{
  "best_metric": 0.9884455399784317,
  "best_model_checkpoint": "swin-large-patch4-window7-224-in22k-finetuned-lora-medmnistv2/checkpoint-4324",
  "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": 2.677260637283325,
      "learning_rate": 0.004990740740740741,
      "loss": 1.6626,
      "step": 10
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.661841630935669,
      "learning_rate": 0.004981481481481482,
      "loss": 1.0711,
      "step": 20
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.4014948606491089,
      "learning_rate": 0.0049722222222222225,
      "loss": 0.9878,
      "step": 30
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.5774717330932617,
      "learning_rate": 0.004962962962962963,
      "loss": 0.832,
      "step": 40
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.944169521331787,
      "learning_rate": 0.004953703703703703,
      "loss": 0.8072,
      "step": 50
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.9590211510658264,
      "learning_rate": 0.004944444444444445,
      "loss": 0.7049,
      "step": 60
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.8535215854644775,
      "learning_rate": 0.004935185185185186,
      "loss": 0.6882,
      "step": 70
    },
    {
      "epoch": 0.15,
      "grad_norm": 2.145578145980835,
      "learning_rate": 0.004925925925925926,
      "loss": 0.8774,
      "step": 80
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.580334186553955,
      "learning_rate": 0.004916666666666666,
      "loss": 0.7384,
      "step": 90
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.8743101358413696,
      "learning_rate": 0.004907407407407408,
      "loss": 0.7732,
      "step": 100
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.0922596454620361,
      "learning_rate": 0.004898148148148148,
      "loss": 0.8463,
      "step": 110
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.5585048198699951,
      "learning_rate": 0.004888888888888889,
      "loss": 0.8079,
      "step": 120
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.7118746042251587,
      "learning_rate": 0.00487962962962963,
      "loss": 0.7567,
      "step": 130
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.6968052387237549,
      "learning_rate": 0.00487037037037037,
      "loss": 0.7144,
      "step": 140
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.3183552026748657,
      "learning_rate": 0.004861111111111111,
      "loss": 0.7303,
      "step": 150
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.2560759782791138,
      "learning_rate": 0.004851851851851852,
      "loss": 0.7694,
      "step": 160
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.185666561126709,
      "learning_rate": 0.004842592592592593,
      "loss": 0.6305,
      "step": 170
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.3163808584213257,
      "learning_rate": 0.004833333333333334,
      "loss": 0.6565,
      "step": 180
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.3728744983673096,
      "learning_rate": 0.004824074074074074,
      "loss": 0.663,
      "step": 190
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.2197016477584839,
      "learning_rate": 0.004814814814814814,
      "loss": 0.7628,
      "step": 200
    },
    {
      "epoch": 0.39,
      "grad_norm": 2.1777913570404053,
      "learning_rate": 0.004805555555555556,
      "loss": 0.5866,
      "step": 210
    },
    {
      "epoch": 0.41,
      "grad_norm": 2.0859122276306152,
      "learning_rate": 0.004796296296296297,
      "loss": 0.6867,
      "step": 220
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.4262019395828247,
      "learning_rate": 0.004787962962962963,
      "loss": 0.6883,
      "step": 230
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.2657884359359741,
      "learning_rate": 0.004778703703703704,
      "loss": 0.651,
      "step": 240
    },
    {
      "epoch": 0.46,
      "grad_norm": 2.6731324195861816,
      "learning_rate": 0.004769444444444444,
      "loss": 0.6012,
      "step": 250
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.3679182529449463,
      "learning_rate": 0.004760185185185185,
      "loss": 0.6211,
      "step": 260
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.6990712881088257,
      "learning_rate": 0.004750925925925926,
      "loss": 0.6239,
      "step": 270
    },
    {
      "epoch": 0.52,
      "grad_norm": 2.043832302093506,
      "learning_rate": 0.004741666666666667,
      "loss": 0.7126,
      "step": 280
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.4476696252822876,
      "learning_rate": 0.004732407407407407,
      "loss": 0.6775,
      "step": 290
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.9128739833831787,
      "learning_rate": 0.004723148148148148,
      "loss": 0.6141,
      "step": 300
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.5211902856826782,
      "learning_rate": 0.004713888888888889,
      "loss": 0.6442,
      "step": 310
    },
    {
      "epoch": 0.59,
      "grad_norm": 3.0712132453918457,
      "learning_rate": 0.00470462962962963,
      "loss": 0.663,
      "step": 320
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.7837520837783813,
      "learning_rate": 0.004695370370370371,
      "loss": 0.674,
      "step": 330
    },
    {
      "epoch": 0.63,
      "grad_norm": 5.605283260345459,
      "learning_rate": 0.004686111111111111,
      "loss": 0.593,
      "step": 340
    },
    {
      "epoch": 0.65,
      "grad_norm": 2.6977906227111816,
      "learning_rate": 0.004676851851851852,
      "loss": 0.6011,
      "step": 350
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.653228998184204,
      "learning_rate": 0.004667592592592593,
      "loss": 0.615,
      "step": 360
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.4604382514953613,
      "learning_rate": 0.004658333333333333,
      "loss": 0.6079,
      "step": 370
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.8553276062011719,
      "learning_rate": 0.004649074074074074,
      "loss": 0.6564,
      "step": 380
    },
    {
      "epoch": 0.72,
      "grad_norm": 2.2343759536743164,
      "learning_rate": 0.004639814814814815,
      "loss": 0.6295,
      "step": 390
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.534677505493164,
      "learning_rate": 0.004630555555555555,
      "loss": 0.6339,
      "step": 400
    },
    {
      "epoch": 0.76,
      "grad_norm": 2.4698429107666016,
      "learning_rate": 0.004621296296296296,
      "loss": 0.7063,
      "step": 410
    },
    {
      "epoch": 0.78,
      "grad_norm": 2.9529638290405273,
      "learning_rate": 0.004612037037037038,
      "loss": 0.6948,
      "step": 420
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.716919183731079,
      "learning_rate": 0.004603703703703704,
      "loss": 0.6224,
      "step": 430
    },
    {
      "epoch": 0.81,
      "grad_norm": 2.205298662185669,
      "learning_rate": 0.004594444444444444,
      "loss": 0.6794,
      "step": 440
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.8521095514297485,
      "learning_rate": 0.004585185185185185,
      "loss": 0.6854,
      "step": 450
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.3433890342712402,
      "learning_rate": 0.004575925925925926,
      "loss": 0.6622,
      "step": 460
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.289505124092102,
      "learning_rate": 0.004566666666666667,
      "loss": 0.5805,
      "step": 470
    },
    {
      "epoch": 0.89,
      "grad_norm": 2.4270637035369873,
      "learning_rate": 0.004557407407407408,
      "loss": 0.5984,
      "step": 480
    },
    {
      "epoch": 0.91,
      "grad_norm": 2.5372653007507324,
      "learning_rate": 0.004548148148148148,
      "loss": 0.6616,
      "step": 490
    },
    {
      "epoch": 0.93,
      "grad_norm": 2.072876453399658,
      "learning_rate": 0.004538888888888889,
      "loss": 0.7577,
      "step": 500
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.5720523595809937,
      "learning_rate": 0.00452962962962963,
      "loss": 0.5948,
      "step": 510
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.810542106628418,
      "learning_rate": 0.00452037037037037,
      "loss": 0.6173,
      "step": 520
    },
    {
      "epoch": 0.98,
      "grad_norm": 2.1614763736724854,
      "learning_rate": 0.004511111111111112,
      "loss": 0.6523,
      "step": 530
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.757303237915039,
      "learning_rate": 0.004501851851851852,
      "loss": 0.6172,
      "step": 540
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9372977969496226,
      "eval_f1": 0.9422208225547629,
      "eval_loss": 0.191307932138443,
      "eval_precision": 0.9480855535269659,
      "eval_recall": 0.9426859850591417,
      "eval_runtime": 105.8196,
      "eval_samples_per_second": 61.34,
      "eval_steps_per_second": 3.837,
      "step": 540
    },
    {
      "epoch": 1.02,
      "grad_norm": 1.6954269409179688,
      "learning_rate": 0.004492592592592592,
      "loss": 0.5184,
      "step": 550
    },
    {
      "epoch": 1.04,
      "grad_norm": 2.1348953247070312,
      "learning_rate": 0.004483333333333333,
      "loss": 0.638,
      "step": 560
    },
    {
      "epoch": 1.05,
      "grad_norm": 2.005446195602417,
      "learning_rate": 0.004474074074074074,
      "loss": 0.646,
      "step": 570
    },
    {
      "epoch": 1.07,
      "grad_norm": 2.4814555644989014,
      "learning_rate": 0.004464814814814815,
      "loss": 0.6302,
      "step": 580
    },
    {
      "epoch": 1.09,
      "grad_norm": 2.1637706756591797,
      "learning_rate": 0.0044555555555555555,
      "loss": 0.5987,
      "step": 590
    },
    {
      "epoch": 1.11,
      "grad_norm": 1.1565651893615723,
      "learning_rate": 0.004446296296296296,
      "loss": 0.5838,
      "step": 600
    },
    {
      "epoch": 1.13,
      "grad_norm": 1.41851007938385,
      "learning_rate": 0.004437037037037037,
      "loss": 0.6723,
      "step": 610
    },
    {
      "epoch": 1.15,
      "grad_norm": 1.4741177558898926,
      "learning_rate": 0.004427777777777778,
      "loss": 0.6424,
      "step": 620
    },
    {
      "epoch": 1.17,
      "grad_norm": 1.8886082172393799,
      "learning_rate": 0.004418518518518519,
      "loss": 0.5423,
      "step": 630
    },
    {
      "epoch": 1.18,
      "grad_norm": 3.1668787002563477,
      "learning_rate": 0.0044092592592592595,
      "loss": 0.6426,
      "step": 640
    },
    {
      "epoch": 1.2,
      "grad_norm": 2.8490853309631348,
      "learning_rate": 0.0044,
      "loss": 0.6451,
      "step": 650
    },
    {
      "epoch": 1.22,
      "grad_norm": 1.922230839729309,
      "learning_rate": 0.004390740740740741,
      "loss": 0.646,
      "step": 660
    },
    {
      "epoch": 1.24,
      "grad_norm": 1.6063907146453857,
      "learning_rate": 0.004381481481481482,
      "loss": 0.631,
      "step": 670
    },
    {
      "epoch": 1.26,
      "grad_norm": 2.440619707107544,
      "learning_rate": 0.004372222222222223,
      "loss": 0.6694,
      "step": 680
    },
    {
      "epoch": 1.28,
      "grad_norm": 2.543225049972534,
      "learning_rate": 0.0043629629629629635,
      "loss": 0.6064,
      "step": 690
    },
    {
      "epoch": 1.3,
      "grad_norm": 2.2633233070373535,
      "learning_rate": 0.004353703703703703,
      "loss": 0.7172,
      "step": 700
    },
    {
      "epoch": 1.31,
      "grad_norm": 2.6559178829193115,
      "learning_rate": 0.004344444444444445,
      "loss": 0.7276,
      "step": 710
    },
    {
      "epoch": 1.33,
      "grad_norm": 2.212648391723633,
      "learning_rate": 0.004335185185185185,
      "loss": 0.6017,
      "step": 720
    },
    {
      "epoch": 1.35,
      "grad_norm": 1.5736876726150513,
      "learning_rate": 0.004325925925925926,
      "loss": 0.6352,
      "step": 730
    },
    {
      "epoch": 1.37,
      "grad_norm": 1.8303929567337036,
      "learning_rate": 0.004316666666666667,
      "loss": 0.5693,
      "step": 740
    },
    {
      "epoch": 1.39,
      "grad_norm": 1.7651495933532715,
      "learning_rate": 0.004307407407407407,
      "loss": 0.5643,
      "step": 750
    },
    {
      "epoch": 1.41,
      "grad_norm": 2.3067328929901123,
      "learning_rate": 0.004298148148148148,
      "loss": 0.6661,
      "step": 760
    },
    {
      "epoch": 1.42,
      "grad_norm": 2.121184825897217,
      "learning_rate": 0.004288888888888889,
      "loss": 0.5675,
      "step": 770
    },
    {
      "epoch": 1.44,
      "grad_norm": 1.3322380781173706,
      "learning_rate": 0.00427962962962963,
      "loss": 0.6631,
      "step": 780
    },
    {
      "epoch": 1.46,
      "grad_norm": 1.6407716274261475,
      "learning_rate": 0.0042703703703703706,
      "loss": 0.6184,
      "step": 790
    },
    {
      "epoch": 1.48,
      "grad_norm": 2.4977192878723145,
      "learning_rate": 0.004261111111111111,
      "loss": 0.5324,
      "step": 800
    },
    {
      "epoch": 1.5,
      "grad_norm": 1.2957355976104736,
      "learning_rate": 0.004251851851851852,
      "loss": 0.6039,
      "step": 810
    },
    {
      "epoch": 1.52,
      "grad_norm": 1.6499762535095215,
      "learning_rate": 0.004242592592592593,
      "loss": 0.6452,
      "step": 820
    },
    {
      "epoch": 1.54,
      "grad_norm": 2.024014949798584,
      "learning_rate": 0.004233333333333334,
      "loss": 0.673,
      "step": 830
    },
    {
      "epoch": 1.55,
      "grad_norm": 1.1818047761917114,
      "learning_rate": 0.004224074074074074,
      "loss": 0.6826,
      "step": 840
    },
    {
      "epoch": 1.57,
      "grad_norm": 1.610490083694458,
      "learning_rate": 0.0042148148148148145,
      "loss": 0.6284,
      "step": 850
    },
    {
      "epoch": 1.59,
      "grad_norm": 4.2171311378479,
      "learning_rate": 0.004205555555555556,
      "loss": 0.583,
      "step": 860
    },
    {
      "epoch": 1.61,
      "grad_norm": 1.984252691268921,
      "learning_rate": 0.004196296296296296,
      "loss": 0.6036,
      "step": 870
    },
    {
      "epoch": 1.63,
      "grad_norm": 1.9518765211105347,
      "learning_rate": 0.004187037037037037,
      "loss": 0.5909,
      "step": 880
    },
    {
      "epoch": 1.65,
      "grad_norm": 2.7360193729400635,
      "learning_rate": 0.0041777777777777785,
      "loss": 0.6071,
      "step": 890
    },
    {
      "epoch": 1.67,
      "grad_norm": 3.141552209854126,
      "learning_rate": 0.0041685185185185184,
      "loss": 0.5846,
      "step": 900
    },
    {
      "epoch": 1.68,
      "grad_norm": 3.816471815109253,
      "learning_rate": 0.004159259259259259,
      "loss": 0.6058,
      "step": 910
    },
    {
      "epoch": 1.7,
      "grad_norm": 1.957501769065857,
      "learning_rate": 0.00415,
      "loss": 0.6491,
      "step": 920
    },
    {
      "epoch": 1.72,
      "grad_norm": 7.639843463897705,
      "learning_rate": 0.004140740740740741,
      "loss": 0.6144,
      "step": 930
    },
    {
      "epoch": 1.74,
      "grad_norm": 2.967031955718994,
      "learning_rate": 0.004131481481481482,
      "loss": 0.7173,
      "step": 940
    },
    {
      "epoch": 1.76,
      "grad_norm": 2.3182485103607178,
      "learning_rate": 0.004122222222222222,
      "loss": 0.6002,
      "step": 950
    },
    {
      "epoch": 1.78,
      "grad_norm": 2.3655409812927246,
      "learning_rate": 0.004112962962962962,
      "loss": 0.5993,
      "step": 960
    },
    {
      "epoch": 1.79,
      "grad_norm": 2.82918643951416,
      "learning_rate": 0.004103703703703704,
      "loss": 0.5827,
      "step": 970
    },
    {
      "epoch": 1.81,
      "grad_norm": 3.4129443168640137,
      "learning_rate": 0.004094444444444445,
      "loss": 0.6606,
      "step": 980
    },
    {
      "epoch": 1.83,
      "grad_norm": 2.3152365684509277,
      "learning_rate": 0.004085185185185185,
      "loss": 0.6762,
      "step": 990
    },
    {
      "epoch": 1.85,
      "grad_norm": 1.8874484300613403,
      "learning_rate": 0.004075925925925926,
      "loss": 0.7619,
      "step": 1000
    },
    {
      "epoch": 1.87,
      "grad_norm": 1.565685510635376,
      "learning_rate": 0.004066666666666667,
      "loss": 0.7038,
      "step": 1010
    },
    {
      "epoch": 1.89,
      "grad_norm": 3.39355206489563,
      "learning_rate": 0.004057407407407407,
      "loss": 0.6543,
      "step": 1020
    },
    {
      "epoch": 1.91,
      "grad_norm": 1.670486330986023,
      "learning_rate": 0.004048148148148148,
      "loss": 0.7117,
      "step": 1030
    },
    {
      "epoch": 1.92,
      "grad_norm": 2.0435540676116943,
      "learning_rate": 0.004038888888888889,
      "loss": 0.7173,
      "step": 1040
    },
    {
      "epoch": 1.94,
      "grad_norm": 2.260481119155884,
      "learning_rate": 0.0040296296296296295,
      "loss": 0.6442,
      "step": 1050
    },
    {
      "epoch": 1.96,
      "grad_norm": 4.03023099899292,
      "learning_rate": 0.00402037037037037,
      "loss": 0.5518,
      "step": 1060
    },
    {
      "epoch": 1.98,
      "grad_norm": 2.807311773300171,
      "learning_rate": 0.004011111111111111,
      "loss": 0.5961,
      "step": 1070
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.02470326423645,
      "learning_rate": 0.004001851851851852,
      "loss": 0.6346,
      "step": 1080
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9759667231551379,
      "eval_f1": 0.9769796000181948,
      "eval_loss": 0.0756431594491005,
      "eval_precision": 0.9799222297554231,
      "eval_recall": 0.9751785999265923,
      "eval_runtime": 105.7688,
      "eval_samples_per_second": 61.37,
      "eval_steps_per_second": 3.839,
      "step": 1081
    },
    {
      "epoch": 2.02,
      "grad_norm": 2.4306721687316895,
      "learning_rate": 0.003992592592592593,
      "loss": 0.5623,
      "step": 1090
    },
    {
      "epoch": 2.04,
      "grad_norm": 3.5204544067382812,
      "learning_rate": 0.0039833333333333335,
      "loss": 0.6119,
      "step": 1100
    },
    {
      "epoch": 2.05,
      "grad_norm": 1.6892510652542114,
      "learning_rate": 0.003974074074074074,
      "loss": 0.6725,
      "step": 1110
    },
    {
      "epoch": 2.07,
      "grad_norm": 1.9901163578033447,
      "learning_rate": 0.003964814814814815,
      "loss": 0.6753,
      "step": 1120
    },
    {
      "epoch": 2.09,
      "grad_norm": 2.7705817222595215,
      "learning_rate": 0.003955555555555556,
      "loss": 0.6667,
      "step": 1130
    },
    {
      "epoch": 2.11,
      "grad_norm": 2.376143217086792,
      "learning_rate": 0.003946296296296296,
      "loss": 0.6805,
      "step": 1140
    },
    {
      "epoch": 2.13,
      "grad_norm": 2.719224214553833,
      "learning_rate": 0.0039370370370370375,
      "loss": 0.66,
      "step": 1150
    },
    {
      "epoch": 2.15,
      "grad_norm": 2.0327024459838867,
      "learning_rate": 0.003927777777777778,
      "loss": 0.6336,
      "step": 1160
    },
    {
      "epoch": 2.16,
      "grad_norm": 1.984775424003601,
      "learning_rate": 0.003918518518518518,
      "loss": 0.5996,
      "step": 1170
    },
    {
      "epoch": 2.18,
      "grad_norm": 2.203216075897217,
      "learning_rate": 0.003909259259259259,
      "loss": 0.6567,
      "step": 1180
    },
    {
      "epoch": 2.2,
      "grad_norm": 1.6019185781478882,
      "learning_rate": 0.0039000000000000003,
      "loss": 0.6497,
      "step": 1190
    },
    {
      "epoch": 2.22,
      "grad_norm": 1.9226101636886597,
      "learning_rate": 0.0038907407407407406,
      "loss": 0.6658,
      "step": 1200
    },
    {
      "epoch": 2.24,
      "grad_norm": 3.1916518211364746,
      "learning_rate": 0.0038814814814814814,
      "loss": 0.7171,
      "step": 1210
    },
    {
      "epoch": 2.26,
      "grad_norm": 1.9380624294281006,
      "learning_rate": 0.0038722222222222226,
      "loss": 0.7219,
      "step": 1220
    },
    {
      "epoch": 2.28,
      "grad_norm": 2.232342481613159,
      "learning_rate": 0.003862962962962963,
      "loss": 0.7147,
      "step": 1230
    },
    {
      "epoch": 2.29,
      "grad_norm": 2.556619644165039,
      "learning_rate": 0.003853703703703704,
      "loss": 0.6061,
      "step": 1240
    },
    {
      "epoch": 2.31,
      "grad_norm": 4.261257648468018,
      "learning_rate": 0.003844444444444444,
      "loss": 0.6232,
      "step": 1250
    },
    {
      "epoch": 2.33,
      "grad_norm": 5.815138339996338,
      "learning_rate": 0.0038351851851851854,
      "loss": 0.6059,
      "step": 1260
    },
    {
      "epoch": 2.35,
      "grad_norm": 2.543003797531128,
      "learning_rate": 0.003825925925925926,
      "loss": 0.7635,
      "step": 1270
    },
    {
      "epoch": 2.37,
      "grad_norm": 3.6232597827911377,
      "learning_rate": 0.0038166666666666666,
      "loss": 0.6533,
      "step": 1280
    },
    {
      "epoch": 2.39,
      "grad_norm": 3.13364839553833,
      "learning_rate": 0.003807407407407408,
      "loss": 0.6824,
      "step": 1290
    },
    {
      "epoch": 2.41,
      "grad_norm": 1.5059545040130615,
      "learning_rate": 0.003798148148148148,
      "loss": 0.6682,
      "step": 1300
    },
    {
      "epoch": 2.42,
      "grad_norm": 2.015592575073242,
      "learning_rate": 0.003788888888888889,
      "loss": 0.6536,
      "step": 1310
    },
    {
      "epoch": 2.44,
      "grad_norm": 1.8305333852767944,
      "learning_rate": 0.0037796296296296297,
      "loss": 0.6503,
      "step": 1320
    },
    {
      "epoch": 2.46,
      "grad_norm": 2.674013376235962,
      "learning_rate": 0.0037703703703703705,
      "loss": 0.6235,
      "step": 1330
    },
    {
      "epoch": 2.48,
      "grad_norm": 4.233829021453857,
      "learning_rate": 0.0037611111111111113,
      "loss": 0.7144,
      "step": 1340
    },
    {
      "epoch": 2.5,
      "grad_norm": 2.49525785446167,
      "learning_rate": 0.0037518518518518517,
      "loss": 0.5574,
      "step": 1350
    },
    {
      "epoch": 2.52,
      "grad_norm": 2.353803873062134,
      "learning_rate": 0.0037425925925925925,
      "loss": 0.687,
      "step": 1360
    },
    {
      "epoch": 2.53,
      "grad_norm": 2.101120710372925,
      "learning_rate": 0.0037333333333333337,
      "loss": 0.5579,
      "step": 1370
    },
    {
      "epoch": 2.55,
      "grad_norm": 1.9760828018188477,
      "learning_rate": 0.003724074074074074,
      "loss": 0.7199,
      "step": 1380
    },
    {
      "epoch": 2.57,
      "grad_norm": 3.785053014755249,
      "learning_rate": 0.003714814814814815,
      "loss": 0.6915,
      "step": 1390
    },
    {
      "epoch": 2.59,
      "grad_norm": 3.4531219005584717,
      "learning_rate": 0.0037055555555555557,
      "loss": 0.6379,
      "step": 1400
    },
    {
      "epoch": 2.61,
      "grad_norm": 2.45957088470459,
      "learning_rate": 0.0036962962962962965,
      "loss": 0.6632,
      "step": 1410
    },
    {
      "epoch": 2.63,
      "grad_norm": 2.7392988204956055,
      "learning_rate": 0.0036870370370370373,
      "loss": 0.6151,
      "step": 1420
    },
    {
      "epoch": 2.65,
      "grad_norm": 1.5700833797454834,
      "learning_rate": 0.0036777777777777776,
      "loss": 0.6362,
      "step": 1430
    },
    {
      "epoch": 2.66,
      "grad_norm": 2.2544972896575928,
      "learning_rate": 0.003668518518518519,
      "loss": 0.6567,
      "step": 1440
    },
    {
      "epoch": 2.68,
      "grad_norm": 2.2333929538726807,
      "learning_rate": 0.0036592592592592592,
      "loss": 0.6005,
      "step": 1450
    },
    {
      "epoch": 2.7,
      "grad_norm": 1.6665147542953491,
      "learning_rate": 0.00365,
      "loss": 0.5697,
      "step": 1460
    },
    {
      "epoch": 2.72,
      "grad_norm": 2.4265899658203125,
      "learning_rate": 0.0036407407407407404,
      "loss": 0.5576,
      "step": 1470
    },
    {
      "epoch": 2.74,
      "grad_norm": 2.1805577278137207,
      "learning_rate": 0.0036314814814814816,
      "loss": 0.5729,
      "step": 1480
    },
    {
      "epoch": 2.76,
      "grad_norm": 1.8916809558868408,
      "learning_rate": 0.0036222222222222224,
      "loss": 0.6201,
      "step": 1490
    },
    {
      "epoch": 2.78,
      "grad_norm": 1.9663447141647339,
      "learning_rate": 0.003612962962962963,
      "loss": 0.5541,
      "step": 1500
    },
    {
      "epoch": 2.79,
      "grad_norm": 1.565356731414795,
      "learning_rate": 0.003603703703703704,
      "loss": 0.6453,
      "step": 1510
    },
    {
      "epoch": 2.81,
      "grad_norm": 1.5807263851165771,
      "learning_rate": 0.003594444444444445,
      "loss": 0.5988,
      "step": 1520
    },
    {
      "epoch": 2.83,
      "grad_norm": 2.013056993484497,
      "learning_rate": 0.003585185185185185,
      "loss": 0.5613,
      "step": 1530
    },
    {
      "epoch": 2.85,
      "grad_norm": 2.1835033893585205,
      "learning_rate": 0.003575925925925926,
      "loss": 0.68,
      "step": 1540
    },
    {
      "epoch": 2.87,
      "grad_norm": 3.3731331825256348,
      "learning_rate": 0.0035666666666666668,
      "loss": 0.6693,
      "step": 1550
    },
    {
      "epoch": 2.89,
      "grad_norm": 2.4271302223205566,
      "learning_rate": 0.0035574074074074076,
      "loss": 0.5841,
      "step": 1560
    },
    {
      "epoch": 2.9,
      "grad_norm": 3.0260560512542725,
      "learning_rate": 0.003548148148148148,
      "loss": 0.5775,
      "step": 1570
    },
    {
      "epoch": 2.92,
      "grad_norm": 2.215425968170166,
      "learning_rate": 0.0035388888888888887,
      "loss": 0.6203,
      "step": 1580
    },
    {
      "epoch": 2.94,
      "grad_norm": 2.4513442516326904,
      "learning_rate": 0.00352962962962963,
      "loss": 0.6032,
      "step": 1590
    },
    {
      "epoch": 2.96,
      "grad_norm": 3.0825629234313965,
      "learning_rate": 0.0035212962962962962,
      "loss": 0.5534,
      "step": 1600
    },
    {
      "epoch": 2.98,
      "grad_norm": 3.933262825012207,
      "learning_rate": 0.003512037037037037,
      "loss": 0.5943,
      "step": 1610
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.9379372596740723,
      "learning_rate": 0.0035027777777777783,
      "loss": 0.6405,
      "step": 1620
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9553227545832691,
      "eval_f1": 0.9537075013208948,
      "eval_loss": 0.13096250593662262,
      "eval_precision": 0.959969032218872,
      "eval_recall": 0.9515227670577108,
      "eval_runtime": 105.8838,
      "eval_samples_per_second": 61.303,
      "eval_steps_per_second": 3.834,
      "step": 1621
    },
    {
      "epoch": 3.02,
      "grad_norm": 4.636434078216553,
      "learning_rate": 0.0034935185185185186,
      "loss": 0.5521,
      "step": 1630
    },
    {
      "epoch": 3.03,
      "grad_norm": 2.875552177429199,
      "learning_rate": 0.0034842592592592594,
      "loss": 0.584,
      "step": 1640
    },
    {
      "epoch": 3.05,
      "grad_norm": 2.166994333267212,
      "learning_rate": 0.003475,
      "loss": 0.6998,
      "step": 1650
    },
    {
      "epoch": 3.07,
      "grad_norm": 2.1271350383758545,
      "learning_rate": 0.003465740740740741,
      "loss": 0.5648,
      "step": 1660
    },
    {
      "epoch": 3.09,
      "grad_norm": 2.3423948287963867,
      "learning_rate": 0.0034564814814814814,
      "loss": 0.6761,
      "step": 1670
    },
    {
      "epoch": 3.11,
      "grad_norm": 4.546567440032959,
      "learning_rate": 0.003447222222222222,
      "loss": 0.6555,
      "step": 1680
    },
    {
      "epoch": 3.13,
      "grad_norm": 2.9972333908081055,
      "learning_rate": 0.0034379629629629634,
      "loss": 0.6147,
      "step": 1690
    },
    {
      "epoch": 3.15,
      "grad_norm": 1.9627290964126587,
      "learning_rate": 0.0034287037037037038,
      "loss": 0.6231,
      "step": 1700
    },
    {
      "epoch": 3.16,
      "grad_norm": 3.252190589904785,
      "learning_rate": 0.0034194444444444446,
      "loss": 0.6926,
      "step": 1710
    },
    {
      "epoch": 3.18,
      "grad_norm": 3.4038755893707275,
      "learning_rate": 0.003410185185185185,
      "loss": 0.6724,
      "step": 1720
    },
    {
      "epoch": 3.2,
      "grad_norm": 2.501878261566162,
      "learning_rate": 0.003400925925925926,
      "loss": 0.6141,
      "step": 1730
    },
    {
      "epoch": 3.22,
      "grad_norm": 2.0606675148010254,
      "learning_rate": 0.003391666666666667,
      "loss": 0.5758,
      "step": 1740
    },
    {
      "epoch": 3.24,
      "grad_norm": 2.631194829940796,
      "learning_rate": 0.0033824074074074073,
      "loss": 0.6344,
      "step": 1750
    },
    {
      "epoch": 3.26,
      "grad_norm": 2.717015027999878,
      "learning_rate": 0.003373148148148148,
      "loss": 0.6729,
      "step": 1760
    },
    {
      "epoch": 3.27,
      "grad_norm": 2.4547502994537354,
      "learning_rate": 0.003363888888888889,
      "loss": 0.6162,
      "step": 1770
    },
    {
      "epoch": 3.29,
      "grad_norm": 4.384427070617676,
      "learning_rate": 0.0033546296296296297,
      "loss": 0.5629,
      "step": 1780
    },
    {
      "epoch": 3.31,
      "grad_norm": 3.193366765975952,
      "learning_rate": 0.00334537037037037,
      "loss": 0.6054,
      "step": 1790
    },
    {
      "epoch": 3.33,
      "grad_norm": 2.8141987323760986,
      "learning_rate": 0.0033361111111111113,
      "loss": 0.6117,
      "step": 1800
    },
    {
      "epoch": 3.35,
      "grad_norm": 1.7347623109817505,
      "learning_rate": 0.003326851851851852,
      "loss": 0.6268,
      "step": 1810
    },
    {
      "epoch": 3.37,
      "grad_norm": 5.643610000610352,
      "learning_rate": 0.0033175925925925925,
      "loss": 0.5513,
      "step": 1820
    },
    {
      "epoch": 3.39,
      "grad_norm": 3.0566015243530273,
      "learning_rate": 0.0033083333333333333,
      "loss": 0.7019,
      "step": 1830
    },
    {
      "epoch": 3.4,
      "grad_norm": 2.0485594272613525,
      "learning_rate": 0.0032990740740740745,
      "loss": 0.5568,
      "step": 1840
    },
    {
      "epoch": 3.42,
      "grad_norm": 2.5610201358795166,
      "learning_rate": 0.003289814814814815,
      "loss": 0.5935,
      "step": 1850
    },
    {
      "epoch": 3.44,
      "grad_norm": 14.397619247436523,
      "learning_rate": 0.0032805555555555557,
      "loss": 0.5506,
      "step": 1860
    },
    {
      "epoch": 3.46,
      "grad_norm": 2.7251882553100586,
      "learning_rate": 0.003271296296296296,
      "loss": 0.6064,
      "step": 1870
    },
    {
      "epoch": 3.48,
      "grad_norm": 2.093822479248047,
      "learning_rate": 0.0032620370370370372,
      "loss": 0.7011,
      "step": 1880
    },
    {
      "epoch": 3.5,
      "grad_norm": 2.208249807357788,
      "learning_rate": 0.003252777777777778,
      "loss": 0.6524,
      "step": 1890
    },
    {
      "epoch": 3.52,
      "grad_norm": 11.702254295349121,
      "learning_rate": 0.0032435185185185184,
      "loss": 0.6089,
      "step": 1900
    },
    {
      "epoch": 3.53,
      "grad_norm": 3.271688938140869,
      "learning_rate": 0.0032342592592592596,
      "loss": 0.7139,
      "step": 1910
    },
    {
      "epoch": 3.55,
      "grad_norm": 5.287911415100098,
      "learning_rate": 0.003225,
      "loss": 0.575,
      "step": 1920
    },
    {
      "epoch": 3.57,
      "grad_norm": 3.2775847911834717,
      "learning_rate": 0.003215740740740741,
      "loss": 0.6546,
      "step": 1930
    },
    {
      "epoch": 3.59,
      "grad_norm": 2.3923332691192627,
      "learning_rate": 0.003206481481481481,
      "loss": 0.5964,
      "step": 1940
    },
    {
      "epoch": 3.61,
      "grad_norm": 2.3694956302642822,
      "learning_rate": 0.0031972222222222224,
      "loss": 0.6172,
      "step": 1950
    },
    {
      "epoch": 3.63,
      "grad_norm": 3.5944793224334717,
      "learning_rate": 0.003187962962962963,
      "loss": 0.574,
      "step": 1960
    },
    {
      "epoch": 3.64,
      "grad_norm": 1.8562227487564087,
      "learning_rate": 0.0031787037037037036,
      "loss": 0.6079,
      "step": 1970
    },
    {
      "epoch": 3.66,
      "grad_norm": 3.2188656330108643,
      "learning_rate": 0.0031694444444444443,
      "loss": 0.5484,
      "step": 1980
    },
    {
      "epoch": 3.68,
      "grad_norm": 3.9696381092071533,
      "learning_rate": 0.0031601851851851856,
      "loss": 0.6438,
      "step": 1990
    },
    {
      "epoch": 3.7,
      "grad_norm": 1.935719609260559,
      "learning_rate": 0.003150925925925926,
      "loss": 0.6175,
      "step": 2000
    },
    {
      "epoch": 3.72,
      "grad_norm": 2.686809778213501,
      "learning_rate": 0.0031416666666666667,
      "loss": 0.5267,
      "step": 2010
    },
    {
      "epoch": 3.74,
      "grad_norm": 1.6655242443084717,
      "learning_rate": 0.0031324074074074075,
      "loss": 0.6053,
      "step": 2020
    },
    {
      "epoch": 3.76,
      "grad_norm": 2.1189332008361816,
      "learning_rate": 0.0031231481481481483,
      "loss": 0.5282,
      "step": 2030
    },
    {
      "epoch": 3.77,
      "grad_norm": 2.889960289001465,
      "learning_rate": 0.0031138888888888887,
      "loss": 0.61,
      "step": 2040
    },
    {
      "epoch": 3.79,
      "grad_norm": 3.7036290168762207,
      "learning_rate": 0.0031046296296296295,
      "loss": 0.6091,
      "step": 2050
    },
    {
      "epoch": 3.81,
      "grad_norm": 4.608720302581787,
      "learning_rate": 0.0030953703703703707,
      "loss": 0.6725,
      "step": 2060
    },
    {
      "epoch": 3.83,
      "grad_norm": 3.6915183067321777,
      "learning_rate": 0.003086111111111111,
      "loss": 0.6051,
      "step": 2070
    },
    {
      "epoch": 3.85,
      "grad_norm": 5.5720133781433105,
      "learning_rate": 0.003076851851851852,
      "loss": 0.6552,
      "step": 2080
    },
    {
      "epoch": 3.87,
      "grad_norm": 3.984224319458008,
      "learning_rate": 0.0030675925925925922,
      "loss": 0.6326,
      "step": 2090
    },
    {
      "epoch": 3.89,
      "grad_norm": 4.13102388381958,
      "learning_rate": 0.0030583333333333335,
      "loss": 0.6286,
      "step": 2100
    },
    {
      "epoch": 3.9,
      "grad_norm": 5.23762845993042,
      "learning_rate": 0.0030490740740740743,
      "loss": 0.6575,
      "step": 2110
    },
    {
      "epoch": 3.92,
      "grad_norm": 3.083714008331299,
      "learning_rate": 0.0030398148148148146,
      "loss": 0.6214,
      "step": 2120
    },
    {
      "epoch": 3.94,
      "grad_norm": 2.6413919925689697,
      "learning_rate": 0.003030555555555556,
      "loss": 0.6019,
      "step": 2130
    },
    {
      "epoch": 3.96,
      "grad_norm": 1.659439206123352,
      "learning_rate": 0.0030212962962962962,
      "loss": 0.6417,
      "step": 2140
    },
    {
      "epoch": 3.98,
      "grad_norm": 3.3802456855773926,
      "learning_rate": 0.003012037037037037,
      "loss": 0.5898,
      "step": 2150
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.054283618927002,
      "learning_rate": 0.003002777777777778,
      "loss": 0.5005,
      "step": 2160
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9662609767370205,
      "eval_f1": 0.972867802534218,
      "eval_loss": 0.1138468086719513,
      "eval_precision": 0.9756603079675337,
      "eval_recall": 0.971821943372282,
      "eval_runtime": 105.7692,
      "eval_samples_per_second": 61.369,
      "eval_steps_per_second": 3.839,
      "step": 2162
    },
    {
      "epoch": 4.01,
      "grad_norm": 3.2501673698425293,
      "learning_rate": 0.0029935185185185186,
      "loss": 0.6619,
      "step": 2170
    },
    {
      "epoch": 4.03,
      "grad_norm": 3.0431628227233887,
      "learning_rate": 0.0029842592592592594,
      "loss": 0.6635,
      "step": 2180
    },
    {
      "epoch": 4.05,
      "grad_norm": 2.6278398036956787,
      "learning_rate": 0.0029749999999999998,
      "loss": 0.6468,
      "step": 2190
    },
    {
      "epoch": 4.07,
      "grad_norm": 1.7868436574935913,
      "learning_rate": 0.002965740740740741,
      "loss": 0.6325,
      "step": 2200
    },
    {
      "epoch": 4.09,
      "grad_norm": 4.32276725769043,
      "learning_rate": 0.002956481481481482,
      "loss": 0.5745,
      "step": 2210
    },
    {
      "epoch": 4.11,
      "grad_norm": 1.9450454711914062,
      "learning_rate": 0.002947222222222222,
      "loss": 0.5264,
      "step": 2220
    },
    {
      "epoch": 4.13,
      "grad_norm": 8.342818260192871,
      "learning_rate": 0.002937962962962963,
      "loss": 0.5547,
      "step": 2230
    },
    {
      "epoch": 4.14,
      "grad_norm": 4.070902347564697,
      "learning_rate": 0.002928703703703704,
      "loss": 0.5232,
      "step": 2240
    },
    {
      "epoch": 4.16,
      "grad_norm": 2.327827215194702,
      "learning_rate": 0.0029194444444444446,
      "loss": 0.5651,
      "step": 2250
    },
    {
      "epoch": 4.18,
      "grad_norm": 2.1814072132110596,
      "learning_rate": 0.0029101851851851854,
      "loss": 0.6027,
      "step": 2260
    },
    {
      "epoch": 4.2,
      "grad_norm": 3.3634531497955322,
      "learning_rate": 0.0029009259259259257,
      "loss": 0.579,
      "step": 2270
    },
    {
      "epoch": 4.22,
      "grad_norm": 2.795017957687378,
      "learning_rate": 0.002891666666666667,
      "loss": 0.6487,
      "step": 2280
    },
    {
      "epoch": 4.24,
      "grad_norm": 2.5090396404266357,
      "learning_rate": 0.0028824074074074073,
      "loss": 0.6351,
      "step": 2290
    },
    {
      "epoch": 4.26,
      "grad_norm": 4.025092601776123,
      "learning_rate": 0.002873148148148148,
      "loss": 0.6395,
      "step": 2300
    },
    {
      "epoch": 4.27,
      "grad_norm": 2.139432668685913,
      "learning_rate": 0.0028638888888888893,
      "loss": 0.5465,
      "step": 2310
    },
    {
      "epoch": 4.29,
      "grad_norm": 2.7012085914611816,
      "learning_rate": 0.0028546296296296297,
      "loss": 0.5872,
      "step": 2320
    },
    {
      "epoch": 4.31,
      "grad_norm": 3.428055763244629,
      "learning_rate": 0.0028453703703703705,
      "loss": 0.5588,
      "step": 2330
    },
    {
      "epoch": 4.33,
      "grad_norm": 3.3620471954345703,
      "learning_rate": 0.002836111111111111,
      "loss": 0.5958,
      "step": 2340
    },
    {
      "epoch": 4.35,
      "grad_norm": 1.7665202617645264,
      "learning_rate": 0.002826851851851852,
      "loss": 0.4828,
      "step": 2350
    },
    {
      "epoch": 4.37,
      "grad_norm": 2.4530186653137207,
      "learning_rate": 0.002817592592592593,
      "loss": 0.6112,
      "step": 2360
    },
    {
      "epoch": 4.38,
      "grad_norm": 1.5522198677062988,
      "learning_rate": 0.0028083333333333333,
      "loss": 0.6276,
      "step": 2370
    },
    {
      "epoch": 4.4,
      "grad_norm": 2.6151957511901855,
      "learning_rate": 0.002799074074074074,
      "loss": 0.5665,
      "step": 2380
    },
    {
      "epoch": 4.42,
      "grad_norm": 2.1417715549468994,
      "learning_rate": 0.002789814814814815,
      "loss": 0.5213,
      "step": 2390
    },
    {
      "epoch": 4.44,
      "grad_norm": 3.485398530960083,
      "learning_rate": 0.0027805555555555556,
      "loss": 0.6137,
      "step": 2400
    },
    {
      "epoch": 4.46,
      "grad_norm": 7.003100395202637,
      "learning_rate": 0.002771296296296296,
      "loss": 0.5872,
      "step": 2410
    },
    {
      "epoch": 4.48,
      "grad_norm": 3.0954577922821045,
      "learning_rate": 0.0027620370370370372,
      "loss": 0.5891,
      "step": 2420
    },
    {
      "epoch": 4.5,
      "grad_norm": 6.439086437225342,
      "learning_rate": 0.002752777777777778,
      "loss": 0.6085,
      "step": 2430
    },
    {
      "epoch": 4.51,
      "grad_norm": 3.5923094749450684,
      "learning_rate": 0.0027435185185185184,
      "loss": 0.6787,
      "step": 2440
    },
    {
      "epoch": 4.53,
      "grad_norm": 2.043170690536499,
      "learning_rate": 0.002734259259259259,
      "loss": 0.5666,
      "step": 2450
    },
    {
      "epoch": 4.55,
      "grad_norm": 2.8164470195770264,
      "learning_rate": 0.0027250000000000004,
      "loss": 0.6586,
      "step": 2460
    },
    {
      "epoch": 4.57,
      "grad_norm": 2.2632648944854736,
      "learning_rate": 0.002715740740740741,
      "loss": 0.6106,
      "step": 2470
    },
    {
      "epoch": 4.59,
      "grad_norm": 2.98234224319458,
      "learning_rate": 0.0027064814814814816,
      "loss": 0.5545,
      "step": 2480
    },
    {
      "epoch": 4.61,
      "grad_norm": 2.2962846755981445,
      "learning_rate": 0.002697222222222222,
      "loss": 0.5755,
      "step": 2490
    },
    {
      "epoch": 4.63,
      "grad_norm": 4.67765474319458,
      "learning_rate": 0.002687962962962963,
      "loss": 0.4411,
      "step": 2500
    },
    {
      "epoch": 4.64,
      "grad_norm": 3.6627418994903564,
      "learning_rate": 0.0026787037037037035,
      "loss": 0.63,
      "step": 2510
    },
    {
      "epoch": 4.66,
      "grad_norm": 3.7983245849609375,
      "learning_rate": 0.0026694444444444443,
      "loss": 0.5024,
      "step": 2520
    },
    {
      "epoch": 4.68,
      "grad_norm": 2.797543525695801,
      "learning_rate": 0.0026601851851851856,
      "loss": 0.5812,
      "step": 2530
    },
    {
      "epoch": 4.7,
      "grad_norm": 2.688131332397461,
      "learning_rate": 0.002650925925925926,
      "loss": 0.6246,
      "step": 2540
    },
    {
      "epoch": 4.72,
      "grad_norm": 3.1840107440948486,
      "learning_rate": 0.0026416666666666667,
      "loss": 0.5858,
      "step": 2550
    },
    {
      "epoch": 4.74,
      "grad_norm": 1.386940598487854,
      "learning_rate": 0.002632407407407407,
      "loss": 0.5242,
      "step": 2560
    },
    {
      "epoch": 4.75,
      "grad_norm": 2.7501676082611084,
      "learning_rate": 0.0026231481481481483,
      "loss": 0.5272,
      "step": 2570
    },
    {
      "epoch": 4.77,
      "grad_norm": 2.656128406524658,
      "learning_rate": 0.002613888888888889,
      "loss": 0.5537,
      "step": 2580
    },
    {
      "epoch": 4.79,
      "grad_norm": 3.3853468894958496,
      "learning_rate": 0.0026046296296296295,
      "loss": 0.6347,
      "step": 2590
    },
    {
      "epoch": 4.81,
      "grad_norm": 3.0706825256347656,
      "learning_rate": 0.0025953703703703703,
      "loss": 0.5902,
      "step": 2600
    },
    {
      "epoch": 4.83,
      "grad_norm": 2.470815658569336,
      "learning_rate": 0.0025861111111111115,
      "loss": 0.5707,
      "step": 2610
    },
    {
      "epoch": 4.85,
      "grad_norm": 2.8364624977111816,
      "learning_rate": 0.002576851851851852,
      "loss": 0.6205,
      "step": 2620
    },
    {
      "epoch": 4.87,
      "grad_norm": 2.347895860671997,
      "learning_rate": 0.0025675925925925927,
      "loss": 0.6068,
      "step": 2630
    },
    {
      "epoch": 4.88,
      "grad_norm": 2.5347235202789307,
      "learning_rate": 0.0025583333333333335,
      "loss": 0.5555,
      "step": 2640
    },
    {
      "epoch": 4.9,
      "grad_norm": 1.7792493104934692,
      "learning_rate": 0.0025490740740740743,
      "loss": 0.6124,
      "step": 2650
    },
    {
      "epoch": 4.92,
      "grad_norm": 1.5881773233413696,
      "learning_rate": 0.0025398148148148146,
      "loss": 0.5311,
      "step": 2660
    },
    {
      "epoch": 4.94,
      "grad_norm": 2.866182327270508,
      "learning_rate": 0.0025305555555555554,
      "loss": 0.5833,
      "step": 2670
    },
    {
      "epoch": 4.96,
      "grad_norm": 2.6761910915374756,
      "learning_rate": 0.0025212962962962967,
      "loss": 0.5367,
      "step": 2680
    },
    {
      "epoch": 4.98,
      "grad_norm": 3.0073177814483643,
      "learning_rate": 0.002512037037037037,
      "loss": 0.5316,
      "step": 2690
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.769132375717163,
      "learning_rate": 0.002502777777777778,
      "loss": 0.5669,
      "step": 2700
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9602526575258049,
      "eval_f1": 0.9665087531565827,
      "eval_loss": 0.11420344561338425,
      "eval_precision": 0.9704223501308576,
      "eval_recall": 0.9647162237423828,
      "eval_runtime": 105.6456,
      "eval_samples_per_second": 61.441,
      "eval_steps_per_second": 3.843,
      "step": 2702
    },
    {
      "epoch": 5.01,
      "grad_norm": 2.4607672691345215,
      "learning_rate": 0.0024935185185185186,
      "loss": 0.5564,
      "step": 2710
    },
    {
      "epoch": 5.03,
      "grad_norm": 1.7590909004211426,
      "learning_rate": 0.0024842592592592594,
      "loss": 0.5319,
      "step": 2720
    },
    {
      "epoch": 5.05,
      "grad_norm": 6.092741966247559,
      "learning_rate": 0.002475,
      "loss": 0.5901,
      "step": 2730
    },
    {
      "epoch": 5.07,
      "grad_norm": 2.170964479446411,
      "learning_rate": 0.002465740740740741,
      "loss": 0.5454,
      "step": 2740
    },
    {
      "epoch": 5.09,
      "grad_norm": 1.7933937311172485,
      "learning_rate": 0.0024564814814814814,
      "loss": 0.5448,
      "step": 2750
    },
    {
      "epoch": 5.11,
      "grad_norm": 1.6023855209350586,
      "learning_rate": 0.002447222222222222,
      "loss": 0.5337,
      "step": 2760
    },
    {
      "epoch": 5.12,
      "grad_norm": 2.6876323223114014,
      "learning_rate": 0.002437962962962963,
      "loss": 0.621,
      "step": 2770
    },
    {
      "epoch": 5.14,
      "grad_norm": 3.007115125656128,
      "learning_rate": 0.0024287037037037038,
      "loss": 0.6147,
      "step": 2780
    },
    {
      "epoch": 5.16,
      "grad_norm": 1.462850570678711,
      "learning_rate": 0.0024194444444444446,
      "loss": 0.5601,
      "step": 2790
    },
    {
      "epoch": 5.18,
      "grad_norm": 2.3480947017669678,
      "learning_rate": 0.0024101851851851853,
      "loss": 0.5242,
      "step": 2800
    },
    {
      "epoch": 5.2,
      "grad_norm": 2.632812023162842,
      "learning_rate": 0.0024009259259259257,
      "loss": 0.5104,
      "step": 2810
    },
    {
      "epoch": 5.22,
      "grad_norm": 4.110454559326172,
      "learning_rate": 0.0023916666666666665,
      "loss": 0.5169,
      "step": 2820
    },
    {
      "epoch": 5.24,
      "grad_norm": 1.5732221603393555,
      "learning_rate": 0.0023824074074074077,
      "loss": 0.5126,
      "step": 2830
    },
    {
      "epoch": 5.25,
      "grad_norm": 2.7537384033203125,
      "learning_rate": 0.002373148148148148,
      "loss": 0.5494,
      "step": 2840
    },
    {
      "epoch": 5.27,
      "grad_norm": 1.9949324131011963,
      "learning_rate": 0.002363888888888889,
      "loss": 0.6381,
      "step": 2850
    },
    {
      "epoch": 5.29,
      "grad_norm": 2.0324654579162598,
      "learning_rate": 0.0023546296296296297,
      "loss": 0.5944,
      "step": 2860
    },
    {
      "epoch": 5.31,
      "grad_norm": 2.691060781478882,
      "learning_rate": 0.0023453703703703705,
      "loss": 0.5745,
      "step": 2870
    },
    {
      "epoch": 5.33,
      "grad_norm": 3.5940804481506348,
      "learning_rate": 0.0023361111111111113,
      "loss": 0.5423,
      "step": 2880
    },
    {
      "epoch": 5.35,
      "grad_norm": 2.7014036178588867,
      "learning_rate": 0.002326851851851852,
      "loss": 0.5708,
      "step": 2890
    },
    {
      "epoch": 5.37,
      "grad_norm": 4.630244255065918,
      "learning_rate": 0.0023175925925925924,
      "loss": 0.5235,
      "step": 2900
    },
    {
      "epoch": 5.38,
      "grad_norm": 4.1839752197265625,
      "learning_rate": 0.0023083333333333332,
      "loss": 0.5385,
      "step": 2910
    },
    {
      "epoch": 5.4,
      "grad_norm": 2.4296653270721436,
      "learning_rate": 0.002299074074074074,
      "loss": 0.4755,
      "step": 2920
    },
    {
      "epoch": 5.42,
      "grad_norm": 3.140748977661133,
      "learning_rate": 0.002289814814814815,
      "loss": 0.5625,
      "step": 2930
    },
    {
      "epoch": 5.44,
      "grad_norm": 3.9935102462768555,
      "learning_rate": 0.0022805555555555556,
      "loss": 0.5864,
      "step": 2940
    },
    {
      "epoch": 5.46,
      "grad_norm": 3.6175553798675537,
      "learning_rate": 0.0022712962962962964,
      "loss": 0.4942,
      "step": 2950
    },
    {
      "epoch": 5.48,
      "grad_norm": 2.6416540145874023,
      "learning_rate": 0.0022620370370370372,
      "loss": 0.5522,
      "step": 2960
    },
    {
      "epoch": 5.49,
      "grad_norm": 2.512601137161255,
      "learning_rate": 0.0022527777777777776,
      "loss": 0.5134,
      "step": 2970
    },
    {
      "epoch": 5.51,
      "grad_norm": 2.6757636070251465,
      "learning_rate": 0.002243518518518519,
      "loss": 0.4753,
      "step": 2980
    },
    {
      "epoch": 5.53,
      "grad_norm": 5.020870208740234,
      "learning_rate": 0.002234259259259259,
      "loss": 0.5596,
      "step": 2990
    },
    {
      "epoch": 5.55,
      "grad_norm": 1.6189966201782227,
      "learning_rate": 0.002225,
      "loss": 0.4691,
      "step": 3000
    },
    {
      "epoch": 5.57,
      "grad_norm": 1.8074935674667358,
      "learning_rate": 0.0022157407407407408,
      "loss": 0.5388,
      "step": 3010
    },
    {
      "epoch": 5.59,
      "grad_norm": 2.1580889225006104,
      "learning_rate": 0.0022064814814814816,
      "loss": 0.5539,
      "step": 3020
    },
    {
      "epoch": 5.61,
      "grad_norm": 3.8424339294433594,
      "learning_rate": 0.0021972222222222224,
      "loss": 0.5362,
      "step": 3030
    },
    {
      "epoch": 5.62,
      "grad_norm": 2.5609259605407715,
      "learning_rate": 0.002187962962962963,
      "loss": 0.4883,
      "step": 3040
    },
    {
      "epoch": 5.64,
      "grad_norm": 3.077549934387207,
      "learning_rate": 0.002178703703703704,
      "loss": 0.5324,
      "step": 3050
    },
    {
      "epoch": 5.66,
      "grad_norm": 3.0118846893310547,
      "learning_rate": 0.0021694444444444443,
      "loss": 0.483,
      "step": 3060
    },
    {
      "epoch": 5.68,
      "grad_norm": 2.3558125495910645,
      "learning_rate": 0.002160185185185185,
      "loss": 0.5453,
      "step": 3070
    },
    {
      "epoch": 5.7,
      "grad_norm": 3.1456894874572754,
      "learning_rate": 0.002150925925925926,
      "loss": 0.5333,
      "step": 3080
    },
    {
      "epoch": 5.72,
      "grad_norm": 2.671800374984741,
      "learning_rate": 0.0021416666666666667,
      "loss": 0.5446,
      "step": 3090
    },
    {
      "epoch": 5.74,
      "grad_norm": 4.476646423339844,
      "learning_rate": 0.0021324074074074075,
      "loss": 0.5464,
      "step": 3100
    },
    {
      "epoch": 5.75,
      "grad_norm": 2.332273483276367,
      "learning_rate": 0.0021231481481481483,
      "loss": 0.4712,
      "step": 3110
    },
    {
      "epoch": 5.77,
      "grad_norm": 2.072690725326538,
      "learning_rate": 0.0021138888888888887,
      "loss": 0.4919,
      "step": 3120
    },
    {
      "epoch": 5.79,
      "grad_norm": 2.1038475036621094,
      "learning_rate": 0.0021046296296296295,
      "loss": 0.4628,
      "step": 3130
    },
    {
      "epoch": 5.81,
      "grad_norm": 3.188100576400757,
      "learning_rate": 0.0020953703703703707,
      "loss": 0.5378,
      "step": 3140
    },
    {
      "epoch": 5.83,
      "grad_norm": 1.139237403869629,
      "learning_rate": 0.002086111111111111,
      "loss": 0.4754,
      "step": 3150
    },
    {
      "epoch": 5.85,
      "grad_norm": 1.6128747463226318,
      "learning_rate": 0.002076851851851852,
      "loss": 0.4934,
      "step": 3160
    },
    {
      "epoch": 5.86,
      "grad_norm": 2.8483083248138428,
      "learning_rate": 0.0020675925925925927,
      "loss": 0.5424,
      "step": 3170
    },
    {
      "epoch": 5.88,
      "grad_norm": 2.0242919921875,
      "learning_rate": 0.0020583333333333335,
      "loss": 0.5385,
      "step": 3180
    },
    {
      "epoch": 5.9,
      "grad_norm": 2.0240182876586914,
      "learning_rate": 0.002049074074074074,
      "loss": 0.5485,
      "step": 3190
    },
    {
      "epoch": 5.92,
      "grad_norm": 4.104608535766602,
      "learning_rate": 0.002039814814814815,
      "loss": 0.5597,
      "step": 3200
    },
    {
      "epoch": 5.94,
      "grad_norm": 2.405686616897583,
      "learning_rate": 0.0020305555555555554,
      "loss": 0.6002,
      "step": 3210
    },
    {
      "epoch": 5.96,
      "grad_norm": 2.7426254749298096,
      "learning_rate": 0.002021296296296296,
      "loss": 0.544,
      "step": 3220
    },
    {
      "epoch": 5.98,
      "grad_norm": 2.5380473136901855,
      "learning_rate": 0.002012037037037037,
      "loss": 0.52,
      "step": 3230
    },
    {
      "epoch": 5.99,
      "grad_norm": 2.7862181663513184,
      "learning_rate": 0.002002777777777778,
      "loss": 0.5548,
      "step": 3240
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9771991988907719,
      "eval_f1": 0.9795134539181521,
      "eval_loss": 0.056862443685531616,
      "eval_precision": 0.9812055832949096,
      "eval_recall": 0.9784771483290412,
      "eval_runtime": 105.3717,
      "eval_samples_per_second": 61.601,
      "eval_steps_per_second": 3.853,
      "step": 3243
    },
    {
      "epoch": 6.01,
      "grad_norm": 3.968397378921509,
      "learning_rate": 0.0019935185185185186,
      "loss": 0.5039,
      "step": 3250
    },
    {
      "epoch": 6.03,
      "grad_norm": 1.8520976305007935,
      "learning_rate": 0.0019842592592592594,
      "loss": 0.498,
      "step": 3260
    },
    {
      "epoch": 6.05,
      "grad_norm": 2.38435959815979,
      "learning_rate": 0.001975,
      "loss": 0.4984,
      "step": 3270
    },
    {
      "epoch": 6.07,
      "grad_norm": 2.9417970180511475,
      "learning_rate": 0.0019657407407407406,
      "loss": 0.489,
      "step": 3280
    },
    {
      "epoch": 6.09,
      "grad_norm": 2.5652129650115967,
      "learning_rate": 0.001956481481481482,
      "loss": 0.5028,
      "step": 3290
    },
    {
      "epoch": 6.11,
      "grad_norm": 3.151473045349121,
      "learning_rate": 0.0019472222222222221,
      "loss": 0.4651,
      "step": 3300
    },
    {
      "epoch": 6.12,
      "grad_norm": 2.743474006652832,
      "learning_rate": 0.001937962962962963,
      "loss": 0.4489,
      "step": 3310
    },
    {
      "epoch": 6.14,
      "grad_norm": 2.15732741355896,
      "learning_rate": 0.0019287037037037035,
      "loss": 0.4736,
      "step": 3320
    },
    {
      "epoch": 6.16,
      "grad_norm": 3.1959879398345947,
      "learning_rate": 0.0019194444444444445,
      "loss": 0.4998,
      "step": 3330
    },
    {
      "epoch": 6.18,
      "grad_norm": 3.215951919555664,
      "learning_rate": 0.0019101851851851853,
      "loss": 0.4991,
      "step": 3340
    },
    {
      "epoch": 6.2,
      "grad_norm": 2.889775276184082,
      "learning_rate": 0.001900925925925926,
      "loss": 0.4817,
      "step": 3350
    },
    {
      "epoch": 6.22,
      "grad_norm": 3.2150237560272217,
      "learning_rate": 0.0018916666666666667,
      "loss": 0.5287,
      "step": 3360
    },
    {
      "epoch": 6.23,
      "grad_norm": 1.6671727895736694,
      "learning_rate": 0.0018824074074074075,
      "loss": 0.4492,
      "step": 3370
    },
    {
      "epoch": 6.25,
      "grad_norm": 2.812654733657837,
      "learning_rate": 0.0018731481481481483,
      "loss": 0.5098,
      "step": 3380
    },
    {
      "epoch": 6.27,
      "grad_norm": 2.0343258380889893,
      "learning_rate": 0.0018638888888888889,
      "loss": 0.4513,
      "step": 3390
    },
    {
      "epoch": 6.29,
      "grad_norm": 2.830615997314453,
      "learning_rate": 0.0018546296296296297,
      "loss": 0.4621,
      "step": 3400
    },
    {
      "epoch": 6.31,
      "grad_norm": 1.4622855186462402,
      "learning_rate": 0.0018453703703703703,
      "loss": 0.4741,
      "step": 3410
    },
    {
      "epoch": 6.33,
      "grad_norm": 2.521087884902954,
      "learning_rate": 0.0018361111111111113,
      "loss": 0.4932,
      "step": 3420
    },
    {
      "epoch": 6.35,
      "grad_norm": 2.6355080604553223,
      "learning_rate": 0.0018268518518518519,
      "loss": 0.4676,
      "step": 3430
    },
    {
      "epoch": 6.36,
      "grad_norm": 6.4964189529418945,
      "learning_rate": 0.0018175925925925927,
      "loss": 0.4437,
      "step": 3440
    },
    {
      "epoch": 6.38,
      "grad_norm": 2.58648943901062,
      "learning_rate": 0.0018083333333333335,
      "loss": 0.4583,
      "step": 3450
    },
    {
      "epoch": 6.4,
      "grad_norm": 2.710174560546875,
      "learning_rate": 0.001799074074074074,
      "loss": 0.5327,
      "step": 3460
    },
    {
      "epoch": 6.42,
      "grad_norm": 4.029747486114502,
      "learning_rate": 0.001789814814814815,
      "loss": 0.5417,
      "step": 3470
    },
    {
      "epoch": 6.44,
      "grad_norm": 1.4080957174301147,
      "learning_rate": 0.0017805555555555556,
      "loss": 0.527,
      "step": 3480
    },
    {
      "epoch": 6.46,
      "grad_norm": 2.5939226150512695,
      "learning_rate": 0.0017712962962962964,
      "loss": 0.4533,
      "step": 3490
    },
    {
      "epoch": 6.48,
      "grad_norm": 1.7863126993179321,
      "learning_rate": 0.001762037037037037,
      "loss": 0.4796,
      "step": 3500
    },
    {
      "epoch": 6.49,
      "grad_norm": 2.5097310543060303,
      "learning_rate": 0.0017527777777777778,
      "loss": 0.5026,
      "step": 3510
    },
    {
      "epoch": 6.51,
      "grad_norm": 3.5115163326263428,
      "learning_rate": 0.0017435185185185184,
      "loss": 0.5033,
      "step": 3520
    },
    {
      "epoch": 6.53,
      "grad_norm": 3.9437549114227295,
      "learning_rate": 0.0017342592592592594,
      "loss": 0.5327,
      "step": 3530
    },
    {
      "epoch": 6.55,
      "grad_norm": 2.61135196685791,
      "learning_rate": 0.001725,
      "loss": 0.4928,
      "step": 3540
    },
    {
      "epoch": 6.57,
      "grad_norm": 2.595803737640381,
      "learning_rate": 0.0017157407407407408,
      "loss": 0.4475,
      "step": 3550
    },
    {
      "epoch": 6.59,
      "grad_norm": 2.401987075805664,
      "learning_rate": 0.0017064814814814816,
      "loss": 0.4093,
      "step": 3560
    },
    {
      "epoch": 6.6,
      "grad_norm": 2.670671224594116,
      "learning_rate": 0.0016972222222222221,
      "loss": 0.4736,
      "step": 3570
    },
    {
      "epoch": 6.62,
      "grad_norm": 2.6677041053771973,
      "learning_rate": 0.0016879629629629632,
      "loss": 0.3864,
      "step": 3580
    },
    {
      "epoch": 6.64,
      "grad_norm": 2.3009049892425537,
      "learning_rate": 0.0016787037037037037,
      "loss": 0.486,
      "step": 3590
    },
    {
      "epoch": 6.66,
      "grad_norm": 3.2118513584136963,
      "learning_rate": 0.0016694444444444445,
      "loss": 0.4634,
      "step": 3600
    },
    {
      "epoch": 6.68,
      "grad_norm": 2.185634136199951,
      "learning_rate": 0.0016601851851851851,
      "loss": 0.4865,
      "step": 3610
    },
    {
      "epoch": 6.7,
      "grad_norm": 4.046822547912598,
      "learning_rate": 0.001650925925925926,
      "loss": 0.4624,
      "step": 3620
    },
    {
      "epoch": 6.72,
      "grad_norm": 2.3376824855804443,
      "learning_rate": 0.0016416666666666665,
      "loss": 0.4652,
      "step": 3630
    },
    {
      "epoch": 6.73,
      "grad_norm": 3.318936347961426,
      "learning_rate": 0.0016324074074074075,
      "loss": 0.4595,
      "step": 3640
    },
    {
      "epoch": 6.75,
      "grad_norm": 2.0213863849639893,
      "learning_rate": 0.0016231481481481483,
      "loss": 0.4768,
      "step": 3650
    },
    {
      "epoch": 6.77,
      "grad_norm": 1.718332052230835,
      "learning_rate": 0.0016138888888888889,
      "loss": 0.5239,
      "step": 3660
    },
    {
      "epoch": 6.79,
      "grad_norm": 1.7439289093017578,
      "learning_rate": 0.0016046296296296297,
      "loss": 0.525,
      "step": 3670
    },
    {
      "epoch": 6.81,
      "grad_norm": 2.79032826423645,
      "learning_rate": 0.0015953703703703703,
      "loss": 0.4299,
      "step": 3680
    },
    {
      "epoch": 6.83,
      "grad_norm": 1.8835906982421875,
      "learning_rate": 0.0015861111111111113,
      "loss": 0.4833,
      "step": 3690
    },
    {
      "epoch": 6.85,
      "grad_norm": 2.179826021194458,
      "learning_rate": 0.0015768518518518519,
      "loss": 0.4499,
      "step": 3700
    },
    {
      "epoch": 6.86,
      "grad_norm": 2.882157325744629,
      "learning_rate": 0.0015675925925925926,
      "loss": 0.4296,
      "step": 3710
    },
    {
      "epoch": 6.88,
      "grad_norm": 2.04009747505188,
      "learning_rate": 0.0015583333333333332,
      "loss": 0.4246,
      "step": 3720
    },
    {
      "epoch": 6.9,
      "grad_norm": 1.883562684059143,
      "learning_rate": 0.0015490740740740742,
      "loss": 0.4877,
      "step": 3730
    },
    {
      "epoch": 6.92,
      "grad_norm": 2.195427179336548,
      "learning_rate": 0.0015398148148148148,
      "loss": 0.4985,
      "step": 3740
    },
    {
      "epoch": 6.94,
      "grad_norm": 1.643837332725525,
      "learning_rate": 0.0015305555555555556,
      "loss": 0.4668,
      "step": 3750
    },
    {
      "epoch": 6.96,
      "grad_norm": 1.886772871017456,
      "learning_rate": 0.0015212962962962964,
      "loss": 0.4143,
      "step": 3760
    },
    {
      "epoch": 6.98,
      "grad_norm": 1.8772404193878174,
      "learning_rate": 0.001512037037037037,
      "loss": 0.4765,
      "step": 3770
    },
    {
      "epoch": 6.99,
      "grad_norm": 2.053422689437866,
      "learning_rate": 0.001502777777777778,
      "loss": 0.4298,
      "step": 3780
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9662609767370205,
      "eval_f1": 0.9735569034773577,
      "eval_loss": 0.09894031286239624,
      "eval_precision": 0.9770102531109687,
      "eval_recall": 0.9722826041431603,
      "eval_runtime": 105.603,
      "eval_samples_per_second": 61.466,
      "eval_steps_per_second": 3.845,
      "step": 3783
    },
    {
      "epoch": 7.01,
      "grad_norm": 2.0501725673675537,
      "learning_rate": 0.0014935185185185186,
      "loss": 0.4187,
      "step": 3790
    },
    {
      "epoch": 7.03,
      "grad_norm": 1.9505852460861206,
      "learning_rate": 0.0014842592592592594,
      "loss": 0.5045,
      "step": 3800
    },
    {
      "epoch": 7.05,
      "grad_norm": 2.5469818115234375,
      "learning_rate": 0.001475,
      "loss": 0.4567,
      "step": 3810
    },
    {
      "epoch": 7.07,
      "grad_norm": 3.28265643119812,
      "learning_rate": 0.0014657407407407408,
      "loss": 0.4738,
      "step": 3820
    },
    {
      "epoch": 7.09,
      "grad_norm": 0.9891900420188904,
      "learning_rate": 0.0014564814814814813,
      "loss": 0.4332,
      "step": 3830
    },
    {
      "epoch": 7.1,
      "grad_norm": 2.7753162384033203,
      "learning_rate": 0.0014472222222222224,
      "loss": 0.4334,
      "step": 3840
    },
    {
      "epoch": 7.12,
      "grad_norm": 2.795895576477051,
      "learning_rate": 0.001437962962962963,
      "loss": 0.5147,
      "step": 3850
    },
    {
      "epoch": 7.14,
      "grad_norm": 1.353401780128479,
      "learning_rate": 0.0014287037037037037,
      "loss": 0.4358,
      "step": 3860
    },
    {
      "epoch": 7.16,
      "grad_norm": 2.1106724739074707,
      "learning_rate": 0.0014194444444444445,
      "loss": 0.3712,
      "step": 3870
    },
    {
      "epoch": 7.18,
      "grad_norm": 3.6988472938537598,
      "learning_rate": 0.0014101851851851851,
      "loss": 0.4664,
      "step": 3880
    },
    {
      "epoch": 7.2,
      "grad_norm": 2.6754424571990967,
      "learning_rate": 0.0014009259259259261,
      "loss": 0.4231,
      "step": 3890
    },
    {
      "epoch": 7.22,
      "grad_norm": 2.6841065883636475,
      "learning_rate": 0.0013916666666666667,
      "loss": 0.4378,
      "step": 3900
    },
    {
      "epoch": 7.23,
      "grad_norm": 2.5743584632873535,
      "learning_rate": 0.0013824074074074075,
      "loss": 0.3719,
      "step": 3910
    },
    {
      "epoch": 7.25,
      "grad_norm": 1.586349606513977,
      "learning_rate": 0.001373148148148148,
      "loss": 0.3748,
      "step": 3920
    },
    {
      "epoch": 7.27,
      "grad_norm": 1.5181403160095215,
      "learning_rate": 0.0013638888888888889,
      "loss": 0.4544,
      "step": 3930
    },
    {
      "epoch": 7.29,
      "grad_norm": 1.539527416229248,
      "learning_rate": 0.0013546296296296295,
      "loss": 0.3758,
      "step": 3940
    },
    {
      "epoch": 7.31,
      "grad_norm": 1.6864418983459473,
      "learning_rate": 0.0013453703703703705,
      "loss": 0.3795,
      "step": 3950
    },
    {
      "epoch": 7.33,
      "grad_norm": 2.111827850341797,
      "learning_rate": 0.0013361111111111113,
      "loss": 0.4379,
      "step": 3960
    },
    {
      "epoch": 7.35,
      "grad_norm": 1.9402803182601929,
      "learning_rate": 0.0013268518518518518,
      "loss": 0.446,
      "step": 3970
    },
    {
      "epoch": 7.36,
      "grad_norm": 1.413064956665039,
      "learning_rate": 0.0013175925925925926,
      "loss": 0.509,
      "step": 3980
    },
    {
      "epoch": 7.38,
      "grad_norm": 1.7911455631256104,
      "learning_rate": 0.0013083333333333332,
      "loss": 0.4938,
      "step": 3990
    },
    {
      "epoch": 7.4,
      "grad_norm": 2.256588935852051,
      "learning_rate": 0.0012990740740740742,
      "loss": 0.4844,
      "step": 4000
    },
    {
      "epoch": 7.42,
      "grad_norm": 1.993911623954773,
      "learning_rate": 0.0012898148148148148,
      "loss": 0.419,
      "step": 4010
    },
    {
      "epoch": 7.44,
      "grad_norm": 1.3340710401535034,
      "learning_rate": 0.0012805555555555556,
      "loss": 0.4464,
      "step": 4020
    },
    {
      "epoch": 7.46,
      "grad_norm": 2.902743339538574,
      "learning_rate": 0.0012712962962962962,
      "loss": 0.4538,
      "step": 4030
    },
    {
      "epoch": 7.47,
      "grad_norm": 2.30949068069458,
      "learning_rate": 0.001262037037037037,
      "loss": 0.4209,
      "step": 4040
    },
    {
      "epoch": 7.49,
      "grad_norm": 2.7385501861572266,
      "learning_rate": 0.0012527777777777778,
      "loss": 0.3801,
      "step": 4050
    },
    {
      "epoch": 7.51,
      "grad_norm": 1.1858464479446411,
      "learning_rate": 0.0012435185185185186,
      "loss": 0.3956,
      "step": 4060
    },
    {
      "epoch": 7.53,
      "grad_norm": 1.694409728050232,
      "learning_rate": 0.0012342592592592594,
      "loss": 0.3786,
      "step": 4070
    },
    {
      "epoch": 7.55,
      "grad_norm": 2.7329421043395996,
      "learning_rate": 0.001225,
      "loss": 0.4497,
      "step": 4080
    },
    {
      "epoch": 7.57,
      "grad_norm": 2.352189540863037,
      "learning_rate": 0.0012157407407407408,
      "loss": 0.3877,
      "step": 4090
    },
    {
      "epoch": 7.59,
      "grad_norm": 4.443812847137451,
      "learning_rate": 0.0012064814814814816,
      "loss": 0.367,
      "step": 4100
    },
    {
      "epoch": 7.6,
      "grad_norm": 2.3426809310913086,
      "learning_rate": 0.0011972222222222221,
      "loss": 0.4192,
      "step": 4110
    },
    {
      "epoch": 7.62,
      "grad_norm": 2.791630268096924,
      "learning_rate": 0.0011879629629629631,
      "loss": 0.407,
      "step": 4120
    },
    {
      "epoch": 7.64,
      "grad_norm": 2.74001145362854,
      "learning_rate": 0.0011787037037037037,
      "loss": 0.3679,
      "step": 4130
    },
    {
      "epoch": 7.66,
      "grad_norm": 3.0963590145111084,
      "learning_rate": 0.0011694444444444445,
      "loss": 0.3808,
      "step": 4140
    },
    {
      "epoch": 7.68,
      "grad_norm": 1.7934086322784424,
      "learning_rate": 0.0011601851851851853,
      "loss": 0.427,
      "step": 4150
    },
    {
      "epoch": 7.7,
      "grad_norm": 2.269930601119995,
      "learning_rate": 0.001150925925925926,
      "loss": 0.3847,
      "step": 4160
    },
    {
      "epoch": 7.72,
      "grad_norm": 1.5360560417175293,
      "learning_rate": 0.0011416666666666667,
      "loss": 0.4534,
      "step": 4170
    },
    {
      "epoch": 7.73,
      "grad_norm": 1.9300364255905151,
      "learning_rate": 0.0011324074074074075,
      "loss": 0.4014,
      "step": 4180
    },
    {
      "epoch": 7.75,
      "grad_norm": 1.613265037536621,
      "learning_rate": 0.001123148148148148,
      "loss": 0.4852,
      "step": 4190
    },
    {
      "epoch": 7.77,
      "grad_norm": 1.625320553779602,
      "learning_rate": 0.0011138888888888889,
      "loss": 0.4753,
      "step": 4200
    },
    {
      "epoch": 7.79,
      "grad_norm": 2.176623821258545,
      "learning_rate": 0.0011046296296296297,
      "loss": 0.4611,
      "step": 4210
    },
    {
      "epoch": 7.81,
      "grad_norm": 2.2461256980895996,
      "learning_rate": 0.0010953703703703705,
      "loss": 0.4326,
      "step": 4220
    },
    {
      "epoch": 7.83,
      "grad_norm": 2.6430141925811768,
      "learning_rate": 0.0010861111111111113,
      "loss": 0.3267,
      "step": 4230
    },
    {
      "epoch": 7.84,
      "grad_norm": 2.5161099433898926,
      "learning_rate": 0.0010768518518518518,
      "loss": 0.4114,
      "step": 4240
    },
    {
      "epoch": 7.86,
      "grad_norm": 2.0569732189178467,
      "learning_rate": 0.0010675925925925926,
      "loss": 0.4323,
      "step": 4250
    },
    {
      "epoch": 7.88,
      "grad_norm": 1.884279489517212,
      "learning_rate": 0.0010583333333333334,
      "loss": 0.4652,
      "step": 4260
    },
    {
      "epoch": 7.9,
      "grad_norm": 1.4818202257156372,
      "learning_rate": 0.001049074074074074,
      "loss": 0.3994,
      "step": 4270
    },
    {
      "epoch": 7.92,
      "grad_norm": 2.8348388671875,
      "learning_rate": 0.0010398148148148148,
      "loss": 0.3829,
      "step": 4280
    },
    {
      "epoch": 7.94,
      "grad_norm": 1.3215655088424683,
      "learning_rate": 0.0010305555555555556,
      "loss": 0.4266,
      "step": 4290
    },
    {
      "epoch": 7.96,
      "grad_norm": 2.0123190879821777,
      "learning_rate": 0.0010212962962962962,
      "loss": 0.3671,
      "step": 4300
    },
    {
      "epoch": 7.97,
      "grad_norm": 2.014122486114502,
      "learning_rate": 0.001012037037037037,
      "loss": 0.4353,
      "step": 4310
    },
    {
      "epoch": 7.99,
      "grad_norm": 5.686545372009277,
      "learning_rate": 0.0010027777777777778,
      "loss": 0.3932,
      "step": 4320
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9884455399784317,
      "eval_f1": 0.989380892500223,
      "eval_loss": 0.03349991887807846,
      "eval_precision": 0.9903161849271732,
      "eval_recall": 0.9887487397554633,
      "eval_runtime": 105.0862,
      "eval_samples_per_second": 61.768,
      "eval_steps_per_second": 3.863,
      "step": 4324
    },
    {
      "epoch": 8.01,
      "grad_norm": 1.598591685295105,
      "learning_rate": 0.0009935185185185186,
      "loss": 0.4489,
      "step": 4330
    },
    {
      "epoch": 8.03,
      "grad_norm": 1.8468151092529297,
      "learning_rate": 0.0009842592592592594,
      "loss": 0.3364,
      "step": 4340
    },
    {
      "epoch": 8.05,
      "grad_norm": 2.6677463054656982,
      "learning_rate": 0.0009750000000000001,
      "loss": 0.4332,
      "step": 4350
    },
    {
      "epoch": 8.07,
      "grad_norm": 2.6927859783172607,
      "learning_rate": 0.0009657407407407408,
      "loss": 0.4052,
      "step": 4360
    },
    {
      "epoch": 8.09,
      "grad_norm": 1.6345399618148804,
      "learning_rate": 0.0009564814814814815,
      "loss": 0.4157,
      "step": 4370
    },
    {
      "epoch": 8.1,
      "grad_norm": 2.7183444499969482,
      "learning_rate": 0.0009472222222222222,
      "loss": 0.4309,
      "step": 4380
    },
    {
      "epoch": 8.12,
      "grad_norm": 1.2181589603424072,
      "learning_rate": 0.0009379629629629629,
      "loss": 0.34,
      "step": 4390
    },
    {
      "epoch": 8.14,
      "grad_norm": 1.3929060697555542,
      "learning_rate": 0.0009287037037037037,
      "loss": 0.3394,
      "step": 4400
    },
    {
      "epoch": 8.16,
      "grad_norm": 2.487053632736206,
      "learning_rate": 0.0009194444444444444,
      "loss": 0.3793,
      "step": 4410
    },
    {
      "epoch": 8.18,
      "grad_norm": 4.430356979370117,
      "learning_rate": 0.0009101851851851851,
      "loss": 0.3657,
      "step": 4420
    },
    {
      "epoch": 8.2,
      "grad_norm": 3.2201452255249023,
      "learning_rate": 0.000900925925925926,
      "loss": 0.4235,
      "step": 4430
    },
    {
      "epoch": 8.21,
      "grad_norm": 1.5028598308563232,
      "learning_rate": 0.0008916666666666667,
      "loss": 0.4478,
      "step": 4440
    },
    {
      "epoch": 8.23,
      "grad_norm": 1.374112844467163,
      "learning_rate": 0.0008824074074074075,
      "loss": 0.4129,
      "step": 4450
    },
    {
      "epoch": 8.25,
      "grad_norm": 1.8899506330490112,
      "learning_rate": 0.0008731481481481482,
      "loss": 0.3859,
      "step": 4460
    },
    {
      "epoch": 8.27,
      "grad_norm": 2.019902229309082,
      "learning_rate": 0.0008638888888888889,
      "loss": 0.4101,
      "step": 4470
    },
    {
      "epoch": 8.29,
      "grad_norm": 1.705667495727539,
      "learning_rate": 0.0008546296296296297,
      "loss": 0.3493,
      "step": 4480
    },
    {
      "epoch": 8.31,
      "grad_norm": 2.031428098678589,
      "learning_rate": 0.0008453703703703704,
      "loss": 0.4074,
      "step": 4490
    },
    {
      "epoch": 8.33,
      "grad_norm": 2.7189221382141113,
      "learning_rate": 0.0008361111111111111,
      "loss": 0.3786,
      "step": 4500
    },
    {
      "epoch": 8.34,
      "grad_norm": 2.454854726791382,
      "learning_rate": 0.0008268518518518518,
      "loss": 0.4436,
      "step": 4510
    },
    {
      "epoch": 8.36,
      "grad_norm": 1.298943281173706,
      "learning_rate": 0.0008175925925925925,
      "loss": 0.4313,
      "step": 4520
    },
    {
      "epoch": 8.38,
      "grad_norm": 2.463287591934204,
      "learning_rate": 0.0008083333333333333,
      "loss": 0.4414,
      "step": 4530
    },
    {
      "epoch": 8.4,
      "grad_norm": 2.80354642868042,
      "learning_rate": 0.0007990740740740741,
      "loss": 0.4188,
      "step": 4540
    },
    {
      "epoch": 8.42,
      "grad_norm": 3.166337251663208,
      "learning_rate": 0.0007898148148148149,
      "loss": 0.4059,
      "step": 4550
    },
    {
      "epoch": 8.44,
      "grad_norm": 1.655117392539978,
      "learning_rate": 0.0007805555555555556,
      "loss": 0.3877,
      "step": 4560
    },
    {
      "epoch": 8.46,
      "grad_norm": 2.249906539916992,
      "learning_rate": 0.0007712962962962963,
      "loss": 0.355,
      "step": 4570
    },
    {
      "epoch": 8.47,
      "grad_norm": 1.6171294450759888,
      "learning_rate": 0.0007620370370370371,
      "loss": 0.3233,
      "step": 4580
    },
    {
      "epoch": 8.49,
      "grad_norm": 2.709031581878662,
      "learning_rate": 0.0007527777777777778,
      "loss": 0.3407,
      "step": 4590
    },
    {
      "epoch": 8.51,
      "grad_norm": 2.6156270503997803,
      "learning_rate": 0.0007435185185185185,
      "loss": 0.3779,
      "step": 4600
    },
    {
      "epoch": 8.53,
      "grad_norm": 1.640161156654358,
      "learning_rate": 0.0007342592592592593,
      "loss": 0.3557,
      "step": 4610
    },
    {
      "epoch": 8.55,
      "grad_norm": 1.8425498008728027,
      "learning_rate": 0.000725,
      "loss": 0.3987,
      "step": 4620
    },
    {
      "epoch": 8.57,
      "grad_norm": 2.946497917175293,
      "learning_rate": 0.0007157407407407406,
      "loss": 0.4044,
      "step": 4630
    },
    {
      "epoch": 8.58,
      "grad_norm": 2.3729591369628906,
      "learning_rate": 0.0007064814814814815,
      "loss": 0.358,
      "step": 4640
    },
    {
      "epoch": 8.6,
      "grad_norm": 1.634698510169983,
      "learning_rate": 0.0006972222222222222,
      "loss": 0.3837,
      "step": 4650
    },
    {
      "epoch": 8.62,
      "grad_norm": 2.881890296936035,
      "learning_rate": 0.000687962962962963,
      "loss": 0.3643,
      "step": 4660
    },
    {
      "epoch": 8.64,
      "grad_norm": 1.3335785865783691,
      "learning_rate": 0.0006787037037037037,
      "loss": 0.3887,
      "step": 4670
    },
    {
      "epoch": 8.66,
      "grad_norm": 2.1087048053741455,
      "learning_rate": 0.0006694444444444445,
      "loss": 0.3354,
      "step": 4680
    },
    {
      "epoch": 8.68,
      "grad_norm": 3.421175956726074,
      "learning_rate": 0.0006601851851851852,
      "loss": 0.4143,
      "step": 4690
    },
    {
      "epoch": 8.7,
      "grad_norm": 1.5602165460586548,
      "learning_rate": 0.0006509259259259259,
      "loss": 0.3928,
      "step": 4700
    },
    {
      "epoch": 8.71,
      "grad_norm": 3.690290689468384,
      "learning_rate": 0.0006416666666666667,
      "loss": 0.3157,
      "step": 4710
    },
    {
      "epoch": 8.73,
      "grad_norm": 2.0590598583221436,
      "learning_rate": 0.0006324074074074074,
      "loss": 0.3986,
      "step": 4720
    },
    {
      "epoch": 8.75,
      "grad_norm": 1.1233676671981812,
      "learning_rate": 0.0006231481481481482,
      "loss": 0.3456,
      "step": 4730
    },
    {
      "epoch": 8.77,
      "grad_norm": 1.8680920600891113,
      "learning_rate": 0.0006138888888888889,
      "loss": 0.4048,
      "step": 4740
    },
    {
      "epoch": 8.79,
      "grad_norm": 1.1571403741836548,
      "learning_rate": 0.0006046296296296297,
      "loss": 0.3535,
      "step": 4750
    },
    {
      "epoch": 8.81,
      "grad_norm": 1.505861520767212,
      "learning_rate": 0.0005953703703703703,
      "loss": 0.3203,
      "step": 4760
    },
    {
      "epoch": 8.83,
      "grad_norm": 1.7797300815582275,
      "learning_rate": 0.0005861111111111111,
      "loss": 0.3308,
      "step": 4770
    },
    {
      "epoch": 8.84,
      "grad_norm": 1.0974406003952026,
      "learning_rate": 0.0005768518518518518,
      "loss": 0.3759,
      "step": 4780
    },
    {
      "epoch": 8.86,
      "grad_norm": 1.5202006101608276,
      "learning_rate": 0.0005675925925925926,
      "loss": 0.3725,
      "step": 4790
    },
    {
      "epoch": 8.88,
      "grad_norm": 1.2904443740844727,
      "learning_rate": 0.0005583333333333333,
      "loss": 0.3535,
      "step": 4800
    },
    {
      "epoch": 8.9,
      "grad_norm": 1.4673963785171509,
      "learning_rate": 0.0005490740740740741,
      "loss": 0.3937,
      "step": 4810
    },
    {
      "epoch": 8.92,
      "grad_norm": 1.0130095481872559,
      "learning_rate": 0.0005398148148148149,
      "loss": 0.3306,
      "step": 4820
    },
    {
      "epoch": 8.94,
      "grad_norm": 1.4989911317825317,
      "learning_rate": 0.0005305555555555556,
      "loss": 0.4111,
      "step": 4830
    },
    {
      "epoch": 8.95,
      "grad_norm": 1.36995267868042,
      "learning_rate": 0.0005212962962962963,
      "loss": 0.3791,
      "step": 4840
    },
    {
      "epoch": 8.97,
      "grad_norm": 2.134408950805664,
      "learning_rate": 0.0005120370370370371,
      "loss": 0.3691,
      "step": 4850
    },
    {
      "epoch": 8.99,
      "grad_norm": 2.304680347442627,
      "learning_rate": 0.0005027777777777778,
      "loss": 0.3409,
      "step": 4860
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9878293021106147,
      "eval_f1": 0.9886590982069928,
      "eval_loss": 0.03709910064935684,
      "eval_precision": 0.989964784159849,
      "eval_recall": 0.9876911349748236,
      "eval_runtime": 105.3983,
      "eval_samples_per_second": 61.585,
      "eval_steps_per_second": 3.852,
      "step": 4864
    },
    {
      "epoch": 9.01,
      "grad_norm": 1.6748008728027344,
      "learning_rate": 0.0004935185185185186,
      "loss": 0.3296,
      "step": 4870
    },
    {
      "epoch": 9.03,
      "grad_norm": 0.8384755849838257,
      "learning_rate": 0.0004842592592592593,
      "loss": 0.3316,
      "step": 4880
    },
    {
      "epoch": 9.05,
      "grad_norm": 2.137038230895996,
      "learning_rate": 0.000475,
      "loss": 0.3698,
      "step": 4890
    },
    {
      "epoch": 9.07,
      "grad_norm": 1.4509310722351074,
      "learning_rate": 0.00046574074074074074,
      "loss": 0.3456,
      "step": 4900
    },
    {
      "epoch": 9.08,
      "grad_norm": 2.2383487224578857,
      "learning_rate": 0.0004564814814814815,
      "loss": 0.3705,
      "step": 4910
    },
    {
      "epoch": 9.1,
      "grad_norm": 1.851224660873413,
      "learning_rate": 0.00044722222222222217,
      "loss": 0.3645,
      "step": 4920
    },
    {
      "epoch": 9.12,
      "grad_norm": 1.9950298070907593,
      "learning_rate": 0.00043796296296296297,
      "loss": 0.3502,
      "step": 4930
    },
    {
      "epoch": 9.14,
      "grad_norm": 2.0750277042388916,
      "learning_rate": 0.0004287037037037037,
      "loss": 0.2884,
      "step": 4940
    },
    {
      "epoch": 9.16,
      "grad_norm": 2.44145131111145,
      "learning_rate": 0.00041944444444444445,
      "loss": 0.3615,
      "step": 4950
    },
    {
      "epoch": 9.18,
      "grad_norm": 1.4347975254058838,
      "learning_rate": 0.0004101851851851852,
      "loss": 0.3275,
      "step": 4960
    },
    {
      "epoch": 9.2,
      "grad_norm": 1.7355468273162842,
      "learning_rate": 0.0004009259259259259,
      "loss": 0.3158,
      "step": 4970
    },
    {
      "epoch": 9.21,
      "grad_norm": 1.5040051937103271,
      "learning_rate": 0.0003916666666666667,
      "loss": 0.3265,
      "step": 4980
    },
    {
      "epoch": 9.23,
      "grad_norm": 2.3825201988220215,
      "learning_rate": 0.0003824074074074074,
      "loss": 0.3635,
      "step": 4990
    },
    {
      "epoch": 9.25,
      "grad_norm": 1.7622706890106201,
      "learning_rate": 0.00037314814814814817,
      "loss": 0.3064,
      "step": 5000
    },
    {
      "epoch": 9.27,
      "grad_norm": 1.3905962705612183,
      "learning_rate": 0.00036388888888888886,
      "loss": 0.3192,
      "step": 5010
    },
    {
      "epoch": 9.29,
      "grad_norm": 1.1275396347045898,
      "learning_rate": 0.0003546296296296296,
      "loss": 0.3793,
      "step": 5020
    },
    {
      "epoch": 9.31,
      "grad_norm": 1.043797254562378,
      "learning_rate": 0.0003453703703703704,
      "loss": 0.3532,
      "step": 5030
    },
    {
      "epoch": 9.32,
      "grad_norm": 2.222782611846924,
      "learning_rate": 0.00033611111111111114,
      "loss": 0.3444,
      "step": 5040
    },
    {
      "epoch": 9.34,
      "grad_norm": 1.1857503652572632,
      "learning_rate": 0.0003268518518518519,
      "loss": 0.2676,
      "step": 5050
    },
    {
      "epoch": 9.36,
      "grad_norm": 1.542576789855957,
      "learning_rate": 0.00031759259259259257,
      "loss": 0.3942,
      "step": 5060
    },
    {
      "epoch": 9.38,
      "grad_norm": 1.9008642435073853,
      "learning_rate": 0.00030833333333333337,
      "loss": 0.3854,
      "step": 5070
    },
    {
      "epoch": 9.4,
      "grad_norm": 1.819470763206482,
      "learning_rate": 0.00029907407407407405,
      "loss": 0.3426,
      "step": 5080
    },
    {
      "epoch": 9.42,
      "grad_norm": 1.4119468927383423,
      "learning_rate": 0.00028981481481481485,
      "loss": 0.3093,
      "step": 5090
    },
    {
      "epoch": 9.44,
      "grad_norm": 2.1372573375701904,
      "learning_rate": 0.00028055555555555554,
      "loss": 0.3178,
      "step": 5100
    },
    {
      "epoch": 9.45,
      "grad_norm": 1.6984546184539795,
      "learning_rate": 0.0002712962962962963,
      "loss": 0.3057,
      "step": 5110
    },
    {
      "epoch": 9.47,
      "grad_norm": 1.6501168012619019,
      "learning_rate": 0.0002620370370370371,
      "loss": 0.3501,
      "step": 5120
    },
    {
      "epoch": 9.49,
      "grad_norm": 1.3018875122070312,
      "learning_rate": 0.00025277777777777777,
      "loss": 0.3547,
      "step": 5130
    },
    {
      "epoch": 9.51,
      "grad_norm": 1.2195825576782227,
      "learning_rate": 0.0002435185185185185,
      "loss": 0.3597,
      "step": 5140
    },
    {
      "epoch": 9.53,
      "grad_norm": 1.5462696552276611,
      "learning_rate": 0.00023425925925925928,
      "loss": 0.378,
      "step": 5150
    },
    {
      "epoch": 9.55,
      "grad_norm": 1.2776215076446533,
      "learning_rate": 0.000225,
      "loss": 0.2943,
      "step": 5160
    },
    {
      "epoch": 9.57,
      "grad_norm": 1.9827752113342285,
      "learning_rate": 0.00021574074074074076,
      "loss": 0.3208,
      "step": 5170
    },
    {
      "epoch": 9.58,
      "grad_norm": 0.7241113185882568,
      "learning_rate": 0.00020648148148148148,
      "loss": 0.3298,
      "step": 5180
    },
    {
      "epoch": 9.6,
      "grad_norm": 1.118054986000061,
      "learning_rate": 0.00019722222222222222,
      "loss": 0.3145,
      "step": 5190
    },
    {
      "epoch": 9.62,
      "grad_norm": 1.2864770889282227,
      "learning_rate": 0.00018796296296296296,
      "loss": 0.3694,
      "step": 5200
    },
    {
      "epoch": 9.64,
      "grad_norm": 2.4528350830078125,
      "learning_rate": 0.0001787037037037037,
      "loss": 0.36,
      "step": 5210
    },
    {
      "epoch": 9.66,
      "grad_norm": 1.4478658437728882,
      "learning_rate": 0.00016944444444444448,
      "loss": 0.3421,
      "step": 5220
    },
    {
      "epoch": 9.68,
      "grad_norm": 1.7859172821044922,
      "learning_rate": 0.0001601851851851852,
      "loss": 0.3438,
      "step": 5230
    },
    {
      "epoch": 9.69,
      "grad_norm": 1.2724491357803345,
      "learning_rate": 0.00015092592592592593,
      "loss": 0.3298,
      "step": 5240
    },
    {
      "epoch": 9.71,
      "grad_norm": 1.7209391593933105,
      "learning_rate": 0.00014166666666666665,
      "loss": 0.3227,
      "step": 5250
    },
    {
      "epoch": 9.73,
      "grad_norm": 1.1616936922073364,
      "learning_rate": 0.00013240740740740742,
      "loss": 0.2874,
      "step": 5260
    },
    {
      "epoch": 9.75,
      "grad_norm": 1.5946358442306519,
      "learning_rate": 0.00012314814814814816,
      "loss": 0.3292,
      "step": 5270
    },
    {
      "epoch": 9.77,
      "grad_norm": 1.7766077518463135,
      "learning_rate": 0.00011388888888888889,
      "loss": 0.3105,
      "step": 5280
    },
    {
      "epoch": 9.79,
      "grad_norm": 1.2184169292449951,
      "learning_rate": 0.00010462962962962962,
      "loss": 0.3077,
      "step": 5290
    },
    {
      "epoch": 9.81,
      "grad_norm": 1.8188039064407349,
      "learning_rate": 9.537037037037036e-05,
      "loss": 0.3829,
      "step": 5300
    },
    {
      "epoch": 9.82,
      "grad_norm": 1.7841200828552246,
      "learning_rate": 8.611111111111112e-05,
      "loss": 0.4086,
      "step": 5310
    },
    {
      "epoch": 9.84,
      "grad_norm": 2.1823301315307617,
      "learning_rate": 7.685185185185186e-05,
      "loss": 0.3443,
      "step": 5320
    },
    {
      "epoch": 9.86,
      "grad_norm": 0.7260684370994568,
      "learning_rate": 6.759259259259259e-05,
      "loss": 0.2493,
      "step": 5330
    },
    {
      "epoch": 9.88,
      "grad_norm": 1.395943522453308,
      "learning_rate": 5.833333333333334e-05,
      "loss": 0.3521,
      "step": 5340
    },
    {
      "epoch": 9.9,
      "grad_norm": 1.731939435005188,
      "learning_rate": 4.9074074074074075e-05,
      "loss": 0.3337,
      "step": 5350
    },
    {
      "epoch": 9.92,
      "grad_norm": 1.1158477067947388,
      "learning_rate": 3.981481481481482e-05,
      "loss": 0.2676,
      "step": 5360
    },
    {
      "epoch": 9.94,
      "grad_norm": 1.2401944398880005,
      "learning_rate": 3.055555555555556e-05,
      "loss": 0.3386,
      "step": 5370
    },
    {
      "epoch": 9.95,
      "grad_norm": 1.0820297002792358,
      "learning_rate": 2.12962962962963e-05,
      "loss": 0.2561,
      "step": 5380
    },
    {
      "epoch": 9.97,
      "grad_norm": 1.5358567237854004,
      "learning_rate": 1.2037037037037039e-05,
      "loss": 0.2979,
      "step": 5390
    },
    {
      "epoch": 9.99,
      "grad_norm": 1.8409550189971924,
      "learning_rate": 2.777777777777778e-06,
      "loss": 0.3111,
      "step": 5400
    },
    {
      "epoch": 9.99,
      "eval_accuracy": 0.9845940533045756,
      "eval_f1": 0.9874036641651607,
      "eval_loss": 0.04327414557337761,
      "eval_precision": 0.9887972691294948,
      "eval_recall": 0.9864475392019028,
      "eval_runtime": 105.3994,
      "eval_samples_per_second": 61.585,
      "eval_steps_per_second": 3.852,
      "step": 5400
    },
    {
      "epoch": 9.99,
      "step": 5400,
      "total_flos": 6.121843696765878e+19,
      "train_loss": 0.5330523451169332,
      "train_runtime": 12454.7917,
      "train_samples_per_second": 27.765,
      "train_steps_per_second": 0.434
    }
  ],
  "logging_steps": 10,
  "max_steps": 5400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 6.121843696765878e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}