666666666666,
      "loss": 0.592,
      "step": 550
    },
    {
      "epoch": 1.04,
      "grad_norm": 3.1778488159179688,
      "learning_rate": 0.004482407407407407,
      "loss": 0.93,
      "step": 560
    },
    {
      "epoch": 1.05,
      "grad_norm": 1.9821240901947021,
      "learning_rate": 0.004473148148148149,
      "loss": 0.8283,
      "step": 570
    },
    {
      "epoch": 1.07,
      "grad_norm": 1.9005764722824097,
      "learning_rate": 0.004463888888888889,
      "loss": 0.7101,
      "step": 580
    },
    {
      "epoch": 1.09,
      "grad_norm": 2.126354217529297,
      "learning_rate": 0.0044546296296296296,
      "loss": 0.6671,
      "step": 590
    },
    {
      "epoch": 1.11,
      "grad_norm": 2.0805599689483643,
      "learning_rate": 0.00444537037037037,
      "loss": 0.6665,
      "step": 600
    },
    {
      "epoch": 1.13,
      "grad_norm": 1.5277152061462402,
      "learning_rate": 0.004436111111111111,
      "loss": 0.7092,
      "step": 610
    },
    {
      "epoch": 1.15,
      "grad_norm": 4.088949680328369,
      "learning_rate": 0.004426851851851852,
      "loss": 0.8679,
      "step": 620
    },
    {
      "epoch": 1.17,
      "grad_norm": 4.5228376388549805,
      "learning_rate": 0.004417592592592593,
      "loss": 0.7648,
      "step": 630
    },
    {
      "epoch": 1.18,
      "grad_norm": 4.287468910217285,
      "learning_rate": 0.0044083333333333335,
      "loss": 0.725,
      "step": 640
    },
    {
      "epoch": 1.2,
      "grad_norm": 3.0292155742645264,
      "learning_rate": 0.004399074074074074,
      "loss": 0.7416,
      "step": 650
    },
    {
      "epoch": 1.22,
      "grad_norm": 2.364569664001465,
      "learning_rate": 0.004389814814814815,
      "loss": 0.7576,
      "step": 660
    },
    {
      "epoch": 1.24,
      "grad_norm": 1.8046070337295532,
      "learning_rate": 0.004380555555555555,
      "loss": 0.6917,
      "step": 670
    },
    {
      "epoch": 1.26,
      "grad_norm": 3.8957250118255615,
      "learning_rate": 0.004371296296296297,
      "loss": 0.6671,
      "step": 680
    },
    {
      "epoch": 1.28,
      "grad_norm": 2.4471426010131836,
      "learning_rate": 0.0043620370370370375,
      "loss": 0.6749,
      "step": 690
    },
    {
      "epoch": 1.3,
      "grad_norm": 1.9041227102279663,
      "learning_rate": 0.0043527777777777775,
      "loss": 0.7423,
      "step": 700
    },
    {
      "epoch": 1.31,
      "grad_norm": 3.001101016998291,
      "learning_rate": 0.004343518518518519,
      "loss": 0.7775,
      "step": 710
    },
    {
      "epoch": 1.33,
      "grad_norm": 2.1323304176330566,
      "learning_rate": 0.004334259259259259,
      "loss": 0.6961,
      "step": 720
    },
    {
      "epoch": 1.35,
      "grad_norm": 3.0839684009552,
      "learning_rate": 0.004325,
      "loss": 0.6791,
      "step": 730
    },
    {
      "epoch": 1.37,
      "grad_norm": 1.924004316329956,
      "learning_rate": 0.004315740740740741,
      "loss": 0.663,
      "step": 740
    },
    {
      "epoch": 1.39,
      "grad_norm": 1.854284644126892,
      "learning_rate": 0.0043064814814814814,
      "loss": 0.706,
      "step": 750
    },
    {
      "epoch": 1.41,
      "grad_norm": 2.4816770553588867,
      "learning_rate": 0.004297222222222222,
      "loss": 0.6623,
      "step": 760
    },
    {
      "epoch": 1.42,
      "grad_norm": 1.8469676971435547,
      "learning_rate": 0.004287962962962963,
      "loss": 0.6025,
      "step": 770
    },
    {
      "epoch": 1.44,
      "grad_norm": 1.6703932285308838,
      "learning_rate": 0.004278703703703704,
      "loss": 0.7209,
      "step": 780
    },
    {
      "epoch": 1.46,
      "grad_norm": 1.3802216053009033,
      "learning_rate": 0.004269444444444445,
      "loss": 0.6823,
      "step": 790
    },
    {
      "epoch": 1.48,
      "grad_norm": 2.1210741996765137,
      "learning_rate": 0.004260185185185185,
      "loss": 0.6577,
      "step": 800
    },
    {
      "epoch": 1.5,
      "grad_norm": 1.841590404510498,
      "learning_rate": 0.004250925925925926,
      "loss": 0.6049,
      "step": 810
    },
    {
      "epoch": 1.52,
      "grad_norm": 2.4515421390533447,
      "learning_rate": 0.004241666666666667,
      "loss": 0.6555,
      "step": 820
    },
    {
      "epoch": 1.54,
      "grad_norm": 2.086291551589966,
      "learning_rate": 0.004232407407407408,
      "loss": 0.7836,
      "step": 830
    },
    {
      "epoch": 1.55,
      "grad_norm": 1.714977741241455,
      "learning_rate": 0.004223148148148149,
      "loss": 0.6868,
      "step": 840
    },
    {
      "epoch": 1.57,
      "grad_norm": 1.9800652265548706,
      "learning_rate": 0.0042138888888888885,
      "loss": 0.7313,
      "step": 850
    },
    {
      "epoch": 1.59,
      "grad_norm": 2.726518154144287,
      "learning_rate": 0.00420462962962963,
      "loss": 0.6946,
      "step": 860
    },
    {
      "epoch": 1.61,
      "grad_norm": 2.3700504302978516,
      "learning_rate": 0.00419537037037037,
      "loss": 0.7765,
      "step": 870
    },
    {
      "epoch": 1.63,
      "grad_norm": 3.294623851776123,
      "learning_rate": 0.004186111111111111,
      "loss": 0.7001,
      "step": 880
    },
    {
      "epoch": 1.65,
      "grad_norm": 1.3391900062561035,
      "learning_rate": 0.004176851851851852,
      "loss": 0.7045,
      "step": 890
    },
    {
      "epoch": 1.67,
      "grad_norm": 3.3717267513275146,
      "learning_rate": 0.0041675925925925925,
      "loss": 0.7396,
      "step": 900
    },
    {
      "epoch": 1.68,
      "grad_norm": 2.7791171073913574,
      "learning_rate": 0.004158333333333333,
      "loss": 0.6946,
      "step": 910
    },
    {
      "epoch": 1.7,
      "grad_norm": 2.9420201778411865,
      "learning_rate": 0.004149074074074074,
      "loss": 0.7215,
      "step": 920
    },
    {
      "epoch": 1.72,
      "grad_norm": 2.337606191635132,
      "learning_rate": 0.004139814814814815,
      "loss": 0.6967,
      "step": 930
    },
    {
      "epoch": 1.74,
      "grad_norm": 2.43172025680542,
      "learning_rate": 0.004130555555555556,
      "loss": 0.7989,
      "step": 940
    },
    {
      "epoch": 1.76,
      "grad_norm": 1.683398723602295,
      "learning_rate": 0.0041212962962962965,
      "loss": 0.6185,
      "step": 950
    },
    {
      "epoch": 1.78,
      "grad_norm": 2.1114211082458496,
      "learning_rate": 0.004112037037037037,
      "loss": 0.7013,
      "step": 960
    },
    {
      "epoch": 1.79,
      "grad_norm": 2.566915512084961,
      "learning_rate": 0.004102777777777778,
      "loss": 0.6091,
      "step": 970
    },
    {
      "epoch": 1.81,
      "grad_norm": 4.286575794219971,
      "learning_rate": 0.004093518518518519,
      "loss": 0.7474,
      "step": 980
    },
    {
      "epoch": 1.83,
      "grad_norm": 1.9286409616470337,
      "learning_rate": 0.004084259259259259,
      "loss": 0.7089,
      "step": 990
    },
    {
      "epoch": 1.85,
      "grad_norm": 2.5214929580688477,
      "learning_rate": 0.004075,
      "loss": 0.7412,
      "step": 1000
    },
    {
      "epoch": 1.87,
      "grad_norm": 2.4169070720672607,
      "learning_rate": 0.004065740740740741,
      "loss": 0.826,
      "step": 1010
    },
    {
      "epoch": 1.89,
      "grad_norm": 2.7352206707000732,
      "learning_rate": 0.004056481481481481,
      "loss": 0.7868,
      "step": 1020
    },
    {
      "epoch": 1.91,
      "grad_norm": 1.969298243522644,
      "learning_rate": 0.004047222222222222,
      "loss": 0.6795,
      "step": 1030
    },
    {
      "epoch": 1.92,
      "grad_norm": 3.015155792236328,
      "learning_rate": 0.004037962962962964,
      "loss": 0.7852,
      "step": 1040
    },
    {
      "epoch": 1.94,
      "grad_norm": 2.2428641319274902,
      "learning_rate": 0.004028703703703704,
      "loss": 0.7271,
      "step": 1050
    },
    {
      "epoch": 1.96,
      "grad_norm": 2.1830852031707764,
      "learning_rate": 0.004019444444444444,
      "loss": 0.6171,
      "step": 1060
    },
    {
      "epoch": 1.98,
      "grad_norm": 4.2572455406188965,
      "learning_rate": 0.004010185185185185,
      "loss": 0.715,
      "step": 1070
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.7257713079452515,
      "learning_rate": 0.004000925925925926,
      "loss": 0.7397,
      "step": 1080
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9406871052226159,
      "eval_f1": 0.9414537311136683,
      "eval_loss": 0.17834338545799255,
      "eval_precision": 0.9539065276276052,
      "eval_recall": 0.9346301030411137,
      "eval_runtime": 55.3829,
      "eval_samples_per_second": 117.202,
      "eval_steps_per_second": 7.331,
      "step": 1081
    },
    {
      "epoch": 2.02,
      "grad_norm": 1.4031952619552612,
      "learning_rate": 0.003991666666666667,
      "loss": 0.652,
      "step": 1090
    },
    {
      "epoch": 2.04,
      "grad_norm": 2.210618257522583,
      "learning_rate": 0.003982407407407408,
      "loss": 0.6434,
      "step": 1100
    },
    {
      "epoch": 2.05,
      "grad_norm": 2.780142068862915,
      "learning_rate": 0.0039731481481481475,
      "loss": 0.6638,
      "step": 1110
    },
    {
      "epoch": 2.07,
      "grad_norm": 2.0362918376922607,
      "learning_rate": 0.003963888888888889,
      "loss": 0.6194,
      "step": 1120
    },
    {
      "epoch": 2.09,
      "grad_norm": 2.7346765995025635,
      "learning_rate": 0.00395462962962963,
      "loss": 0.6721,
      "step": 1130
    },
    {
      "epoch": 2.11,
      "grad_norm": 1.8441007137298584,
      "learning_rate": 0.00394537037037037,
      "loss": 0.7129,
      "step": 1140
    },
    {
      "epoch": 2.13,
      "grad_norm": 2.510021448135376,
      "learning_rate": 0.003936111111111112,
      "loss": 0.669,
      "step": 1150
    },
    {
      "epoch": 2.15,
      "grad_norm": 2.9301140308380127,
      "learning_rate": 0.003926851851851852,
      "loss": 0.6342,
      "step": 1160
    },
    {
      "epoch": 2.16,
      "grad_norm": 1.851173996925354,
      "learning_rate": 0.003917592592592592,
      "loss": 0.6602,
      "step": 1170
    },
    {
      "epoch": 2.18,
      "grad_norm": 2.490886926651001,
      "learning_rate": 0.003908333333333333,
      "loss": 0.6195,
      "step": 1180
    },
    {
      "epoch": 2.2,
      "grad_norm": 1.5079678297042847,
      "learning_rate": 0.0038990740740740743,
      "loss": 0.8126,
      "step": 1190
    },
    {
      "epoch": 2.22,
      "grad_norm": 1.9084036350250244,
      "learning_rate": 0.003889814814814815,
      "loss": 0.6769,
      "step": 1200
    },
    {
      "epoch": 2.24,
      "grad_norm": 1.8284987211227417,
      "learning_rate": 0.0038814814814814814,
      "loss": 0.7317,
      "step": 1210
    },
    {
      "epoch": 2.26,
      "grad_norm": 3.351156234741211,
      "learning_rate": 0.0038722222222222226,
      "loss": 0.7049,
      "step": 1220
    },
    {
      "epoch": 2.28,
      "grad_norm": 2.314993143081665,
      "learning_rate": 0.003862962962962963,
      "loss": 0.6291,
      "step": 1230
    },
    {
      "epoch": 2.29,
      "grad_norm": 2.27850341796875,
      "learning_rate": 0.003853703703703704,
      "loss": 0.6749,
      "step": 1240
    },
    {
      "epoch": 2.31,
      "grad_norm": 2.4541099071502686,
      "learning_rate": 0.003844444444444444,
      "loss": 0.7365,
      "step": 1250
    },
    {
      "epoch": 2.33,
      "grad_norm": 1.6301252841949463,
      "learning_rate": 0.0038351851851851854,
      "loss": 0.6236,
      "step": 1260
    },
    {
      "epoch": 2.35,
      "grad_norm": 2.035531520843506,
      "learning_rate": 0.003825925925925926,
      "loss": 0.7176,
      "step": 1270
    },
    {
      "epoch": 2.37,
      "grad_norm": 2.2524566650390625,
      "learning_rate": 0.0038166666666666666,
      "loss": 0.7119,
      "step": 1280
    },
    {
      "epoch": 2.39,
      "grad_norm": 1.778144121170044,
      "learning_rate": 0.003807407407407408,
      "loss": 0.7021,
      "step": 1290
    },
    {
      "epoch": 2.41,
      "grad_norm": 1.8808122873306274,
      "learning_rate": 0.003798148148148148,
      "loss": 0.6876,
      "step": 1300
    },
    {
      "epoch": 2.42,
      "grad_norm": 2.720100164413452,
      "learning_rate": 0.003788888888888889,
      "loss": 0.7559,
      "step": 1310
    },
    {
      "epoch": 2.44,
      "grad_norm": 1.9608194828033447,
      "learning_rate": 0.0037796296296296297,
      "loss": 0.6799,
      "step": 1320
    },
    {
      "epoch": 2.46,
      "grad_norm": 1.8149750232696533,
      "learning_rate": 0.0037703703703703705,
      "loss": 0.6701,
      "step": 1330
    },
    {
      "epoch": 2.48,
      "grad_norm": 1.8840572834014893,
      "learning_rate": 0.0037611111111111113,
      "loss": 0.7146,
      "step": 1340
    },
    {
      "epoch": 2.5,
      "grad_norm": 1.7513312101364136,
      "learning_rate": 0.0037518518518518517,
      "loss": 0.6712,
      "step": 1350
    },
    {
      "epoch": 2.52,
      "grad_norm": 1.748304009437561,
      "learning_rate": 0.0037425925925925925,
      "loss": 0.7049,
      "step": 1360
    },
    {
      "epoch": 2.53,
      "grad_norm": 1.2645372152328491,
      "learning_rate": 0.0037333333333333337,
      "loss": 0.6088,
      "step": 1370
    },
    {
      "epoch": 2.55,
      "grad_norm": 3.14270281791687,
      "learning_rate": 0.003724074074074074,
      "loss": 0.7163,
      "step": 1380
    },
    {
      "epoch": 2.57,
      "grad_norm": 2.0332484245300293,
      "learning_rate": 0.003714814814814815,
      "loss": 0.7047,
      "step": 1390
    },
    {
      "epoch": 2.59,
      "grad_norm": 1.6158325672149658,
      "learning_rate": 0.0037055555555555557,
      "loss": 0.6448,
      "step": 1400
    },
    {
      "epoch": 2.61,
      "grad_norm": 1.9182847738265991,
      "learning_rate": 0.0036962962962962965,
      "loss": 0.711,
      "step": 1410
    },
    {
      "epoch": 2.63,
      "grad_norm": 2.140184164047241,
      "learning_rate": 0.0036870370370370373,
      "loss": 0.6617,
      "step": 1420
    },
    {
      "epoch": 2.65,
      "grad_norm": 1.746243953704834,
      "learning_rate": 0.0036777777777777776,
      "loss": 0.7103,
      "step": 1430
    },
    {
      "epoch": 2.66,
      "grad_norm": 2.6956822872161865,
      "learning_rate": 0.003668518518518519,
      "loss": 0.7163,
      "step": 1440
    },
    {
      "epoch": 2.68,
      "grad_norm": 2.7663381099700928,
      "learning_rate": 0.0036592592592592592,
      "loss": 0.6509,
      "step": 1450
    },
    {
      "epoch": 2.7,
      "grad_norm": 2.410922050476074,
      "learning_rate": 0.00365,
      "loss": 0.6046,
      "step": 1460
    },
    {
      "epoch": 2.72,
      "grad_norm": 1.8373881578445435,
      "learning_rate": 0.0036407407407407404,
      "loss": 0.701,
      "step": 1470
    },
    {
      "epoch": 2.74,
      "grad_norm": 1.8466787338256836,
      "learning_rate": 0.0036314814814814816,
      "loss": 0.6391,
      "step": 1480
    },
    {
      "epoch": 2.76,
      "grad_norm": 2.0422921180725098,
      "learning_rate": 0.0036222222222222224,
      "loss": 0.7001,
      "step": 1490
    },
    {
      "epoch": 2.78,
      "grad_norm": 1.739327311515808,
      "learning_rate": 0.003612962962962963,
      "loss": 0.6194,
      "step": 1500
    },
    {
      "epoch": 2.79,
      "grad_norm": 2.094775915145874,
      "learning_rate": 0.003603703703703704,
      "loss": 0.5824,
      "step": 1510
    },
    {
      "epoch": 2.81,
      "grad_norm": 2.5036680698394775,
      "learning_rate": 0.003594444444444445,
      "loss": 0.6477,
      "step": 1520
    },
    {
      "epoch": 2.83,
      "grad_norm": 1.7037166357040405,
      "learning_rate": 0.003585185185185185,
      "loss": 0.6822,
      "step": 1530
    },
    {
      "epoch": 2.85,
      "grad_norm": 2.193714141845703,
      "learning_rate": 0.003575925925925926,
      "loss": 0.7301,
      "step": 1540
    },
    {
      "epoch": 2.87,
      "grad_norm": 4.449968338012695,
      "learning_rate": 0.0035666666666666668,
      "loss": 0.7114,
      "step": 1550
    },
    {
      "epoch": 2.89,
      "grad_norm": 1.640833854675293,
      "learning_rate": 0.0035574074074074076,
      "loss": 0.5631,
      "step": 1560
    },
    {
      "epoch": 2.9,
      "grad_norm": 2.4068825244903564,
      "learning_rate": 0.003548148148148148,
      "loss": 0.5896,
      "step": 1570
    },
    {
      "epoch": 2.92,
      "grad_norm": 2.7234010696411133,
      "learning_rate": 0.0035388888888888887,
      "loss": 0.7563,
      "step": 1580
    },
    {
      "epoch": 2.94,
      "grad_norm": 1.8094590902328491,
      "learning_rate": 0.00352962962962963,
      "loss": 0.7116,
      "step": 1590
    },
    {
      "epoch": 2.96,
      "grad_norm": 2.4650044441223145,
      "learning_rate": 0.0035203703703703703,
      "loss": 0.6583,
      "step": 1600
    },
    {
      "epoch": 2.98,
      "grad_norm": 1.8076032400131226,
      "learning_rate": 0.003511111111111111,
      "loss": 0.5859,
      "step": 1610
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.924779176712036,
      "learning_rate": 0.0035018518518518523,
      "loss": 0.7151,
      "step": 1620
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9551686951163149,
      "eval_f1": 0.957151302239187,
      "eval_loss": 0.12974682450294495,
      "eval_precision": 0.961140704317014,
      "eval_recall": 0.9555174635857528,
      "eval_runtime": 55.4288,
      "eval_samples_per_second": 117.105,
      "eval_steps_per_second": 7.325,
      "step": 1621
    },
    {
      "epoch": 3.02,
      "grad_norm": 2.4308605194091797,
      "learning_rate": 0.0034925925925925927,
      "loss": 0.6032,
      "step": 1630
    },
    {
      "epoch": 3.03,
      "grad_norm": 1.6040972471237183,
      "learning_rate": 0.0034833333333333335,
      "loss": 0.6163,
      "step": 1640
    },
    {
      "epoch": 3.05,
      "grad_norm": 1.7093302011489868,
      "learning_rate": 0.003474074074074074,
      "loss": 0.6484,
      "step": 1650
    },
    {
      "epoch": 3.07,
      "grad_norm": 2.0212676525115967,
      "learning_rate": 0.003464814814814815,
      "loss": 0.5747,
      "step": 1660
    },
    {
      "epoch": 3.09,
      "grad_norm": 3.725684642791748,
      "learning_rate": 0.0034555555555555555,
      "loss": 0.6488,
      "step": 1670
    },
    {
      "epoch": 3.11,
      "grad_norm": 2.5696334838867188,
      "learning_rate": 0.0034462962962962963,
      "loss": 0.7223,
      "step": 1680
    },
    {
      "epoch": 3.13,
      "grad_norm": 1.7049779891967773,
      "learning_rate": 0.003437037037037037,
      "loss": 0.641,
      "step": 1690
    },
    {
      "epoch": 3.15,
      "grad_norm": 2.035477638244629,
      "learning_rate": 0.003427777777777778,
      "loss": 0.6237,
      "step": 1700
    },
    {
      "epoch": 3.16,
      "grad_norm": 1.8789253234863281,
      "learning_rate": 0.0034185185185185187,
      "loss": 0.6312,
      "step": 1710
    },
    {
      "epoch": 3.18,
      "grad_norm": 2.2245302200317383,
      "learning_rate": 0.003409259259259259,
      "loss": 0.5985,
      "step": 1720
    },
    {
      "epoch": 3.2,
      "grad_norm": 2.150331735610962,
      "learning_rate": 0.0034000000000000002,
      "loss": 0.612,
      "step": 1730
    },
    {
      "epoch": 3.22,
      "grad_norm": 1.986059546470642,
      "learning_rate": 0.003390740740740741,
      "loss": 0.6001,
      "step": 1740
    },
    {
      "epoch": 3.24,
      "grad_norm": 1.4887031316757202,
      "learning_rate": 0.0033814814814814814,
      "loss": 0.6574,
      "step": 1750
    },
    {
      "epoch": 3.26,
      "grad_norm": 1.1965304613113403,
      "learning_rate": 0.003372222222222222,
      "loss": 0.6766,
      "step": 1760
    },
    {
      "epoch": 3.27,
      "grad_norm": 1.564257025718689,
      "learning_rate": 0.0033629629629629634,
      "loss": 0.6955,
      "step": 1770
    },
    {
      "epoch": 3.29,
      "grad_norm": 5.508277416229248,
      "learning_rate": 0.003353703703703704,
      "loss": 0.6097,
      "step": 1780
    },
    {
      "epoch": 3.31,
      "grad_norm": 1.9245549440383911,
      "learning_rate": 0.0033444444444444446,
      "loss": 0.6515,
      "step": 1790
    },
    {
      "epoch": 3.33,
      "grad_norm": 2.3552732467651367,
      "learning_rate": 0.003335185185185185,
      "loss": 0.6401,
      "step": 1800
    },
    {
      "epoch": 3.35,
      "grad_norm": 1.622826099395752,
      "learning_rate": 0.003325925925925926,
      "loss": 0.7,
      "step": 1810
    },
    {
      "epoch": 3.37,
      "grad_norm": 1.6205705404281616,
      "learning_rate": 0.0033166666666666665,
      "loss": 0.5968,
      "step": 1820
    },
    {
      "epoch": 3.39,
      "grad_norm": 1.9269992113113403,
      "learning_rate": 0.0033074074074074073,
      "loss": 0.6,
      "step": 1830
    },
    {
      "epoch": 3.4,
      "grad_norm": 2.3607101440429688,
      "learning_rate": 0.0032981481481481486,
      "loss": 0.5655,
      "step": 1840
    },
    {
      "epoch": 3.42,
      "grad_norm": 1.4156101942062378,
      "learning_rate": 0.003288888888888889,
      "loss": 0.6071,
      "step": 1850
    },
    {
      "epoch": 3.44,
      "grad_norm": 2.652838706970215,
      "learning_rate": 0.0032796296296296297,
      "loss": 0.5959,
      "step": 1860
    },
    {
      "epoch": 3.46,
      "grad_norm": 1.7326959371566772,
      "learning_rate": 0.00327037037037037,
      "loss": 0.7051,
      "step": 1870
    },
    {
      "epoch": 3.48,
      "grad_norm": 1.4709230661392212,
      "learning_rate": 0.0032611111111111113,
      "loss": 0.68,
      "step": 1880
    },
    {
      "epoch": 3.5,
      "grad_norm": 2.1479110717773438,
      "learning_rate": 0.003251851851851852,
      "loss": 0.6491,
      "step": 1890
    },
    {
      "epoch": 3.52,
      "grad_norm": 2.462580442428589,
      "learning_rate": 0.0032425925925925925,
      "loss": 0.6129,
      "step": 1900
    },
    {
      "epoch": 3.53,
      "grad_norm": 2.8849940299987793,
      "learning_rate": 0.0032333333333333333,
      "loss": 0.6553,
      "step": 1910
    },
    {
      "epoch": 3.55,
      "grad_norm": 1.6244118213653564,
      "learning_rate": 0.003224074074074074,
      "loss": 0.5845,
      "step": 1920
    },
    {
      "epoch": 3.57,
      "grad_norm": 2.803816795349121,
      "learning_rate": 0.003214814814814815,
      "loss": 0.6695,
      "step": 1930
    },
    {
      "epoch": 3.59,
      "grad_norm": 2.4138996601104736,
      "learning_rate": 0.0032055555555555552,
      "loss": 0.707,
      "step": 1940
    },
    {
      "epoch": 3.61,
      "grad_norm": 1.1193430423736572,
      "learning_rate": 0.0031962962962962965,
      "loss": 0.6003,
      "step": 1950
    },
    {
      "epoch": 3.63,
      "grad_norm": 1.8593242168426514,
      "learning_rate": 0.0031870370370370373,
      "loss": 0.5724,
      "step": 1960
    },
    {
      "epoch": 3.64,
      "grad_norm": 2.084005117416382,
      "learning_rate": 0.0031777777777777776,
      "loss": 0.6246,
      "step": 1970
    },
    {
      "epoch": 3.66,
      "grad_norm": 1.32781982421875,
      "learning_rate": 0.0031685185185185184,
      "loss": 0.6075,
      "step": 1980
    },
    {
      "epoch": 3.68,
      "grad_norm": 2.452892303466797,
      "learning_rate": 0.0031592592592592597,
      "loss": 0.6104,
      "step": 1990
    },
    {
      "epoch": 3.7,
      "grad_norm": 1.6961791515350342,
      "learning_rate": 0.00315,
      "loss": 0.5483,
      "step": 2000
    },
    {
      "epoch": 3.72,
      "grad_norm": 2.94992733001709,
      "learning_rate": 0.003140740740740741,
      "loss": 0.5403,
      "step": 2010
    },
    {
      "epoch": 3.74,
      "grad_norm": 2.3624625205993652,
      "learning_rate": 0.0031314814814814816,
      "loss": 0.6748,
      "step": 2020
    },
    {
      "epoch": 3.76,
      "grad_norm": 2.0412802696228027,
      "learning_rate": 0.0031222222222222224,
      "loss": 0.4809,
      "step": 2030
    },
    {
      "epoch": 3.77,
      "grad_norm": 1.5728198289871216,
      "learning_rate": 0.003112962962962963,
      "loss": 0.5755,
      "step": 2040
    },
    {
      "epoch": 3.79,
      "grad_norm": 1.3974460363388062,
      "learning_rate": 0.0031037037037037036,
      "loss": 0.5436,
      "step": 2050
    },
    {
      "epoch": 3.81,
      "grad_norm": 1.5604791641235352,
      "learning_rate": 0.003094444444444445,
      "loss": 0.6197,
      "step": 2060
    },
    {
      "epoch": 3.83,
      "grad_norm": 2.040276288986206,
      "learning_rate": 0.003085185185185185,
      "loss": 0.6062,
      "step": 2070
    },
    {
      "epoch": 3.85,
      "grad_norm": 3.173293113708496,
      "learning_rate": 0.003075925925925926,
      "loss": 0.717,
      "step": 2080
    },
    {
      "epoch": 3.87,
      "grad_norm": 2.018458843231201,
      "learning_rate": 0.0030666666666666663,
      "loss": 0.6418,
      "step": 2090
    },
    {
      "epoch": 3.89,
      "grad_norm": 1.7057709693908691,
      "learning_rate": 0.0030574074074074076,
      "loss": 0.5948,
      "step": 2100
    },
    {
      "epoch": 3.9,
      "grad_norm": 2.133119821548462,
      "learning_rate": 0.0030481481481481484,
      "loss": 0.6099,
      "step": 2110
    },
    {
      "epoch": 3.92,
      "grad_norm": 2.217524290084839,
      "learning_rate": 0.0030388888888888887,
      "loss": 0.6263,
      "step": 2120
    },
    {
      "epoch": 3.94,
      "grad_norm": 2.3708674907684326,
      "learning_rate": 0.00302962962962963,
      "loss": 0.5851,
      "step": 2130
    },
    {
      "epoch": 3.96,
      "grad_norm": 1.4289004802703857,
      "learning_rate": 0.0030203703703703707,
      "loss": 0.5971,
      "step": 2140
    },
    {
      "epoch": 3.98,
      "grad_norm": 2.463360071182251,
      "learning_rate": 0.003011111111111111,
      "loss": 0.5889,
      "step": 2150
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.5189619064331055,
      "learning_rate": 0.003001851851851852,
      "loss": 0.4964,
      "step": 2160
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.97350177168387,
      "eval_f1": 0.9729772233764705,
      "eval_loss": 0.07412911206483841,
      "eval_precision": 0.9765336155795084,
      "eval_recall": 0.9701628526053132,
      "eval_runtime": 55.303,
      "eval_samples_per_second": 117.372,
      "eval_steps_per_second": 7.341,
      "step": 2162
    },
    {
      "epoch": 4.01,
      "grad_norm": 2.313246965408325,
      "learning_rate": 0.0029925925925925927,
      "loss": 0.5979,
      "step": 2170
    },
    {
      "epoch": 4.03,
      "grad_norm": 1.554916501045227,
      "learning_rate": 0.0029833333333333335,
      "loss": 0.621,
      "step": 2180
    },
    {
      "epoch": 4.05,
      "grad_norm": 2.7611353397369385,
      "learning_rate": 0.002974074074074074,
      "loss": 0.6662,
      "step": 2190
    },
    {
      "epoch": 4.07,
      "grad_norm": 1.206809639930725,
      "learning_rate": 0.0029648148148148147,
      "loss": 0.6061,
      "step": 2200
    },
    {
      "epoch": 4.09,
      "grad_norm": 1.876120686531067,
      "learning_rate": 0.002955555555555556,
      "loss": 0.5656,
      "step": 2210
    },
    {
      "epoch": 4.11,
      "grad_norm": 2.117579936981201,
      "learning_rate": 0.0029462962962962963,
      "loss": 0.611,
      "step": 2220
    },
    {
      "epoch": 4.13,
      "grad_norm": 2.014042377471924,
      "learning_rate": 0.002937037037037037,
      "loss": 0.5893,
      "step": 2230
    },
    {
      "epoch": 4.14,
      "grad_norm": 1.6792641878128052,
      "learning_rate": 0.0029277777777777783,
      "loss": 0.5691,
      "step": 2240
    },
    {
      "epoch": 4.16,
      "grad_norm": 1.2664446830749512,
      "learning_rate": 0.0029185185185185186,
      "loss": 0.5729,
      "step": 2250
    },
    {
      "epoch": 4.18,
      "grad_norm": 3.1098408699035645,
      "learning_rate": 0.0029092592592592594,
      "loss": 0.6115,
      "step": 2260
    },
    {
      "epoch": 4.2,
      "grad_norm": 2.1186132431030273,
      "learning_rate": 0.0029,
      "loss": 0.596,
      "step": 2270
    },
    {
      "epoch": 4.22,
      "grad_norm": 1.8666993379592896,
      "learning_rate": 0.002890740740740741,
      "loss": 0.6021,
      "step": 2280
    },
    {
      "epoch": 4.24,
      "grad_norm": 1.655548334121704,
      "learning_rate": 0.0028814814814814814,
      "loss": 0.5928,
      "step": 2290
    },
    {
      "epoch": 4.26,
      "grad_norm": 2.1081762313842773,
      "learning_rate": 0.002872222222222222,
      "loss": 0.5683,
      "step": 2300
    },
    {
      "epoch": 4.27,
      "grad_norm": 1.4986019134521484,
      "learning_rate": 0.0028629629629629626,
      "loss": 0.562,
      "step": 2310
    },
    {
      "epoch": 4.29,
      "grad_norm": 2.6111056804656982,
      "learning_rate": 0.002853703703703704,
      "loss": 0.5804,
      "step": 2320
    },
    {
      "epoch": 4.31,
      "grad_norm": 1.8590542078018188,
      "learning_rate": 0.0028444444444444446,
      "loss": 0.5409,
      "step": 2330
    },
    {
      "epoch": 4.33,
      "grad_norm": 1.897940754890442,
      "learning_rate": 0.002835185185185185,
      "loss": 0.6299,
      "step": 2340
    },
    {
      "epoch": 4.35,
      "grad_norm": 1.9114952087402344,
      "learning_rate": 0.002825925925925926,
      "loss": 0.5333,
      "step": 2350
    },
    {
      "epoch": 4.37,
      "grad_norm": 2.015868663787842,
      "learning_rate": 0.002816666666666667,
      "loss": 0.5611,
      "step": 2360
    },
    {
      "epoch": 4.38,
      "grad_norm": 1.5808693170547485,
      "learning_rate": 0.0028074074074074073,
      "loss": 0.6212,
      "step": 2370
    },
    {
      "epoch": 4.4,
      "grad_norm": 1.8288767337799072,
      "learning_rate": 0.002798148148148148,
      "loss": 0.604,
      "step": 2380
    },
    {
      "epoch": 4.42,
      "grad_norm": 1.389092206954956,
      "learning_rate": 0.002788888888888889,
      "loss": 0.4657,
      "step": 2390
    },
    {
      "epoch": 4.44,
      "grad_norm": 2.551008701324463,
      "learning_rate": 0.0027796296296296297,
      "loss": 0.6214,
      "step": 2400
    },
    {
      "epoch": 4.46,
      "grad_norm": 2.26882266998291,
      "learning_rate": 0.0027703703703703705,
      "loss": 0.5511,
      "step": 2410
    },
    {
      "epoch": 4.48,
      "grad_norm": 2.3850460052490234,
      "learning_rate": 0.002761111111111111,
      "loss": 0.5768,
      "step": 2420
    },
    {
      "epoch": 4.5,
      "grad_norm": 1.9451379776000977,
      "learning_rate": 0.002751851851851852,
      "loss": 0.5732,
      "step": 2430
    },
    {
      "epoch": 4.51,
      "grad_norm": 1.6541566848754883,
      "learning_rate": 0.0027425925925925925,
      "loss": 0.549,
      "step": 2440
    },
    {
      "epoch": 4.53,
      "grad_norm": 1.3802433013916016,
      "learning_rate": 0.0027333333333333333,
      "loss": 0.5486,
      "step": 2450
    },
    {
      "epoch": 4.55,
      "grad_norm": 1.941901683807373,
      "learning_rate": 0.0027240740740740745,
      "loss": 0.6119,
      "step": 2460
    },
    {
      "epoch": 4.57,
      "grad_norm": 1.9203604459762573,
      "learning_rate": 0.002714814814814815,
      "loss": 0.632,
      "step": 2470
    },
    {
      "epoch": 4.59,
      "grad_norm": 2.170319080352783,
      "learning_rate": 0.0027055555555555557,
      "loss": 0.5819,
      "step": 2480
    },
    {
      "epoch": 4.61,
      "grad_norm": 1.8074395656585693,
      "learning_rate": 0.002696296296296296,
      "loss": 0.6101,
      "step": 2490
    },
    {
      "epoch": 4.63,
      "grad_norm": 2.010608196258545,
      "learning_rate": 0.0026870370370370373,
      "loss": 0.492,
      "step": 2500
    },
    {
      "epoch": 4.64,
      "grad_norm": 1.3907346725463867,
      "learning_rate": 0.002677777777777778,
      "loss": 0.5658,
      "step": 2510
    },
    {
      "epoch": 4.66,
      "grad_norm": 1.8057767152786255,
      "learning_rate": 0.0026685185185185184,
      "loss": 0.509,
      "step": 2520
    },
    {
      "epoch": 4.68,
      "grad_norm": 2.036869764328003,
      "learning_rate": 0.002659259259259259,
      "loss": 0.5711,
      "step": 2530
    },
    {
      "epoch": 4.7,
      "grad_norm": 1.673647165298462,
      "learning_rate": 0.00265,
      "loss": 0.5692,
      "step": 2540
    },
    {
      "epoch": 4.72,
      "grad_norm": 2.044961929321289,
      "learning_rate": 0.002640740740740741,
      "loss": 0.6153,
      "step": 2550
    },
    {
      "epoch": 4.74,
      "grad_norm": 1.3360215425491333,
      "learning_rate": 0.002631481481481481,
      "loss": 0.5785,
      "step": 2560
    },
    {
      "epoch": 4.75,
      "grad_norm": 2.4942688941955566,
      "learning_rate": 0.0026222222222222224,
      "loss": 0.5423,
      "step": 2570
    },
    {
      "epoch": 4.77,
      "grad_norm": 1.556859016418457,
      "learning_rate": 0.002612962962962963,
      "loss": 0.5783,
      "step": 2580
    },
    {
      "epoch": 4.79,
      "grad_norm": 3.010946750640869,
      "learning_rate": 0.0026037037037037036,
      "loss": 0.5786,
      "step": 2590
    },
    {
      "epoch": 4.81,
      "grad_norm": 1.5896031856536865,
      "learning_rate": 0.0025944444444444444,
      "loss": 0.6075,
      "step": 2600
    },
    {
      "epoch": 4.83,
      "grad_norm": 1.9418363571166992,
      "learning_rate": 0.0025851851851851856,
      "loss": 0.5482,
      "step": 2610
    },
    {
      "epoch": 4.85,
      "grad_norm": 1.9787338972091675,
      "learning_rate": 0.002576851851851852,
      "loss": 0.6026,
      "step": 2620
    },
    {
      "epoch": 4.87,
      "grad_norm": 2.7021665573120117,
      "learning_rate": 0.0025675925925925927,
      "loss": 0.4977,
      "step": 2630
    },
    {
      "epoch": 4.88,
      "grad_norm": 2.2835893630981445,
      "learning_rate": 0.0025583333333333335,
      "loss": 0.5375,
      "step": 2640
    },
    {
      "epoch": 4.9,
      "grad_norm": 1.9748303890228271,
      "learning_rate": 0.0025490740740740743,
      "loss": 0.5263,
      "step": 2650
    },
    {
      "epoch": 4.92,
      "grad_norm": 1.6271171569824219,
      "learning_rate": 0.0025398148148148146,
      "loss": 0.5275,
      "step": 2660
    },
    {
      "epoch": 4.94,
      "grad_norm": 2.3140103816986084,
      "learning_rate": 0.0025305555555555554,
      "loss": 0.5493,
      "step": 2670
    },
    {
      "epoch": 4.96,
      "grad_norm": 2.2506508827209473,
      "learning_rate": 0.0025212962962962967,
      "loss": 0.5756,
      "step": 2680
    },
    {
      "epoch": 4.98,
      "grad_norm": 2.5812437534332275,
      "learning_rate": 0.002512037037037037,
      "loss": 0.511,
      "step": 2690
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.886813998222351,
      "learning_rate": 0.002502777777777778,
      "loss": 0.5509,
      "step": 2700
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9770451394238175,
      "eval_f1": 0.9783131597050811,
      "eval_loss": 0.06713523715734482,
      "eval_precision": 0.9775503610869588,
      "eval_recall": 0.9795962655538212,
      "eval_runtime": 55.3992,
      "eval_samples_per_second": 117.168,
      "eval_steps_per_second": 7.329,
      "step": 2702
    },
    {
      "epoch": 5.01,
      "grad_norm": 2.990206241607666,
      "learning_rate": 0.0024935185185185186,
      "loss": 0.5212,
      "step": 2710
    },
    {
      "epoch": 5.03,
      "grad_norm": 1.2570441961288452,
      "learning_rate": 0.0024842592592592594,
      "loss": 0.4703,
      "step": 2720
    },
    {
      "epoch": 5.05,
      "grad_norm": 1.6648197174072266,
      "learning_rate": 0.002475,
      "loss": 0.6173,
      "step": 2730
    },
    {
      "epoch": 5.07,
      "grad_norm": 1.9291369915008545,
      "learning_rate": 0.002465740740740741,
      "loss": 0.5929,
      "step": 2740
    },
    {
      "epoch": 5.09,
      "grad_norm": 2.2894320487976074,
      "learning_rate": 0.0024564814814814814,
      "loss": 0.5282,
      "step": 2750
    },
    {
      "epoch": 5.11,
      "grad_norm": NaN,
      "learning_rate": 0.002448148148148148,
      "loss": 0.5085,
      "step": 2760
    },
    {
      "epoch": 5.12,
      "grad_norm": 2.4065709114074707,
      "learning_rate": 0.002438888888888889,
      "loss": 0.5983,
      "step": 2770
    },
    {
      "epoch": 5.14,
      "grad_norm": 1.4784891605377197,
      "learning_rate": 0.0024296296296296297,
      "loss": 0.5586,
      "step": 2780
    },
    {
      "epoch": 5.16,
      "grad_norm": 3.6442089080810547,
      "learning_rate": 0.0024203703703703705,
      "loss": 0.5824,
      "step": 2790
    },
    {
      "epoch": 5.18,
      "grad_norm": 1.4887707233428955,
      "learning_rate": 0.0024111111111111113,
      "loss": 0.6023,
      "step": 2800
    },
    {
      "epoch": 5.2,
      "grad_norm": 1.4839690923690796,
      "learning_rate": 0.002401851851851852,
      "loss": 0.5486,
      "step": 2810
    },
    {
      "epoch": 5.22,
      "grad_norm": 1.522079586982727,
      "learning_rate": 0.0023925925925925924,
      "loss": 0.5803,
      "step": 2820
    },
    {
      "epoch": 5.24,
      "grad_norm": 1.6155500411987305,
      "learning_rate": 0.0023833333333333337,
      "loss": 0.5765,
      "step": 2830
    },
    {
      "epoch": 5.25,
      "grad_norm": 1.4608234167099,
      "learning_rate": 0.002374074074074074,
      "loss": 0.5163,
      "step": 2840
    },
    {
      "epoch": 5.27,
      "grad_norm": 1.8083865642547607,
      "learning_rate": 0.002364814814814815,
      "loss": 0.5533,
      "step": 2850
    },
    {
      "epoch": 5.29,
      "grad_norm": 1.7674206495285034,
      "learning_rate": 0.0023555555555555556,
      "loss": 0.5558,
      "step": 2860
    },
    {
      "epoch": 5.31,
      "grad_norm": 1.7677381038665771,
      "learning_rate": 0.0023462962962962964,
      "loss": 0.551,
      "step": 2870
    },
    {
      "epoch": 5.33,
      "grad_norm": 2.006486415863037,
      "learning_rate": 0.0023370370370370368,
      "loss": 0.5311,
      "step": 2880
    },
    {
      "epoch": 5.35,
      "grad_norm": 1.3190593719482422,
      "learning_rate": 0.002327777777777778,
      "loss": 0.5462,
      "step": 2890
    },
    {
      "epoch": 5.37,
      "grad_norm": 2.1701364517211914,
      "learning_rate": 0.002318518518518519,
      "loss": 0.5056,
      "step": 2900
    },
    {
      "epoch": 5.38,
      "grad_norm": 3.071043014526367,
      "learning_rate": 0.002309259259259259,
      "loss": 0.5611,
      "step": 2910
    },
    {
      "epoch": 5.4,
      "grad_norm": 1.7673370838165283,
      "learning_rate": 0.0023,
      "loss": 0.5369,
      "step": 2920
    },
    {
      "epoch": 5.42,
      "grad_norm": 1.6603857278823853,
      "learning_rate": 0.0022907407407407408,
      "loss": 0.6108,
      "step": 2930
    },
    {
      "epoch": 5.44,
      "grad_norm": 3.754216194152832,
      "learning_rate": 0.0022814814814814816,
      "loss": 0.5355,
      "step": 2940
    },
    {
      "epoch": 5.46,
      "grad_norm": 2.2037136554718018,
      "learning_rate": 0.0022722222222222224,
      "loss": 0.4998,
      "step": 2950
    },
    {
      "epoch": 5.48,
      "grad_norm": 2.2701072692871094,
      "learning_rate": 0.002262962962962963,
      "loss": 0.5672,
      "step": 2960
    },
    {
      "epoch": 5.49,
      "grad_norm": 2.246368408203125,
      "learning_rate": 0.0022537037037037035,
      "loss": 0.5309,
      "step": 2970
    },
    {
      "epoch": 5.51,
      "grad_norm": 1.4481878280639648,
      "learning_rate": 0.0022444444444444443,
      "loss": 0.4786,
      "step": 2980
    },
    {
      "epoch": 5.53,
      "grad_norm": 1.7569130659103394,
      "learning_rate": 0.002235185185185185,
      "loss": 0.579,
      "step": 2990
    },
    {
      "epoch": 5.55,
      "grad_norm": 2.219710350036621,
      "learning_rate": 0.002225925925925926,
      "loss": 0.4768,
      "step": 3000
    },
    {
      "epoch": 5.57,
      "grad_norm": 1.3611927032470703,
      "learning_rate": 0.0022166666666666667,
      "loss": 0.4504,
      "step": 3010
    },
    {
      "epoch": 5.59,
      "grad_norm": 1.528106451034546,
      "learning_rate": 0.0022074074074074075,
      "loss": 0.491,
      "step": 3020
    },
    {
      "epoch": 5.61,
      "grad_norm": 1.934448003768921,
      "learning_rate": 0.0021981481481481483,
      "loss": 0.5553,
      "step": 3030
    },
    {
      "epoch": 5.62,
      "grad_norm": 1.9539951086044312,
      "learning_rate": 0.002188888888888889,
      "loss": 0.514,
      "step": 3040
    },
    {
      "epoch": 5.64,
      "grad_norm": 3.347055196762085,
      "learning_rate": 0.00217962962962963,
      "loss": 0.565,
      "step": 3050
    },
    {
      "epoch": 5.66,
      "grad_norm": 2.0984296798706055,
      "learning_rate": 0.0021703703703703702,
      "loss": 0.4917,
      "step": 3060
    },
    {
      "epoch": 5.68,
      "grad_norm": 4.030628681182861,
      "learning_rate": 0.002161111111111111,
      "loss": 0.5485,
      "step": 3070
    },
    {
      "epoch": 5.7,
      "grad_norm": 1.62152898311615,
      "learning_rate": 0.002151851851851852,
      "loss": 0.5468,
      "step": 3080
    },
    {
      "epoch": 5.72,
      "grad_norm": 2.245067596435547,
      "learning_rate": 0.0021425925925925926,
      "loss": 0.4992,
      "step": 3090
    },
    {
      "epoch": 5.74,
      "grad_norm": 2.4511606693267822,
      "learning_rate": 0.0021333333333333334,
      "loss": 0.5478,
      "step": 3100
    },
    {
      "epoch": 5.75,
      "grad_norm": 3.2524254322052,
      "learning_rate": 0.0021240740740740742,
      "loss": 0.5006,
      "step": 3110
    },
    {
      "epoch": 5.77,
      "grad_norm": 1.8704936504364014,
      "learning_rate": 0.002114814814814815,
      "loss": 0.5177,
      "step": 3120
    },
    {
      "epoch": 5.79,
      "grad_norm": 1.6047933101654053,
      "learning_rate": 0.0021064814814814813,
      "loss": 0.5491,
      "step": 3130
    },
    {
      "epoch": 5.81,
      "grad_norm": 2.038342237472534,
      "learning_rate": 0.002097222222222222,
      "loss": 0.5096,
      "step": 3140
    },
    {
      "epoch": 5.83,
      "grad_norm": 1.1834994554519653,
      "learning_rate": 0.002087962962962963,
      "loss": 0.4611,
      "step": 3150
    },
    {
      "epoch": 5.85,
      "grad_norm": 3.6137051582336426,
      "learning_rate": 0.0020787037037037037,
      "loss": 0.5056,
      "step": 3160
    },
    {
      "epoch": 5.86,
      "grad_norm": 1.7401528358459473,
      "learning_rate": 0.0020694444444444445,
      "loss": 0.5285,
      "step": 3170
    },
    {
      "epoch": 5.88,
      "grad_norm": 1.1825543642044067,
      "learning_rate": 0.0020601851851851853,
      "loss": 0.5166,
      "step": 3180
    },
    {
      "epoch": 5.9,
      "grad_norm": 1.469771385192871,
      "learning_rate": 0.002050925925925926,
      "loss": 0.5246,
      "step": 3190
    },
    {
      "epoch": 5.92,
      "grad_norm": 1.9187328815460205,
      "learning_rate": 0.0020416666666666665,
      "loss": 0.5687,
      "step": 3200
    },
    {
      "epoch": 5.94,
      "grad_norm": 1.27570641040802,
      "learning_rate": 0.0020324074074074077,
      "loss": 0.5367,
      "step": 3210
    },
    {
      "epoch": 5.96,
      "grad_norm": 2.436992883682251,
      "learning_rate": 0.002023148148148148,
      "loss": 0.5108,
      "step": 3220
    },
    {
      "epoch": 5.98,
      "grad_norm": 2.1716506481170654,
      "learning_rate": 0.002013888888888889,
      "loss": 0.5081,
      "step": 3230
    },
    {
      "epoch": 5.99,
      "grad_norm": 2.3047661781311035,
      "learning_rate": 0.0020046296296296296,
      "loss": 0.5746,
      "step": 3240
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9753504852873209,
      "eval_f1": 0.9794812735446677,
      "eval_loss": 0.06419458985328674,
      "eval_precision": 0.9810236528071776,
      "eval_recall": 0.9788061689635984,
      "eval_runtime": 55.1158,
      "eval_samples_per_second": 117.77,
      "eval_steps_per_second": 7.366,
      "step": 3243
    },
    {
      "epoch": 6.01,
      "grad_norm": 2.193244457244873,
      "learning_rate": 0.0019953703703703704,
      "loss": 0.4844,
      "step": 3250
    },
    {
      "epoch": 6.03,
      "grad_norm": 1.7774107456207275,
      "learning_rate": 0.0019861111111111112,
      "loss": 0.5153,
      "step": 3260
    },
    {
      "epoch": 6.05,
      "grad_norm": 1.8430052995681763,
      "learning_rate": 0.001976851851851852,
      "loss": 0.5332,
      "step": 3270
    },
    {
      "epoch": 6.07,
      "grad_norm": 2.268313407897949,
      "learning_rate": 0.001967592592592593,
      "loss": 0.4762,
      "step": 3280
    },
    {
      "epoch": 6.09,
      "grad_norm": 2.3177573680877686,
      "learning_rate": 0.001958333333333333,
      "loss": 0.5045,
      "step": 3290
    },
    {
      "epoch": 6.11,
      "grad_norm": 2.091614007949829,
      "learning_rate": 0.0019490740740740742,
      "loss": 0.426,
      "step": 3300
    },
    {
      "epoch": 6.12,
      "grad_norm": 2.227275848388672,
      "learning_rate": 0.0019398148148148148,
      "loss": 0.5102,
      "step": 3310
    },
    {
      "epoch": 6.14,
      "grad_norm": 1.47825026512146,
      "learning_rate": 0.0019305555555555556,
      "loss": 0.4559,
      "step": 3320
    },
    {
      "epoch": 6.16,
      "grad_norm": 2.3970351219177246,
      "learning_rate": 0.0019212962962962962,
      "loss": 0.5202,
      "step": 3330
    },
    {
      "epoch": 6.18,
      "grad_norm": 1.6059942245483398,
      "learning_rate": 0.0019120370370370372,
      "loss": 0.4684,
      "step": 3340
    },
    {
      "epoch": 6.2,
      "grad_norm": 1.4332858324050903,
      "learning_rate": 0.0019027777777777778,
      "loss": 0.5029,
      "step": 3350
    },
    {
      "epoch": 6.22,
      "grad_norm": 2.158965587615967,
      "learning_rate": 0.0018935185185185186,
      "loss": 0.5185,
      "step": 3360
    },
    {
      "epoch": 6.23,
      "grad_norm": 2.077367067337036,
      "learning_rate": 0.0018842592592592591,
      "loss": 0.472,
      "step": 3370
    },
    {
      "epoch": 6.25,
      "grad_norm": 2.0703649520874023,
      "learning_rate": 0.001875,
      "loss": 0.4818,
      "step": 3380
    },
    {
      "epoch": 6.27,
      "grad_norm": 2.000805139541626,
      "learning_rate": 0.001865740740740741,
      "loss": 0.4643,
      "step": 3390
    },
    {
      "epoch": 6.29,
      "grad_norm": 1.556027889251709,
      "learning_rate": 0.0018564814814814815,
      "loss": 0.468,
      "step": 3400
    },
    {
      "epoch": 6.31,
      "grad_norm": 1.2694203853607178,
      "learning_rate": 0.0018472222222222223,
      "loss": 0.4437,
      "step": 3410
    },
    {
      "epoch": 6.33,
      "grad_norm": 2.0669643878936768,
      "learning_rate": 0.001837962962962963,
      "loss": 0.4565,
      "step": 3420
    },
    {
      "epoch": 6.35,
      "grad_norm": 1.6692304611206055,
      "learning_rate": 0.001828703703703704,
      "loss": 0.4573,
      "step": 3430
    },
    {
      "epoch": 6.36,
      "grad_norm": 1.6164546012878418,
      "learning_rate": 0.0018194444444444445,
      "loss": 0.4282,
      "step": 3440
    },
    {
      "epoch": 6.38,
      "grad_norm": 1.9918891191482544,
      "learning_rate": 0.0018101851851851853,
      "loss": 0.4522,
      "step": 3450
    },
    {
      "epoch": 6.4,
      "grad_norm": 1.4127854108810425,
      "learning_rate": 0.0018009259259259259,
      "loss": 0.5317,
      "step": 3460
    },
    {
      "epoch": 6.42,
      "grad_norm": 1.8190535306930542,
      "learning_rate": 0.0017916666666666667,
      "loss": 0.5193,
      "step": 3470
    },
    {
      "epoch": 6.44,
      "grad_norm": 1.8793113231658936,
      "learning_rate": 0.0017824074074074072,
      "loss": 0.5322,
      "step": 3480
    },
    {
      "epoch": 6.46,
      "grad_norm": 1.1124266386032104,
      "learning_rate": 0.0017731481481481483,
      "loss": 0.4849,
      "step": 3490
    },
    {
      "epoch": 6.48,
      "grad_norm": 1.5575320720672607,
      "learning_rate": 0.001763888888888889,
      "loss": 0.5254,
      "step": 3500
    },
    {
      "epoch": 6.49,
      "grad_norm": 1.3481545448303223,
      "learning_rate": 0.0017546296296296296,
      "loss": 0.4869,
      "step": 3510
    },
    {
      "epoch": 6.51,
      "grad_norm": 1.0486959218978882,
      "learning_rate": 0.0017453703703703704,
      "loss": 0.5048,
      "step": 3520
    },
    {
      "epoch": 6.53,
      "grad_norm": 1.848264217376709,
      "learning_rate": 0.001736111111111111,
      "loss": 0.5064,
      "step": 3530
    },
    {
      "epoch": 6.55,
      "grad_norm": 1.6750601530075073,
      "learning_rate": 0.001726851851851852,
      "loss": 0.4459,
      "step": 3540
    },
    {
      "epoch": 6.57,
      "grad_norm": 1.5514721870422363,
      "learning_rate": 0.0017175925925925926,
      "loss": 0.4382,
      "step": 3550
    },
    {
      "epoch": 6.59,
      "grad_norm": 1.3411380052566528,
      "learning_rate": 0.0017083333333333334,
      "loss": 0.4007,
      "step": 3560
    },
    {
      "epoch": 6.6,
      "grad_norm": 1.8607478141784668,
      "learning_rate": 0.001699074074074074,
      "loss": 0.4776,
      "step": 3570
    },
    {
      "epoch": 6.62,
      "grad_norm": 1.6788519620895386,
      "learning_rate": 0.0016898148148148148,
      "loss": 0.4499,
      "step": 3580
    },
    {
      "epoch": 6.64,
      "grad_norm": 1.4055275917053223,
      "learning_rate": 0.0016805555555555558,
      "loss": 0.5017,
      "step": 3590
    },
    {
      "epoch": 6.66,
      "grad_norm": 1.9145076274871826,
      "learning_rate": 0.0016712962962962964,
      "loss": 0.4196,
      "step": 3600
    },
    {
      "epoch": 6.68,
      "grad_norm": 1.7690577507019043,
      "learning_rate": 0.0016620370370370372,
      "loss": 0.4792,
      "step": 3610
    },
    {
      "epoch": 6.7,
      "grad_norm": 1.713212490081787,
      "learning_rate": 0.0016527777777777778,
      "loss": 0.4418,
      "step": 3620
    },
    {
      "epoch": 6.72,
      "grad_norm": 2.4528722763061523,
      "learning_rate": 0.0016435185185185185,
      "loss": 0.4519,
      "step": 3630
    },
    {
      "epoch": 6.73,
      "grad_norm": 1.8348699808120728,
      "learning_rate": 0.0016342592592592591,
      "loss": 0.3959,
      "step": 3640
    },
    {
      "epoch": 6.75,
      "grad_norm": 1.38786780834198,
      "learning_rate": 0.0016250000000000001,
      "loss": 0.4433,
      "step": 3650
    },
    {
      "epoch": 6.77,
      "grad_norm": 0.8413279056549072,
      "learning_rate": 0.0016157407407407407,
      "loss": 0.4869,
      "step": 3660
    },
    {
      "epoch": 6.79,
      "grad_norm": 1.3703649044036865,
      "learning_rate": 0.0016064814814814815,
      "loss": 0.5021,
      "step": 3670
    },
    {
      "epoch": 6.81,
      "grad_norm": 1.6931146383285522,
      "learning_rate": 0.001597222222222222,
      "loss": 0.416,
      "step": 3680
    },
    {
      "epoch": 6.83,
      "grad_norm": 1.314029574394226,
      "learning_rate": 0.001587962962962963,
      "loss": 0.4489,
      "step": 3690
    },
    {
      "epoch": 6.85,
      "grad_norm": 1.6631927490234375,
      "learning_rate": 0.001578703703703704,
      "loss": 0.4334,
      "step": 3700
    },
    {
      "epoch": 6.86,
      "grad_norm": 1.821455478668213,
      "learning_rate": 0.0015694444444444445,
      "loss": 0.4307,
      "step": 3710
    },
    {
      "epoch": 6.88,
      "grad_norm": 1.4613134860992432,
      "learning_rate": 0.0015601851851851853,
      "loss": 0.414,
      "step": 3720
    },
    {
      "epoch": 6.9,
      "grad_norm": 1.9156243801116943,
      "learning_rate": 0.0015509259259259259,
      "loss": 0.5022,
      "step": 3730
    },
    {
      "epoch": 6.92,
      "grad_norm": 1.9036259651184082,
      "learning_rate": 0.0015416666666666669,
      "loss": 0.4562,
      "step": 3740
    },
    {
      "epoch": 6.94,
      "grad_norm": 1.6884894371032715,
      "learning_rate": 0.0015324074074074075,
      "loss": 0.4418,
      "step": 3750
    },
    {
      "epoch": 6.96,
      "grad_norm": 1.4793485403060913,
      "learning_rate": 0.0015231481481481483,
      "loss": 0.4091,
      "step": 3760
    },
    {
      "epoch": 6.98,
      "grad_norm": 1.9029099941253662,
      "learning_rate": 0.0015138888888888888,
      "loss": 0.4827,
      "step": 3770
    },
    {
      "epoch": 6.99,
      "grad_norm": 1.2703421115875244,
      "learning_rate": 0.0015046296296296296,
      "loss": 0.4066,
      "step": 3780
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9565552303189031,
      "eval_f1": 0.96141521763694,
      "eval_loss": 0.11959208548069,
      "eval_precision": 0.9693468985322017,
      "eval_recall": 0.9563045431610278,
      "eval_runtime": 55.0215,
      "eval_samples_per_second": 117.972,
      "eval_steps_per_second": 7.379,
      "step": 3783
    },
    {
      "epoch": 7.01,
      "grad_norm": 1.4109677076339722,
      "learning_rate": 0.0014953703703703702,
      "loss": 0.434,
      "step": 3790
    },
    {
      "epoch": 7.03,
      "grad_norm": 1.6477872133255005,
      "learning_rate": 0.0014861111111111112,
      "loss": 0.466,
      "step": 3800
    },
    {
      "epoch": 7.05,
      "grad_norm": 0.9931581020355225,
      "learning_rate": 0.001476851851851852,
      "loss": 0.4282,
      "step": 3810
    },
    {
      "epoch": 7.07,
      "grad_norm": 2.313886880874634,
      "learning_rate": 0.0014675925925925926,
      "loss": 0.4249,
      "step": 3820
    },
    {
      "epoch": 7.09,
      "grad_norm": 1.5501474142074585,
      "learning_rate": 0.0014583333333333334,
      "loss": 0.4062,
      "step": 3830
    },
    {
      "epoch": 7.1,
      "grad_norm": 2.267270088195801,
      "learning_rate": 0.001449074074074074,
      "loss": 0.4546,
      "step": 3840
    },
    {
      "epoch": 7.12,
      "grad_norm": 2.5071542263031006,
      "learning_rate": 0.001439814814814815,
      "loss": 0.4828,
      "step": 3850
    },
    {
      "epoch": 7.14,
      "grad_norm": 1.464120864868164,
      "learning_rate": 0.0014305555555555556,
      "loss": 0.416,
      "step": 3860
    },
    {
      "epoch": 7.16,
      "grad_norm": 1.3284567594528198,
      "learning_rate": 0.0014212962962962964,
      "loss": 0.3704,
      "step": 3870
    },
    {
      "epoch": 7.18,
      "grad_norm": 1.5376476049423218,
      "learning_rate": 0.001412037037037037,
      "loss": 0.454,
      "step": 3880
    },
    {
      "epoch": 7.2,
      "grad_norm": 1.0760722160339355,
      "learning_rate": 0.0014027777777777777,
      "loss": 0.4538,
      "step": 3890
    },
    {
      "epoch": 7.22,
      "grad_norm": 2.0362236499786377,
      "learning_rate": 0.0013935185185185188,
      "loss": 0.4178,
      "step": 3900
    },
    {
      "epoch": 7.23,
      "grad_norm": 1.949922800064087,
      "learning_rate": 0.0013842592592592593,
      "loss": 0.3457,
      "step": 3910
    },
    {
      "epoch": 7.25,
      "grad_norm": 1.173089623451233,
      "learning_rate": 0.0013750000000000001,
      "loss": 0.389,
      "step": 3920
    },
    {
      "epoch": 7.27,
      "grad_norm": 1.4733856916427612,
      "learning_rate": 0.0013657407407407407,
      "loss": 0.456,
      "step": 3930
    },
    {
      "epoch": 7.29,
      "grad_norm": 1.5104173421859741,
      "learning_rate": 0.0013564814814814815,
      "loss": 0.406,
      "step": 3940
    },
    {
      "epoch": 7.31,
      "grad_norm": 2.4948642253875732,
      "learning_rate": 0.001347222222222222,
      "loss": 0.4195,
      "step": 3950
    },
    {
      "epoch": 7.33,
      "grad_norm": 2.137315511703491,
      "learning_rate": 0.001337962962962963,
      "loss": 0.4857,
      "step": 3960
    },
    {
      "epoch": 7.35,
      "grad_norm": 1.20357084274292,
      "learning_rate": 0.0013287037037037037,
      "loss": 0.4287,
      "step": 3970
    },
    {
      "epoch": 7.36,
      "grad_norm": 1.5241217613220215,
      "learning_rate": 0.0013194444444444445,
      "loss": 0.4535,
      "step": 3980
    },
    {
      "epoch": 7.38,
      "grad_norm": 1.4119207859039307,
      "learning_rate": 0.001310185185185185,
      "loss": 0.486,
      "step": 3990
    },
    {
      "epoch": 7.4,
      "grad_norm": 1.3262003660202026,
      "learning_rate": 0.0013009259259259259,
      "loss": 0.4833,
      "step": 4000
    },
    {
      "epoch": 7.42,
      "grad_norm": 1.0021650791168213,
      "learning_rate": 0.0012916666666666669,
      "loss": 0.3996,
      "step": 4010
    },
    {
      "epoch": 7.44,
      "grad_norm": 1.8423185348510742,
      "learning_rate": 0.0012824074074074075,
      "loss": 0.4569,
      "step": 4020
    },
    {
      "epoch": 7.46,
      "grad_norm": 2.010634183883667,
      "learning_rate": 0.0012731481481481483,
      "loss": 0.4673,
      "step": 4030
    },
    {
      "epoch": 7.47,
      "grad_norm": 2.974384307861328,
      "learning_rate": 0.0012638888888888888,
      "loss": 0.3992,
      "step": 4040
    },
    {
      "epoch": 7.49,
      "grad_norm": 1.022455096244812,
      "learning_rate": 0.0012546296296296296,
      "loss": 0.3598,
      "step": 4050
    },
    {
      "epoch": 7.51,
      "grad_norm": 1.5852398872375488,
      "learning_rate": 0.0012453703703703704,
      "loss": 0.4115,
      "step": 4060
    },
    {
      "epoch": 7.53,
      "grad_norm": 1.3596739768981934,
      "learning_rate": 0.0012361111111111112,
      "loss": 0.3534,
      "step": 4070
    },
    {
      "epoch": 7.55,
      "grad_norm": 1.9260822534561157,
      "learning_rate": 0.001226851851851852,
      "loss": 0.4297,
      "step": 4080
    },
    {
      "epoch": 7.57,
      "grad_norm": 1.5783950090408325,
      "learning_rate": 0.0012175925925925926,
      "loss": 0.368,
      "step": 4090
    },
    {
      "epoch": 7.59,
      "grad_norm": 1.7927485704421997,
      "learning_rate": 0.0012083333333333334,
      "loss": 0.3701,
      "step": 4100
    },
    {
      "epoch": 7.6,
      "grad_norm": 1.1680549383163452,
      "learning_rate": 0.0011990740740740742,
      "loss": 0.3883,
      "step": 4110
    },
    {
      "epoch": 7.62,
      "grad_norm": 1.4916154146194458,
      "learning_rate": 0.0011898148148148148,
      "loss": 0.373,
      "step": 4120
    },
    {
      "epoch": 7.64,
      "grad_norm": 1.3623180389404297,
      "learning_rate": 0.0011805555555555556,
      "loss": 0.3697,
      "step": 4130
    },
    {
      "epoch": 7.66,
      "grad_norm": 1.2618119716644287,
      "learning_rate": 0.0011712962962962964,
      "loss": 0.4114,
      "step": 4140
    },
    {
      "epoch": 7.68,
      "grad_norm": 1.7506533861160278,
      "learning_rate": 0.001162037037037037,
      "loss": 0.4003,
      "step": 4150
    },
    {
      "epoch": 7.7,
      "grad_norm": 1.0284491777420044,
      "learning_rate": 0.001152777777777778,
      "loss": 0.3749,
      "step": 4160
    },
    {
      "epoch": 7.72,
      "grad_norm": 1.344994306564331,
      "learning_rate": 0.0011435185185185185,
      "loss": 0.4025,
      "step": 4170
    },
    {
      "epoch": 7.73,
      "grad_norm": 3.066920042037964,
      "learning_rate": 0.0011342592592592593,
      "loss": 0.4203,
      "step": 4180
    },
    {
      "epoch": 7.75,
      "grad_norm": 1.4180749654769897,
      "learning_rate": 0.0011250000000000001,
      "loss": 0.4364,
      "step": 4190
    },
    {
      "epoch": 7.77,
      "grad_norm": 1.5211139917373657,
      "learning_rate": 0.0011157407407407407,
      "loss": 0.4133,
      "step": 4200
    },
    {
      "epoch": 7.79,
      "grad_norm": 1.9516364336013794,
      "learning_rate": 0.0011064814814814815,
      "loss": 0.4232,
      "step": 4210
    },
    {
      "epoch": 7.81,
      "grad_norm": 2.458101511001587,
      "learning_rate": 0.0010972222222222223,
      "loss": 0.434,
      "step": 4220
    },
    {
      "epoch": 7.83,
      "grad_norm": 1.2099213600158691,
      "learning_rate": 0.0010879629629629629,
      "loss": 0.3169,
      "step": 4230
    },
    {
      "epoch": 7.84,
      "grad_norm": 1.0573714971542358,
      "learning_rate": 0.0010787037037037037,
      "loss": 0.3777,
      "step": 4240
    },
    {
      "epoch": 7.86,
      "grad_norm": 1.4449975490570068,
      "learning_rate": 0.0010694444444444445,
      "loss": 0.4426,
      "step": 4250
    },
    {
      "epoch": 7.88,
      "grad_norm": 1.335119366645813,
      "learning_rate": 0.001060185185185185,
      "loss": 0.4643,
      "step": 4260
    },
    {
      "epoch": 7.9,
      "grad_norm": 2.1732048988342285,
      "learning_rate": 0.001050925925925926,
      "loss": 0.3903,
      "step": 4270
    },
    {
      "epoch": 7.92,
      "grad_norm": 1.193305492401123,
      "learning_rate": 0.0010416666666666667,
      "loss": 0.3744,
      "step": 4280
    },
    {
      "epoch": 7.94,
      "grad_norm": 0.9779842495918274,
      "learning_rate": 0.0010324074074074074,
      "loss": 0.4124,
      "step": 4290
    },
    {
      "epoch": 7.96,
      "grad_norm": 2.7513339519500732,
      "learning_rate": 0.0010231481481481482,
      "loss": 0.3917,
      "step": 4300
    },
    {
      "epoch": 7.97,
      "grad_norm": 1.756395936012268,
      "learning_rate": 0.0010138888888888888,
      "loss": 0.4301,
      "step": 4310
    },
    {
      "epoch": 7.99,
      "grad_norm": 1.5804221630096436,
      "learning_rate": 0.0010046296296296296,
      "loss": 0.4046,
      "step": 4320
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.979818209828994,
      "eval_f1": 0.9833721625743178,
      "eval_loss": 0.04691172018647194,
      "eval_precision": 0.9853003976747072,
      "eval_recall": 0.9820557579494299,
      "eval_runtime": 54.8438,
      "eval_samples_per_second": 118.354,
      "eval_steps_per_second": 7.403,
      "step": 4324
    },
    {
      "epoch": 8.01,
      "grad_norm": 1.8008298873901367,
      "learning_rate": 0.0009953703703703704,
      "loss": 0.4452,
      "step": 4330
    },
    {
      "epoch": 8.03,
      "grad_norm": 3.602733850479126,
      "learning_rate": 0.000986111111111111,
      "loss": 0.3281,
      "step": 4340
    },
    {
      "epoch": 8.05,
      "grad_norm": 1.5571168661117554,
      "learning_rate": 0.0009768518518518518,
      "loss": 0.432,
      "step": 4350
    },
    {
      "epoch": 8.07,
      "grad_norm": 1.5973519086837769,
      "learning_rate": 0.0009675925925925926,
      "loss": 0.3682,
      "step": 4360
    },
    {
      "epoch": 8.09,
      "grad_norm": 1.1785231828689575,
      "learning_rate": 0.0009583333333333334,
      "loss": 0.3233,
      "step": 4370
    },
    {
      "epoch": 8.1,
      "grad_norm": 1.9037755727767944,
      "learning_rate": 0.0009490740740740742,
      "loss": 0.3733,
      "step": 4380
    },
    {
      "epoch": 8.12,
      "grad_norm": 1.3710215091705322,
      "learning_rate": 0.0009398148148148149,
      "loss": 0.3562,
      "step": 4390
    },
    {
      "epoch": 8.14,
      "grad_norm": 1.4512821435928345,
      "learning_rate": 0.0009305555555555556,
      "loss": 0.3391,
      "step": 4400
    },
    {
      "epoch": 8.16,
      "grad_norm": 1.5068678855895996,
      "learning_rate": 0.0009212962962962964,
      "loss": 0.3703,
      "step": 4410
    },
    {
      "epoch": 8.18,
      "grad_norm": 1.2810806035995483,
      "learning_rate": 0.000912037037037037,
      "loss": 0.3729,
      "step": 4420
    },
    {
      "epoch": 8.2,
      "grad_norm": 1.9714285135269165,
      "learning_rate": 0.0009027777777777777,
      "loss": 0.3988,
      "step": 4430
    },
    {
      "epoch": 8.21,
      "grad_norm": 1.3327577114105225,
      "learning_rate": 0.0008935185185185185,
      "loss": 0.4024,
      "step": 4440
    },
    {
      "epoch": 8.23,
      "grad_norm": 1.0256530046463013,
      "learning_rate": 0.0008842592592592592,
      "loss": 0.3818,
      "step": 4450
    },
    {
      "epoch": 8.25,
      "grad_norm": 2.233092784881592,
      "learning_rate": 0.0008749999999999999,
      "loss": 0.3668,
      "step": 4460
    },
    {
      "epoch": 8.27,
      "grad_norm": 1.3951568603515625,
      "learning_rate": 0.0008657407407407407,
      "loss": 0.44,
      "step": 4470
    },
    {
      "epoch": 8.29,
      "grad_norm": 1.827919840812683,
      "learning_rate": 0.0008564814814814815,
      "loss": 0.3801,
      "step": 4480
    },
    {
      "epoch": 8.31,
      "grad_norm": 2.934967041015625,
      "learning_rate": 0.0008472222222222223,
      "loss": 0.422,
      "step": 4490
    },
    {
      "epoch": 8.33,
      "grad_norm": 1.2205818891525269,
      "learning_rate": 0.000837962962962963,
      "loss": 0.3991,
      "step": 4500
    },
    {
      "epoch": 8.34,
      "grad_norm": 1.424593448638916,
      "learning_rate": 0.0008287037037037038,
      "loss": 0.418,
      "step": 4510
    },
    {
      "epoch": 8.36,
      "grad_norm": 1.0254913568496704,
      "learning_rate": 0.0008194444444444445,
      "loss": 0.3876,
      "step": 4520
    },
    {
      "epoch": 8.38,
      "grad_norm": 0.9939424991607666,
      "learning_rate": 0.0008101851851851852,
      "loss": 0.4267,
      "step": 4530
    },
    {
      "epoch": 8.4,
      "grad_norm": 1.7678611278533936,
      "learning_rate": 0.000800925925925926,
      "loss": 0.3916,
      "step": 4540
    },
    {
      "epoch": 8.42,
      "grad_norm": 1.0343542098999023,
      "learning_rate": 0.0007916666666666666,
      "loss": 0.3939,
      "step": 4550
    },
    {
      "epoch": 8.44,
      "grad_norm": 0.9549526572227478,
      "learning_rate": 0.0007824074074074073,
      "loss": 0.3785,
      "step": 4560
    },
    {
      "epoch": 8.46,
      "grad_norm": 1.0010316371917725,
      "learning_rate": 0.0007731481481481481,
      "loss": 0.3508,
      "step": 4570
    },
    {
      "epoch": 8.47,
      "grad_norm": 2.0805888175964355,
      "learning_rate": 0.0007638888888888889,
      "loss": 0.3614,
      "step": 4580
    },
    {
      "epoch": 8.49,
      "grad_norm": 1.181626558303833,
      "learning_rate": 0.0007546296296296297,
      "loss": 0.359,
      "step": 4590
    },
    {
      "epoch": 8.51,
      "grad_norm": 1.138376235961914,
      "learning_rate": 0.0007453703703703704,
      "loss": 0.3513,
      "step": 4600
    },
    {
      "epoch": 8.53,
      "grad_norm": 1.0916022062301636,
      "learning_rate": 0.0007361111111111111,
      "loss": 0.3199,
      "step": 4610
    },
    {
      "epoch": 8.55,
      "grad_norm": 2.3587114810943604,
      "learning_rate": 0.0007268518518518519,
      "loss": 0.3658,
      "step": 4620
    },
    {
      "epoch": 8.57,
      "grad_norm": 1.2504198551177979,
      "learning_rate": 0.0007175925925925926,
      "loss": 0.3952,
      "step": 4630
    },
    {
      "epoch": 8.58,
      "grad_norm": 1.3484505414962769,
      "learning_rate": 0.0007083333333333333,
      "loss": 0.3542,
      "step": 4640
    },
    {
      "epoch": 8.6,
      "grad_norm": 2.4245710372924805,
      "learning_rate": 0.0006990740740740741,
      "loss": 0.3421,
      "step": 4650
    },
    {
      "epoch": 8.62,
      "grad_norm": 1.2606480121612549,
      "learning_rate": 0.0006898148148148148,
      "loss": 0.3404,
      "step": 4660
    },
    {
      "epoch": 8.64,
      "grad_norm": 1.913200855255127,
      "learning_rate": 0.0006805555555555556,
      "loss": 0.4015,
      "step": 4670
    },
    {
      "epoch": 8.66,
      "grad_norm": 1.5260008573532104,
      "learning_rate": 0.0006712962962962964,
      "loss": 0.3425,
      "step": 4680
    },
    {
      "epoch": 8.68,
      "grad_norm": 1.221198320388794,
      "learning_rate": 0.0006620370370370372,
      "loss": 0.4078,
      "step": 4690
    },
    {
      "epoch": 8.7,
      "grad_norm": 1.630136489868164,
      "learning_rate": 0.0006527777777777778,
      "loss": 0.3508,
      "step": 4700
    },
    {
      "epoch": 8.71,
      "grad_norm": 1.3053523302078247,
      "learning_rate": 0.0006435185185185185,
      "loss": 0.2958,
      "step": 4710
    },
    {
      "epoch": 8.73,
      "grad_norm": 1.5121403932571411,
      "learning_rate": 0.0006342592592592593,
      "loss": 0.3552,
      "step": 4720
    },
    {
      "epoch": 8.75,
      "grad_norm": 1.1451386213302612,
      "learning_rate": 0.000625,
      "loss": 0.3688,
      "step": 4730
    },
    {
      "epoch": 8.77,
      "grad_norm": 2.4750773906707764,
      "learning_rate": 0.0006157407407407407,
      "loss": 0.3435,
      "step": 4740
    },
    {
      "epoch": 8.79,
      "grad_norm": 0.6995977163314819,
      "learning_rate": 0.0006064814814814815,
      "loss": 0.3152,
      "step": 4750
    },
    {
      "epoch": 8.81,
      "grad_norm": 1.6774075031280518,
      "learning_rate": 0.0005972222222222223,
      "loss": 0.2898,
      "step": 4760
    },
    {
      "epoch": 8.83,
      "grad_norm": 1.1041797399520874,
      "learning_rate": 0.000587962962962963,
      "loss": 0.3086,
      "step": 4770
    },
    {
      "epoch": 8.84,
      "grad_norm": 1.2594033479690552,
      "learning_rate": 0.0005787037037037037,
      "loss": 0.3981,
      "step": 4780
    },
    {
      "epoch": 8.86,
      "grad_norm": 0.9759043455123901,
      "learning_rate": 0.0005694444444444445,
      "loss": 0.356,
      "step": 4790
    },
    {
      "epoch": 8.88,
      "grad_norm": 1.4041826725006104,
      "learning_rate": 0.0005601851851851852,
      "loss": 0.3514,
      "step": 4800
    },
    {
      "epoch": 8.9,
      "grad_norm": 0.9928153157234192,
      "learning_rate": 0.000550925925925926,
      "loss": 0.388,
      "step": 4810
    },
    {
      "epoch": 8.92,
      "grad_norm": 1.0990333557128906,
      "learning_rate": 0.0005416666666666668,
      "loss": 0.3013,
      "step": 4820
    },
    {
      "epoch": 8.94,
      "grad_norm": 1.2321481704711914,
      "learning_rate": 0.0005324074074074074,
      "loss": 0.4016,
      "step": 4830
    },
    {
      "epoch": 8.95,
      "grad_norm": 1.1745514869689941,
      "learning_rate": 0.0005231481481481481,
      "loss": 0.3601,
      "step": 4840
    },
    {
      "epoch": 8.97,
      "grad_norm": 1.633880615234375,
      "learning_rate": 0.0005138888888888889,
      "loss": 0.3291,
      "step": 4850
    },
    {
      "epoch": 8.99,
      "grad_norm": 1.1216607093811035,
      "learning_rate": 0.0005046296296296297,
      "loss": 0.3314,
      "step": 4860
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.986134647974118,
      "eval_f1": 0.987440563210657,
      "eval_loss": 0.038755957037210464,
      "eval_precision": 0.9891828577753293,
      "eval_recall": 0.98596866524898,
      "eval_runtime": 55.0297,
      "eval_samples_per_second": 117.955,
      "eval_steps_per_second": 7.378,
      "step": 4864
    },
    {
      "epoch": 9.01,
      "grad_norm": 1.2497656345367432,
      "learning_rate": 0.0004953703703703704,
      "loss": 0.3183,
      "step": 4870
    },
    {
      "epoch": 9.03,
      "grad_norm": 1.5892603397369385,
      "learning_rate": 0.00048611111111111115,
      "loss": 0.3222,
      "step": 4880
    },
    {
      "epoch": 9.05,
      "grad_norm": 1.4494744539260864,
      "learning_rate": 0.00047685185185185184,
      "loss": 0.3573,
      "step": 4890
    },
    {
      "epoch": 9.07,
      "grad_norm": 1.7034820318222046,
      "learning_rate": 0.0004675925925925926,
      "loss": 0.3194,
      "step": 4900
    },
    {
      "epoch": 9.08,
      "grad_norm": 0.8668962121009827,
      "learning_rate": 0.0004583333333333333,
      "loss": 0.3483,
      "step": 4910
    },
    {
      "epoch": 9.1,
      "grad_norm": 1.081874132156372,
      "learning_rate": 0.0004490740740740741,
      "loss": 0.3626,
      "step": 4920
    },
    {
      "epoch": 9.12,
      "grad_norm": 1.78743314743042,
      "learning_rate": 0.0004398148148148148,
      "loss": 0.3651,
      "step": 4930
    },
    {
      "epoch": 9.14,
      "grad_norm": 1.1324867010116577,
      "learning_rate": 0.00043055555555555555,
      "loss": 0.2714,
      "step": 4940
    },
    {
      "epoch": 9.16,
      "grad_norm": 1.9258543252944946,
      "learning_rate": 0.0004212962962962963,
      "loss": 0.3635,
      "step": 4950
    },
    {
      "epoch": 9.18,
      "grad_norm": 1.2044012546539307,
      "learning_rate": 0.00041203703703703704,
      "loss": 0.3126,
      "step": 4960
    },
    {
      "epoch": 9.2,
      "grad_norm": 1.3639435768127441,
      "learning_rate": 0.00040277777777777783,
      "loss": 0.3203,
      "step": 4970
    },
    {
      "epoch": 9.21,
      "grad_norm": 1.7811472415924072,
      "learning_rate": 0.0003935185185185185,
      "loss": 0.3208,
      "step": 4980
    },
    {
      "epoch": 9.23,
      "grad_norm": 1.4253028631210327,
      "learning_rate": 0.00038425925925925927,
      "loss": 0.3326,
      "step": 4990
    },
    {
      "epoch": 9.25,
      "grad_norm": 1.4700595140457153,
      "learning_rate": 0.000375,
      "loss": 0.3015,
      "step": 5000
    },
    {
      "epoch": 9.27,
      "grad_norm": 2.1158525943756104,
      "learning_rate": 0.0003657407407407407,
      "loss": 0.3472,
      "step": 5010
    },
    {
      "epoch": 9.29,
      "grad_norm": 1.325528621673584,
      "learning_rate": 0.0003564814814814815,
      "loss": 0.3365,
      "step": 5020
    },
    {
      "epoch": 9.31,
      "grad_norm": 1.173003077507019,
      "learning_rate": 0.00034722222222222224,
      "loss": 0.3213,
      "step": 5030
    },
    {
      "epoch": 9.32,
      "grad_norm": 2.355008840560913,
      "learning_rate": 0.000337962962962963,
      "loss": 0.3386,
      "step": 5040
    },
    {
      "epoch": 9.34,
      "grad_norm": 1.198581576347351,
      "learning_rate": 0.0003287037037037037,
      "loss": 0.2985,
      "step": 5050
    },
    {
      "epoch": 9.36,
      "grad_norm": 1.7199565172195435,
      "learning_rate": 0.0003194444444444444,
      "loss": 0.3666,
      "step": 5060
    },
    {
      "epoch": 9.38,
      "grad_norm": 1.359098196029663,
      "learning_rate": 0.0003101851851851852,
      "loss": 0.327,
      "step": 5070
    },
    {
      "epoch": 9.4,
      "grad_norm": 1.2990363836288452,
      "learning_rate": 0.0003009259259259259,
      "loss": 0.3673,
      "step": 5080
    },
    {
      "epoch": 9.42,
      "grad_norm": 1.5110586881637573,
      "learning_rate": 0.0002916666666666667,
      "loss": 0.3124,
      "step": 5090
    },
    {
      "epoch": 9.44,
      "grad_norm": 1.6193897724151611,
      "learning_rate": 0.0002824074074074074,
      "loss": 0.3056,
      "step": 5100
    },
    {
      "epoch": 9.45,
      "grad_norm": 1.2994916439056396,
      "learning_rate": 0.0002731481481481482,
      "loss": 0.3041,
      "step": 5110
    },
    {
      "epoch": 9.47,
      "grad_norm": 1.1582506895065308,
      "learning_rate": 0.0002638888888888889,
      "loss": 0.325,
      "step": 5120
    },
    {
      "epoch": 9.49,
      "grad_norm": 1.2252171039581299,
      "learning_rate": 0.0002546296296296296,
      "loss": 0.3541,
      "step": 5130
    },
    {
      "epoch": 9.51,
      "grad_norm": 1.0199075937271118,
      "learning_rate": 0.0002453703703703704,
      "loss": 0.3763,
      "step": 5140
    },
    {
      "epoch": 9.53,
      "grad_norm": 1.113032341003418,
      "learning_rate": 0.00023611111111111112,
      "loss": 0.3459,
      "step": 5150
    },
    {
      "epoch": 9.55,
      "grad_norm": 1.1221752166748047,
      "learning_rate": 0.00022685185185185186,
      "loss": 0.284,
      "step": 5160
    },
    {
      "epoch": 9.57,
      "grad_norm": 1.0831656455993652,
      "learning_rate": 0.0002175925925925926,
      "loss": 0.3117,
      "step": 5170
    },
    {
      "epoch": 9.58,
      "grad_norm": 0.723981499671936,
      "learning_rate": 0.00020833333333333332,
      "loss": 0.3231,
      "step": 5180
    },
    {
      "epoch": 9.6,
      "grad_norm": 1.0349414348602295,
      "learning_rate": 0.0001990740740740741,
      "loss": 0.3344,
      "step": 5190
    },
    {
      "epoch": 9.62,
      "grad_norm": 0.9438874125480652,
      "learning_rate": 0.0001898148148148148,
      "loss": 0.3272,
      "step": 5200
    },
    {
      "epoch": 9.64,
      "grad_norm": 1.0570435523986816,
      "learning_rate": 0.00018055555555555555,
      "loss": 0.3309,
      "step": 5210
    },
    {
      "epoch": 9.66,
      "grad_norm": 0.9774361848831177,
      "learning_rate": 0.00017129629629629632,
      "loss": 0.3129,
      "step": 5220
    },
    {
      "epoch": 9.68,
      "grad_norm": 1.1588292121887207,
      "learning_rate": 0.00016203703703703703,
      "loss": 0.3287,
      "step": 5230
    },
    {
      "epoch": 9.69,
      "grad_norm": 1.1384470462799072,
      "learning_rate": 0.00015277777777777777,
      "loss": 0.2838,
      "step": 5240
    },
    {
      "epoch": 9.71,
      "grad_norm": 1.3432296514511108,
      "learning_rate": 0.00014351851851851852,
      "loss": 0.3424,
      "step": 5250
    },
    {
      "epoch": 9.73,
      "grad_norm": 0.6609634160995483,
      "learning_rate": 0.00013425925925925926,
      "loss": 0.2843,
      "step": 5260
    },
    {
      "epoch": 9.75,
      "grad_norm": 1.2155038118362427,
      "learning_rate": 0.000125,
      "loss": 0.3159,
      "step": 5270
    },
    {
      "epoch": 9.77,
      "grad_norm": 1.5623424053192139,
      "learning_rate": 0.00011574074074074073,
      "loss": 0.2891,
      "step": 5280
    },
    {
      "epoch": 9.79,
      "grad_norm": 1.9647743701934814,
      "learning_rate": 0.00010648148148148149,
      "loss": 0.3394,
      "step": 5290
    },
    {
      "epoch": 9.81,
      "grad_norm": 1.1939747333526611,
      "learning_rate": 9.722222222222223e-05,
      "loss": 0.3545,
      "step": 5300
    },
    {
      "epoch": 9.82,
      "grad_norm": 1.5705734491348267,
      "learning_rate": 8.796296296296297e-05,
      "loss": 0.372,
      "step": 5310
    },
    {
      "epoch": 9.84,
      "grad_norm": 1.9765841960906982,
      "learning_rate": 7.87037037037037e-05,
      "loss": 0.3555,
      "step": 5320
    },
    {
      "epoch": 9.86,
      "grad_norm": 1.168407917022705,
      "learning_rate": 6.944444444444444e-05,
      "loss": 0.2545,
      "step": 5330
    },
    {
      "epoch": 9.88,
      "grad_norm": 0.8777741193771362,
      "learning_rate": 6.018518518518519e-05,
      "loss": 0.3408,
      "step": 5340
    },
    {
      "epoch": 9.9,
      "grad_norm": 1.2033889293670654,
      "learning_rate": 5.092592592592593e-05,
      "loss": 0.3518,
      "step": 5350
    },
    {
      "epoch": 9.92,
      "grad_norm": 1.9429413080215454,
      "learning_rate": 4.1666666666666665e-05,
      "loss": 0.2779,
      "step": 5360
    },
    {
      "epoch": 9.94,
      "grad_norm": 1.5541284084320068,
      "learning_rate": 3.240740740740741e-05,
      "loss": 0.3169,
      "step": 5370
    },
    {
      "epoch": 9.95,
      "grad_norm": 0.8645866513252258,
      "learning_rate": 2.3148148148148147e-05,
      "loss": 0.2976,
      "step": 5380
    },
    {
      "epoch": 9.97,
      "grad_norm": 1.3550152778625488,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.2891,
      "step": 5390
    },
    {
      "epoch": 9.99,
      "grad_norm": 1.666207194328308,
      "learning_rate": 4.6296296296296296e-06,
      "loss": 0.2865,
      "step": 5400
    },
    {
      "epoch": 9.99,
      "eval_accuracy": 0.9830534586350331,
      "eval_f1": 0.9868535728073075,
      "eval_loss": 0.0450444296002388,
      "eval_precision": 0.9880250018061109,
      "eval_recall": 0.986225426158784,
      "eval_runtime": 55.3663,
      "eval_samples_per_second": 117.237,
      "eval_steps_per_second": 7.333,
      "step": 5400
    },
    {
      "epoch": 9.99,
      "step": 5400,
      "total_flos": 2.695096772600047e+19,
      "train_loss": 0.5508582998205114,
      "train_runtime": 6179.724,
      "train_samples_per_second": 55.959,
      "train_steps_per_second": 0.874
    },
    {
      "epoch": 9.99,
      "eval_accuracy": 0.9328945888176398,
      "eval_f1": 0.9340102214756545,
      "eval_loss": 0.23724627494812012,
      "eval_precision": 0.9416197933012,
      "eval_recall": 0.9295532071921557,
      "eval_runtime": 150.877,
      "eval_samples_per_second": 117.831,
      "eval_steps_per_second": 7.37,
      "step": 5400
    },
    {
      "epoch": 9.99,
      "eval_accuracy": 0.9328945888176398,
      "eval_f1": 0.9340102214756545,
      "eval_loss": 0.23724627494812012,
      "eval_precision": 0.9416197933012,
      "eval_recall": 0.9295532071921557,
      "eval_runtime": 150.8426,
      "eval_samples_per_second": 117.858,
      "eval_steps_per_second": 7.372,
      "step": 5400
    }
  ],
  "logging_steps": 10,
  "max_steps": 5400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 2.695096772600047e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}