{
  "best_global_step": 1100,
  "best_metric": 2.467733144760132,
  "best_model_checkpoint": "./swin-ena24/checkpoint-1100",
  "epoch": 7.0,
  "eval_steps": 100,
  "global_step": 2779,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02518891687657431,
      "grad_norm": 254123.9375,
      "learning_rate": 0.00019935228499460238,
      "loss": 2.9691,
      "step": 10
    },
    {
      "epoch": 0.05037783375314862,
      "grad_norm": 426010.46875,
      "learning_rate": 0.00019863260165527168,
      "loss": 2.8919,
      "step": 20
    },
    {
      "epoch": 0.07556675062972293,
      "grad_norm": 438250.125,
      "learning_rate": 0.000197912918315941,
      "loss": 2.7195,
      "step": 30
    },
    {
      "epoch": 0.10075566750629723,
      "grad_norm": 643403.375,
      "learning_rate": 0.00019719323497661032,
      "loss": 2.567,
      "step": 40
    },
    {
      "epoch": 0.12594458438287154,
      "grad_norm": 416875.5,
      "learning_rate": 0.0001964735516372796,
      "loss": 2.6196,
      "step": 50
    },
    {
      "epoch": 0.15113350125944586,
      "grad_norm": 412576.78125,
      "learning_rate": 0.0001957538682979489,
      "loss": 2.4359,
      "step": 60
    },
    {
      "epoch": 0.17632241813602015,
      "grad_norm": 371569.9375,
      "learning_rate": 0.00019503418495861824,
      "loss": 2.4815,
      "step": 70
    },
    {
      "epoch": 0.20151133501259447,
      "grad_norm": 297408.5625,
      "learning_rate": 0.0001943145016192875,
      "loss": 2.508,
      "step": 80
    },
    {
      "epoch": 0.22670025188916876,
      "grad_norm": 405330.71875,
      "learning_rate": 0.00019359481827995682,
      "loss": 2.385,
      "step": 90
    },
    {
      "epoch": 0.2518891687657431,
      "grad_norm": 578770.75,
      "learning_rate": 0.00019287513494062612,
      "loss": 1.9888,
      "step": 100
    },
    {
      "epoch": 0.2518891687657431,
      "eval_accuracy": 0.1630859375,
      "eval_f1_macro": 0.08928735942932195,
      "eval_loss": 3.412175178527832,
      "eval_runtime": 11.0442,
      "eval_samples_per_second": 92.718,
      "eval_steps_per_second": 5.795,
      "step": 100
    },
    {
      "epoch": 0.2770780856423174,
      "grad_norm": 510874.125,
      "learning_rate": 0.00019215545160129545,
      "loss": 2.188,
      "step": 110
    },
    {
      "epoch": 0.3022670025188917,
      "grad_norm": 513859.9375,
      "learning_rate": 0.00019143576826196473,
      "loss": 2.1743,
      "step": 120
    },
    {
      "epoch": 0.327455919395466,
      "grad_norm": 527887.5,
      "learning_rate": 0.00019071608492263404,
      "loss": 2.2732,
      "step": 130
    },
    {
      "epoch": 0.3526448362720403,
      "grad_norm": 460399.375,
      "learning_rate": 0.00018999640158330337,
      "loss": 2.1564,
      "step": 140
    },
    {
      "epoch": 0.3778337531486146,
      "grad_norm": 569056.8125,
      "learning_rate": 0.00018927671824397267,
      "loss": 1.8225,
      "step": 150
    },
    {
      "epoch": 0.40302267002518893,
      "grad_norm": 909920.6875,
      "learning_rate": 0.00018855703490464195,
      "loss": 1.771,
      "step": 160
    },
    {
      "epoch": 0.4282115869017632,
      "grad_norm": 494884.5,
      "learning_rate": 0.00018783735156531128,
      "loss": 1.8349,
      "step": 170
    },
    {
      "epoch": 0.4534005037783375,
      "grad_norm": 550597.375,
      "learning_rate": 0.0001871176682259806,
      "loss": 1.5891,
      "step": 180
    },
    {
      "epoch": 0.47858942065491183,
      "grad_norm": 462934.75,
      "learning_rate": 0.00018639798488664987,
      "loss": 1.8926,
      "step": 190
    },
    {
      "epoch": 0.5037783375314862,
      "grad_norm": 724803.5,
      "learning_rate": 0.00018567830154731917,
      "loss": 1.6111,
      "step": 200
    },
    {
      "epoch": 0.5037783375314862,
      "eval_accuracy": 0.2578125,
      "eval_f1_macro": 0.15353256242114557,
      "eval_loss": 2.9077086448669434,
      "eval_runtime": 11.368,
      "eval_samples_per_second": 90.078,
      "eval_steps_per_second": 5.63,
      "step": 200
    },
    {
      "epoch": 0.5289672544080605,
      "grad_norm": 559555.8125,
      "learning_rate": 0.0001849586182079885,
      "loss": 1.6636,
      "step": 210
    },
    {
      "epoch": 0.5541561712846348,
      "grad_norm": 1021320.3125,
      "learning_rate": 0.0001842389348686578,
      "loss": 1.462,
      "step": 220
    },
    {
      "epoch": 0.5793450881612091,
      "grad_norm": 692250.375,
      "learning_rate": 0.0001835192515293271,
      "loss": 1.5007,
      "step": 230
    },
    {
      "epoch": 0.6045340050377834,
      "grad_norm": 774498.1875,
      "learning_rate": 0.00018279956818999642,
      "loss": 1.5122,
      "step": 240
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 497269.53125,
      "learning_rate": 0.00018207988485066572,
      "loss": 1.3379,
      "step": 250
    },
    {
      "epoch": 0.654911838790932,
      "grad_norm": 786711.4375,
      "learning_rate": 0.00018136020151133503,
      "loss": 1.5591,
      "step": 260
    },
    {
      "epoch": 0.6801007556675063,
      "grad_norm": 561642.375,
      "learning_rate": 0.00018064051817200433,
      "loss": 1.3499,
      "step": 270
    },
    {
      "epoch": 0.7052896725440806,
      "grad_norm": 372517.21875,
      "learning_rate": 0.00017992083483267364,
      "loss": 1.3296,
      "step": 280
    },
    {
      "epoch": 0.7304785894206549,
      "grad_norm": 783448.4375,
      "learning_rate": 0.00017920115149334294,
      "loss": 1.2106,
      "step": 290
    },
    {
      "epoch": 0.7556675062972292,
      "grad_norm": 748525.0,
      "learning_rate": 0.00017848146815401222,
      "loss": 1.1276,
      "step": 300
    },
    {
      "epoch": 0.7556675062972292,
      "eval_accuracy": 0.357421875,
      "eval_f1_macro": 0.28272669387224364,
      "eval_loss": 2.6503801345825195,
      "eval_runtime": 11.5522,
      "eval_samples_per_second": 88.641,
      "eval_steps_per_second": 5.54,
      "step": 300
    },
    {
      "epoch": 0.7808564231738035,
      "grad_norm": 481374.1875,
      "learning_rate": 0.00017776178481468155,
      "loss": 1.1727,
      "step": 310
    },
    {
      "epoch": 0.8060453400503779,
      "grad_norm": 1038731.1875,
      "learning_rate": 0.00017704210147535086,
      "loss": 1.1783,
      "step": 320
    },
    {
      "epoch": 0.8312342569269522,
      "grad_norm": 369614.8125,
      "learning_rate": 0.00017632241813602016,
      "loss": 1.4703,
      "step": 330
    },
    {
      "epoch": 0.8564231738035264,
      "grad_norm": 411708.9375,
      "learning_rate": 0.00017560273479668947,
      "loss": 1.0237,
      "step": 340
    },
    {
      "epoch": 0.8816120906801007,
      "grad_norm": 777554.375,
      "learning_rate": 0.00017488305145735877,
      "loss": 1.3113,
      "step": 350
    },
    {
      "epoch": 0.906801007556675,
      "grad_norm": 589113.125,
      "learning_rate": 0.00017416336811802808,
      "loss": 1.4282,
      "step": 360
    },
    {
      "epoch": 0.9319899244332494,
      "grad_norm": 537991.9375,
      "learning_rate": 0.00017344368477869738,
      "loss": 1.0802,
      "step": 370
    },
    {
      "epoch": 0.9571788413098237,
      "grad_norm": 745339.875,
      "learning_rate": 0.0001727240014393667,
      "loss": 1.1078,
      "step": 380
    },
    {
      "epoch": 0.982367758186398,
      "grad_norm": 515597.875,
      "learning_rate": 0.000172004318100036,
      "loss": 1.3261,
      "step": 390
    },
    {
      "epoch": 1.0075566750629723,
      "grad_norm": 438800.96875,
      "learning_rate": 0.0001712846347607053,
      "loss": 1.0234,
      "step": 400
    },
    {
      "epoch": 1.0075566750629723,
      "eval_accuracy": 0.390625,
      "eval_f1_macro": 0.31557618159267287,
      "eval_loss": 2.572810173034668,
      "eval_runtime": 11.7819,
      "eval_samples_per_second": 86.913,
      "eval_steps_per_second": 5.432,
      "step": 400
    },
    {
      "epoch": 1.0327455919395465,
      "grad_norm": 360076.0,
      "learning_rate": 0.0001705649514213746,
      "loss": 1.0415,
      "step": 410
    },
    {
      "epoch": 1.057934508816121,
      "grad_norm": 468955.34375,
      "learning_rate": 0.0001698452680820439,
      "loss": 0.8579,
      "step": 420
    },
    {
      "epoch": 1.0831234256926952,
      "grad_norm": 765571.4375,
      "learning_rate": 0.0001691255847427132,
      "loss": 0.9408,
      "step": 430
    },
    {
      "epoch": 1.1083123425692696,
      "grad_norm": 682154.8125,
      "learning_rate": 0.00016840590140338252,
      "loss": 1.0962,
      "step": 440
    },
    {
      "epoch": 1.1335012594458438,
      "grad_norm": 861960.5,
      "learning_rate": 0.00016768621806405182,
      "loss": 0.8124,
      "step": 450
    },
    {
      "epoch": 1.1586901763224182,
      "grad_norm": 583161.8125,
      "learning_rate": 0.00016696653472472113,
      "loss": 1.1014,
      "step": 460
    },
    {
      "epoch": 1.1838790931989924,
      "grad_norm": 390205.40625,
      "learning_rate": 0.00016624685138539046,
      "loss": 0.8475,
      "step": 470
    },
    {
      "epoch": 1.2090680100755669,
      "grad_norm": 293254.3125,
      "learning_rate": 0.00016552716804605974,
      "loss": 1.0482,
      "step": 480
    },
    {
      "epoch": 1.234256926952141,
      "grad_norm": 571449.25,
      "learning_rate": 0.00016480748470672904,
      "loss": 0.8776,
      "step": 490
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 362773.25,
      "learning_rate": 0.00016408780136739835,
      "loss": 0.8909,
      "step": 500
    },
    {
      "epoch": 1.2594458438287153,
      "eval_accuracy": 0.421875,
      "eval_f1_macro": 0.3388453334601594,
      "eval_loss": 2.500704765319824,
      "eval_runtime": 11.4557,
      "eval_samples_per_second": 89.388,
      "eval_steps_per_second": 5.587,
      "step": 500
    },
    {
      "epoch": 1.2846347607052897,
      "grad_norm": 351467.09375,
      "learning_rate": 0.00016336811802806765,
      "loss": 0.8244,
      "step": 510
    },
    {
      "epoch": 1.309823677581864,
      "grad_norm": 894287.0625,
      "learning_rate": 0.00016264843468873696,
      "loss": 1.048,
      "step": 520
    },
    {
      "epoch": 1.3350125944584383,
      "grad_norm": 474119.75,
      "learning_rate": 0.00016192875134940626,
      "loss": 0.9196,
      "step": 530
    },
    {
      "epoch": 1.3602015113350125,
      "grad_norm": 1069011.125,
      "learning_rate": 0.0001612090680100756,
      "loss": 0.804,
      "step": 540
    },
    {
      "epoch": 1.385390428211587,
      "grad_norm": 587531.6875,
      "learning_rate": 0.00016048938467074487,
      "loss": 0.6696,
      "step": 550
    },
    {
      "epoch": 1.4105793450881612,
      "grad_norm": 879147.25,
      "learning_rate": 0.00015976970133141418,
      "loss": 0.7831,
      "step": 560
    },
    {
      "epoch": 1.4357682619647356,
      "grad_norm": 219215.859375,
      "learning_rate": 0.0001590500179920835,
      "loss": 0.5681,
      "step": 570
    },
    {
      "epoch": 1.4609571788413098,
      "grad_norm": 447798.375,
      "learning_rate": 0.00015833033465275279,
      "loss": 0.8663,
      "step": 580
    },
    {
      "epoch": 1.486146095717884,
      "grad_norm": 708370.625,
      "learning_rate": 0.0001576106513134221,
      "loss": 0.7913,
      "step": 590
    },
    {
      "epoch": 1.5113350125944585,
      "grad_norm": 822008.0625,
      "learning_rate": 0.0001568909679740914,
      "loss": 0.8008,
      "step": 600
    },
    {
      "epoch": 1.5113350125944585,
      "eval_accuracy": 0.404296875,
      "eval_f1_macro": 0.3618779716937303,
      "eval_loss": 2.7039053440093994,
      "eval_runtime": 11.8087,
      "eval_samples_per_second": 86.716,
      "eval_steps_per_second": 5.42,
      "step": 600
    },
    {
      "epoch": 1.536523929471033,
      "grad_norm": 467169.0625,
      "learning_rate": 0.00015617128463476073,
      "loss": 0.7902,
      "step": 610
    },
    {
      "epoch": 1.561712846347607,
      "grad_norm": 385000.3125,
      "learning_rate": 0.00015545160129543,
      "loss": 0.7146,
      "step": 620
    },
    {
      "epoch": 1.5869017632241813,
      "grad_norm": 366275.21875,
      "learning_rate": 0.0001547319179560993,
      "loss": 0.6558,
      "step": 630
    },
    {
      "epoch": 1.6120906801007555,
      "grad_norm": 432902.1875,
      "learning_rate": 0.00015401223461676864,
      "loss": 0.7669,
      "step": 640
    },
    {
      "epoch": 1.63727959697733,
      "grad_norm": 426131.71875,
      "learning_rate": 0.00015329255127743795,
      "loss": 0.7121,
      "step": 650
    },
    {
      "epoch": 1.6624685138539044,
      "grad_norm": 426302.0,
      "learning_rate": 0.00015257286793810722,
      "loss": 0.9564,
      "step": 660
    },
    {
      "epoch": 1.6876574307304786,
      "grad_norm": 405949.34375,
      "learning_rate": 0.00015185318459877656,
      "loss": 0.6072,
      "step": 670
    },
    {
      "epoch": 1.7128463476070528,
      "grad_norm": 319128.53125,
      "learning_rate": 0.00015113350125944586,
      "loss": 0.5537,
      "step": 680
    },
    {
      "epoch": 1.7380352644836272,
      "grad_norm": 405533.625,
      "learning_rate": 0.00015041381792011514,
      "loss": 0.8143,
      "step": 690
    },
    {
      "epoch": 1.7632241813602016,
      "grad_norm": 357302.3125,
      "learning_rate": 0.00014969413458078447,
      "loss": 0.6885,
      "step": 700
    },
    {
      "epoch": 1.7632241813602016,
      "eval_accuracy": 0.3701171875,
      "eval_f1_macro": 0.2926079223907959,
      "eval_loss": 3.1089859008789062,
      "eval_runtime": 11.4817,
      "eval_samples_per_second": 89.186,
      "eval_steps_per_second": 5.574,
      "step": 700
    },
    {
      "epoch": 1.7884130982367759,
      "grad_norm": 448783.3125,
      "learning_rate": 0.00014897445124145378,
      "loss": 0.8981,
      "step": 710
    },
    {
      "epoch": 1.81360201511335,
      "grad_norm": 663808.1875,
      "learning_rate": 0.00014825476790212308,
      "loss": 0.6493,
      "step": 720
    },
    {
      "epoch": 1.8387909319899243,
      "grad_norm": 1166550.75,
      "learning_rate": 0.00014753508456279236,
      "loss": 0.6288,
      "step": 730
    },
    {
      "epoch": 1.8639798488664987,
      "grad_norm": 774232.6875,
      "learning_rate": 0.0001468154012234617,
      "loss": 0.8717,
      "step": 740
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 565763.0,
      "learning_rate": 0.000146095717884131,
      "loss": 0.7199,
      "step": 750
    },
    {
      "epoch": 1.9143576826196473,
      "grad_norm": 365545.625,
      "learning_rate": 0.0001453760345448003,
      "loss": 0.721,
      "step": 760
    },
    {
      "epoch": 1.9395465994962215,
      "grad_norm": 616748.25,
      "learning_rate": 0.0001446563512054696,
      "loss": 0.6793,
      "step": 770
    },
    {
      "epoch": 1.964735516372796,
      "grad_norm": 503789.65625,
      "learning_rate": 0.0001439366678661389,
      "loss": 0.5703,
      "step": 780
    },
    {
      "epoch": 1.9899244332493704,
      "grad_norm": 619143.4375,
      "learning_rate": 0.00014321698452680822,
      "loss": 0.679,
      "step": 790
    },
    {
      "epoch": 2.0151133501259446,
      "grad_norm": 661451.3125,
      "learning_rate": 0.00014249730118747752,
      "loss": 0.839,
      "step": 800
    },
    {
      "epoch": 2.0151133501259446,
      "eval_accuracy": 0.484375,
      "eval_f1_macro": 0.414907768165119,
      "eval_loss": 2.584502935409546,
      "eval_runtime": 11.5153,
      "eval_samples_per_second": 88.925,
      "eval_steps_per_second": 5.558,
      "step": 800
    },
    {
      "epoch": 2.040302267002519,
      "grad_norm": 316634.78125,
      "learning_rate": 0.00014177761784814683,
      "loss": 0.4317,
      "step": 810
    },
    {
      "epoch": 2.065491183879093,
      "grad_norm": 347729.625,
      "learning_rate": 0.00014105793450881613,
      "loss": 0.3792,
      "step": 820
    },
    {
      "epoch": 2.0906801007556677,
      "grad_norm": 659445.4375,
      "learning_rate": 0.00014033825116948544,
      "loss": 0.452,
      "step": 830
    },
    {
      "epoch": 2.115869017632242,
      "grad_norm": 735795.1875,
      "learning_rate": 0.00013961856783015474,
      "loss": 0.5447,
      "step": 840
    },
    {
      "epoch": 2.141057934508816,
      "grad_norm": 107777.5390625,
      "learning_rate": 0.00013889888449082404,
      "loss": 0.5247,
      "step": 850
    },
    {
      "epoch": 2.1662468513853903,
      "grad_norm": 577874.5,
      "learning_rate": 0.00013817920115149335,
      "loss": 0.5108,
      "step": 860
    },
    {
      "epoch": 2.1914357682619645,
      "grad_norm": 441668.625,
      "learning_rate": 0.00013745951781216265,
      "loss": 0.4324,
      "step": 870
    },
    {
      "epoch": 2.216624685138539,
      "grad_norm": 489653.59375,
      "learning_rate": 0.00013673983447283196,
      "loss": 0.4427,
      "step": 880
    },
    {
      "epoch": 2.2418136020151134,
      "grad_norm": 296346.84375,
      "learning_rate": 0.00013602015113350126,
      "loss": 0.5105,
      "step": 890
    },
    {
      "epoch": 2.2670025188916876,
      "grad_norm": 187145.796875,
      "learning_rate": 0.00013530046779417057,
      "loss": 0.325,
      "step": 900
    },
    {
      "epoch": 2.2670025188916876,
      "eval_accuracy": 0.5068359375,
      "eval_f1_macro": 0.41278416064519763,
      "eval_loss": 2.5142855644226074,
      "eval_runtime": 12.173,
      "eval_samples_per_second": 84.12,
      "eval_steps_per_second": 5.258,
      "step": 900
    },
    {
      "epoch": 2.292191435768262,
      "grad_norm": 406714.6875,
      "learning_rate": 0.00013458078445483987,
      "loss": 0.4576,
      "step": 910
    },
    {
      "epoch": 2.3173803526448364,
      "grad_norm": 379892.40625,
      "learning_rate": 0.00013386110111550918,
      "loss": 0.4179,
      "step": 920
    },
    {
      "epoch": 2.3425692695214106,
      "grad_norm": 540661.9375,
      "learning_rate": 0.00013314141777617848,
      "loss": 0.3386,
      "step": 930
    },
    {
      "epoch": 2.367758186397985,
      "grad_norm": 850949.25,
      "learning_rate": 0.0001324217344368478,
      "loss": 0.5695,
      "step": 940
    },
    {
      "epoch": 2.392947103274559,
      "grad_norm": 363627.53125,
      "learning_rate": 0.0001317020510975171,
      "loss": 0.3362,
      "step": 950
    },
    {
      "epoch": 2.4181360201511337,
      "grad_norm": 644468.0625,
      "learning_rate": 0.0001309823677581864,
      "loss": 0.444,
      "step": 960
    },
    {
      "epoch": 2.443324937027708,
      "grad_norm": 164236.78125,
      "learning_rate": 0.00013026268441885573,
      "loss": 0.285,
      "step": 970
    },
    {
      "epoch": 2.468513853904282,
      "grad_norm": 688494.375,
      "learning_rate": 0.000129543001079525,
      "loss": 0.4024,
      "step": 980
    },
    {
      "epoch": 2.4937027707808563,
      "grad_norm": 486211.0625,
      "learning_rate": 0.0001288233177401943,
      "loss": 0.4544,
      "step": 990
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 273390.03125,
      "learning_rate": 0.00012810363440086365,
      "loss": 0.4501,
      "step": 1000
    },
    {
      "epoch": 2.5188916876574305,
      "eval_accuracy": 0.4482421875,
      "eval_f1_macro": 0.40562802574511003,
      "eval_loss": 2.7683629989624023,
      "eval_runtime": 11.8873,
      "eval_samples_per_second": 86.142,
      "eval_steps_per_second": 5.384,
      "step": 1000
    },
    {
      "epoch": 2.544080604534005,
      "grad_norm": 730712.125,
      "learning_rate": 0.00012738395106153292,
      "loss": 0.5234,
      "step": 1010
    },
    {
      "epoch": 2.5692695214105794,
      "grad_norm": 369909.34375,
      "learning_rate": 0.00012666426772220223,
      "loss": 0.4051,
      "step": 1020
    },
    {
      "epoch": 2.5944584382871536,
      "grad_norm": 507635.78125,
      "learning_rate": 0.00012594458438287153,
      "loss": 0.2967,
      "step": 1030
    },
    {
      "epoch": 2.619647355163728,
      "grad_norm": 688013.125,
      "learning_rate": 0.00012522490104354087,
      "loss": 0.5321,
      "step": 1040
    },
    {
      "epoch": 2.644836272040302,
      "grad_norm": 505216.40625,
      "learning_rate": 0.00012450521770421014,
      "loss": 0.4807,
      "step": 1050
    },
    {
      "epoch": 2.6700251889168767,
      "grad_norm": 252679.53125,
      "learning_rate": 0.00012378553436487945,
      "loss": 0.4088,
      "step": 1060
    },
    {
      "epoch": 2.695214105793451,
      "grad_norm": 235546.9375,
      "learning_rate": 0.00012306585102554878,
      "loss": 0.393,
      "step": 1070
    },
    {
      "epoch": 2.720403022670025,
      "grad_norm": 398018.84375,
      "learning_rate": 0.00012234616768621808,
      "loss": 0.3694,
      "step": 1080
    },
    {
      "epoch": 2.7455919395465997,
      "grad_norm": 374467.3125,
      "learning_rate": 0.00012162648434688738,
      "loss": 0.3599,
      "step": 1090
    },
    {
      "epoch": 2.770780856423174,
      "grad_norm": 533788.9375,
      "learning_rate": 0.0001209068010075567,
      "loss": 0.3191,
      "step": 1100
    },
    {
      "epoch": 2.770780856423174,
      "eval_accuracy": 0.5146484375,
      "eval_f1_macro": 0.432831730682114,
      "eval_loss": 2.467733144760132,
      "eval_runtime": 11.9402,
      "eval_samples_per_second": 85.761,
      "eval_steps_per_second": 5.36,
      "step": 1100
    },
    {
      "epoch": 2.795969773299748,
      "grad_norm": 27070.849609375,
      "learning_rate": 0.00012018711766822599,
      "loss": 0.2222,
      "step": 1110
    },
    {
      "epoch": 2.8211586901763224,
      "grad_norm": 742011.0625,
      "learning_rate": 0.00011946743432889529,
      "loss": 0.3194,
      "step": 1120
    },
    {
      "epoch": 2.8463476070528966,
      "grad_norm": 508084.53125,
      "learning_rate": 0.00011874775098956458,
      "loss": 0.5068,
      "step": 1130
    },
    {
      "epoch": 2.8715365239294712,
      "grad_norm": 472090.21875,
      "learning_rate": 0.00011802806765023391,
      "loss": 0.3457,
      "step": 1140
    },
    {
      "epoch": 2.8967254408060454,
      "grad_norm": 345946.625,
      "learning_rate": 0.0001173083843109032,
      "loss": 0.4024,
      "step": 1150
    },
    {
      "epoch": 2.9219143576826196,
      "grad_norm": 116557.78125,
      "learning_rate": 0.00011658870097157251,
      "loss": 0.3248,
      "step": 1160
    },
    {
      "epoch": 2.947103274559194,
      "grad_norm": 544007.125,
      "learning_rate": 0.00011586901763224183,
      "loss": 0.2957,
      "step": 1170
    },
    {
      "epoch": 2.972292191435768,
      "grad_norm": 558989.0,
      "learning_rate": 0.00011514933429291112,
      "loss": 0.3386,
      "step": 1180
    },
    {
      "epoch": 2.9974811083123427,
      "grad_norm": 509623.65625,
      "learning_rate": 0.00011442965095358043,
      "loss": 0.4525,
      "step": 1190
    },
    {
      "epoch": 3.022670025188917,
      "grad_norm": 374462.28125,
      "learning_rate": 0.00011370996761424974,
      "loss": 0.1664,
      "step": 1200
    },
    {
      "epoch": 3.022670025188917,
      "eval_accuracy": 0.5361328125,
      "eval_f1_macro": 0.4597358092759388,
      "eval_loss": 2.477670192718506,
      "eval_runtime": 11.8885,
      "eval_samples_per_second": 86.134,
      "eval_steps_per_second": 5.383,
      "step": 1200
    },
    {
      "epoch": 3.047858942065491,
      "grad_norm": 417895.8125,
      "learning_rate": 0.00011299028427491905,
      "loss": 0.2631,
      "step": 1210
    },
    {
      "epoch": 3.0730478589420653,
      "grad_norm": 307081.21875,
      "learning_rate": 0.00011227060093558834,
      "loss": 0.1544,
      "step": 1220
    },
    {
      "epoch": 3.09823677581864,
      "grad_norm": 1034528.125,
      "learning_rate": 0.00011155091759625764,
      "loss": 0.2107,
      "step": 1230
    },
    {
      "epoch": 3.123425692695214,
      "grad_norm": 659800.0625,
      "learning_rate": 0.00011083123425692696,
      "loss": 0.3405,
      "step": 1240
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 129759.8671875,
      "learning_rate": 0.00011011155091759627,
      "loss": 0.2225,
      "step": 1250
    },
    {
      "epoch": 3.1738035264483626,
      "grad_norm": 560157.25,
      "learning_rate": 0.00010939186757826556,
      "loss": 0.1545,
      "step": 1260
    },
    {
      "epoch": 3.1989924433249373,
      "grad_norm": 253132.671875,
      "learning_rate": 0.00010867218423893488,
      "loss": 0.3073,
      "step": 1270
    },
    {
      "epoch": 3.2241813602015115,
      "grad_norm": 407181.90625,
      "learning_rate": 0.00010795250089960418,
      "loss": 0.3247,
      "step": 1280
    },
    {
      "epoch": 3.2493702770780857,
      "grad_norm": 433259.09375,
      "learning_rate": 0.00010723281756027347,
      "loss": 0.127,
      "step": 1290
    },
    {
      "epoch": 3.27455919395466,
      "grad_norm": 187479.984375,
      "learning_rate": 0.0001065131342209428,
      "loss": 0.1469,
      "step": 1300
    },
    {
      "epoch": 3.27455919395466,
      "eval_accuracy": 0.5205078125,
      "eval_f1_macro": 0.4495261806770087,
      "eval_loss": 2.6402528285980225,
      "eval_runtime": 11.5773,
      "eval_samples_per_second": 88.449,
      "eval_steps_per_second": 5.528,
      "step": 1300
    },
    {
      "epoch": 3.299748110831234,
      "grad_norm": 736520.0625,
      "learning_rate": 0.0001057934508816121,
      "loss": 0.2567,
      "step": 1310
    },
    {
      "epoch": 3.3249370277078087,
      "grad_norm": 376107.03125,
      "learning_rate": 0.0001050737675422814,
      "loss": 0.1442,
      "step": 1320
    },
    {
      "epoch": 3.350125944584383,
      "grad_norm": 723020.3125,
      "learning_rate": 0.0001043540842029507,
      "loss": 0.1422,
      "step": 1330
    },
    {
      "epoch": 3.375314861460957,
      "grad_norm": 60796.48046875,
      "learning_rate": 0.00010363440086362001,
      "loss": 0.2077,
      "step": 1340
    },
    {
      "epoch": 3.4005037783375314,
      "grad_norm": 624072.0,
      "learning_rate": 0.00010291471752428932,
      "loss": 0.1422,
      "step": 1350
    },
    {
      "epoch": 3.4256926952141056,
      "grad_norm": 191143.5625,
      "learning_rate": 0.00010219503418495862,
      "loss": 0.2425,
      "step": 1360
    },
    {
      "epoch": 3.4508816120906802,
      "grad_norm": 291693.75,
      "learning_rate": 0.00010147535084562794,
      "loss": 0.2693,
      "step": 1370
    },
    {
      "epoch": 3.4760705289672544,
      "grad_norm": 516212.4375,
      "learning_rate": 0.00010075566750629723,
      "loss": 0.2397,
      "step": 1380
    },
    {
      "epoch": 3.5012594458438286,
      "grad_norm": 273365.9375,
      "learning_rate": 0.00010003598416696654,
      "loss": 0.1774,
      "step": 1390
    },
    {
      "epoch": 3.5264483627204033,
      "grad_norm": 518260.21875,
      "learning_rate": 9.931630082763584e-05,
      "loss": 0.3063,
      "step": 1400
    },
    {
      "epoch": 3.5264483627204033,
      "eval_accuracy": 0.5009765625,
      "eval_f1_macro": 0.441499104615912,
      "eval_loss": 2.7999606132507324,
      "eval_runtime": 11.7898,
      "eval_samples_per_second": 86.855,
      "eval_steps_per_second": 5.428,
      "step": 1400
    },
    {
      "epoch": 3.551637279596977,
      "grad_norm": 552209.5,
      "learning_rate": 9.859661748830516e-05,
      "loss": 0.2254,
      "step": 1410
    },
    {
      "epoch": 3.5768261964735517,
      "grad_norm": 7195.90625,
      "learning_rate": 9.787693414897445e-05,
      "loss": 0.1378,
      "step": 1420
    },
    {
      "epoch": 3.602015113350126,
      "grad_norm": 879434.0625,
      "learning_rate": 9.715725080964376e-05,
      "loss": 0.2466,
      "step": 1430
    },
    {
      "epoch": 3.6272040302267,
      "grad_norm": 116491.9609375,
      "learning_rate": 9.643756747031306e-05,
      "loss": 0.2094,
      "step": 1440
    },
    {
      "epoch": 3.652392947103275,
      "grad_norm": 524990.0625,
      "learning_rate": 9.571788413098237e-05,
      "loss": 0.2522,
      "step": 1450
    },
    {
      "epoch": 3.677581863979849,
      "grad_norm": 939990.8125,
      "learning_rate": 9.499820079165168e-05,
      "loss": 0.2382,
      "step": 1460
    },
    {
      "epoch": 3.702770780856423,
      "grad_norm": 337334.625,
      "learning_rate": 9.427851745232098e-05,
      "loss": 0.299,
      "step": 1470
    },
    {
      "epoch": 3.7279596977329974,
      "grad_norm": 322309.28125,
      "learning_rate": 9.35588341129903e-05,
      "loss": 0.2307,
      "step": 1480
    },
    {
      "epoch": 3.7531486146095716,
      "grad_norm": 80700.328125,
      "learning_rate": 9.283915077365959e-05,
      "loss": 0.1816,
      "step": 1490
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 667783.125,
      "learning_rate": 9.21194674343289e-05,
      "loss": 0.1786,
      "step": 1500
    },
    {
      "epoch": 3.7783375314861463,
      "eval_accuracy": 0.533203125,
      "eval_f1_macro": 0.45250014527358284,
      "eval_loss": 2.8164846897125244,
      "eval_runtime": 11.728,
      "eval_samples_per_second": 87.312,
      "eval_steps_per_second": 5.457,
      "step": 1500
    },
    {
      "epoch": 3.8035264483627205,
      "grad_norm": 340034.53125,
      "learning_rate": 9.139978409499821e-05,
      "loss": 0.1934,
      "step": 1510
    },
    {
      "epoch": 3.8287153652392947,
      "grad_norm": 569343.8125,
      "learning_rate": 9.068010075566751e-05,
      "loss": 0.3571,
      "step": 1520
    },
    {
      "epoch": 3.853904282115869,
      "grad_norm": 73828.2578125,
      "learning_rate": 8.996041741633682e-05,
      "loss": 0.2595,
      "step": 1530
    },
    {
      "epoch": 3.879093198992443,
      "grad_norm": 610885.5625,
      "learning_rate": 8.924073407700611e-05,
      "loss": 0.2454,
      "step": 1540
    },
    {
      "epoch": 3.9042821158690177,
      "grad_norm": 549705.5,
      "learning_rate": 8.852105073767543e-05,
      "loss": 0.2246,
      "step": 1550
    },
    {
      "epoch": 3.929471032745592,
      "grad_norm": 946495.1875,
      "learning_rate": 8.780136739834473e-05,
      "loss": 0.3156,
      "step": 1560
    },
    {
      "epoch": 3.954659949622166,
      "grad_norm": 89126.0546875,
      "learning_rate": 8.708168405901404e-05,
      "loss": 0.127,
      "step": 1570
    },
    {
      "epoch": 3.979848866498741,
      "grad_norm": 364322.40625,
      "learning_rate": 8.636200071968334e-05,
      "loss": 0.2114,
      "step": 1580
    },
    {
      "epoch": 4.005037783375315,
      "grad_norm": 646444.625,
      "learning_rate": 8.564231738035265e-05,
      "loss": 0.1633,
      "step": 1590
    },
    {
      "epoch": 4.030226700251889,
      "grad_norm": 281335.90625,
      "learning_rate": 8.492263404102195e-05,
      "loss": 0.0687,
      "step": 1600
    },
    {
      "epoch": 4.030226700251889,
      "eval_accuracy": 0.568359375,
      "eval_f1_macro": 0.49420184108051124,
      "eval_loss": 2.9026849269866943,
      "eval_runtime": 11.365,
      "eval_samples_per_second": 90.101,
      "eval_steps_per_second": 5.631,
      "step": 1600
    },
    {
      "epoch": 4.055415617128464,
      "grad_norm": 12626.9794921875,
      "learning_rate": 8.420295070169126e-05,
      "loss": 0.0641,
      "step": 1610
    },
    {
      "epoch": 4.080604534005038,
      "grad_norm": 55986.578125,
      "learning_rate": 8.348326736236056e-05,
      "loss": 0.0647,
      "step": 1620
    },
    {
      "epoch": 4.105793450881612,
      "grad_norm": 509303.3125,
      "learning_rate": 8.276358402302987e-05,
      "loss": 0.0746,
      "step": 1630
    },
    {
      "epoch": 4.130982367758186,
      "grad_norm": 21481.740234375,
      "learning_rate": 8.204390068369917e-05,
      "loss": 0.0746,
      "step": 1640
    },
    {
      "epoch": 4.156171284634761,
      "grad_norm": 11360.7412109375,
      "learning_rate": 8.132421734436848e-05,
      "loss": 0.0555,
      "step": 1650
    },
    {
      "epoch": 4.181360201511335,
      "grad_norm": 396739.0,
      "learning_rate": 8.06045340050378e-05,
      "loss": 0.0595,
      "step": 1660
    },
    {
      "epoch": 4.206549118387909,
      "grad_norm": 8099.65478515625,
      "learning_rate": 7.988485066570709e-05,
      "loss": 0.1391,
      "step": 1670
    },
    {
      "epoch": 4.231738035264484,
      "grad_norm": 208596.65625,
      "learning_rate": 7.916516732637639e-05,
      "loss": 0.1038,
      "step": 1680
    },
    {
      "epoch": 4.2569269521410575,
      "grad_norm": 359488.34375,
      "learning_rate": 7.84454839870457e-05,
      "loss": 0.025,
      "step": 1690
    },
    {
      "epoch": 4.282115869017632,
      "grad_norm": 247004.875,
      "learning_rate": 7.7725800647715e-05,
      "loss": 0.0427,
      "step": 1700
    },
    {
      "epoch": 4.282115869017632,
      "eval_accuracy": 0.4912109375,
      "eval_f1_macro": 0.4362345681944286,
      "eval_loss": 3.321627616882324,
      "eval_runtime": 11.8095,
      "eval_samples_per_second": 86.71,
      "eval_steps_per_second": 5.419,
      "step": 1700
    },
    {
      "epoch": 4.307304785894207,
      "grad_norm": 22134.611328125,
      "learning_rate": 7.700611730838432e-05,
      "loss": 0.1015,
      "step": 1710
    },
    {
      "epoch": 4.332493702770781,
      "grad_norm": 163005.84375,
      "learning_rate": 7.628643396905361e-05,
      "loss": 0.0524,
      "step": 1720
    },
    {
      "epoch": 4.357682619647355,
      "grad_norm": 25011.78515625,
      "learning_rate": 7.556675062972293e-05,
      "loss": 0.1,
      "step": 1730
    },
    {
      "epoch": 4.382871536523929,
      "grad_norm": 606683.75,
      "learning_rate": 7.484706729039224e-05,
      "loss": 0.1503,
      "step": 1740
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 411167.28125,
      "learning_rate": 7.412738395106154e-05,
      "loss": 0.1075,
      "step": 1750
    },
    {
      "epoch": 4.433249370277078,
      "grad_norm": 20846.5390625,
      "learning_rate": 7.340770061173085e-05,
      "loss": 0.0793,
      "step": 1760
    },
    {
      "epoch": 4.458438287153652,
      "grad_norm": 35871.06640625,
      "learning_rate": 7.268801727240015e-05,
      "loss": 0.0607,
      "step": 1770
    },
    {
      "epoch": 4.483627204030227,
      "grad_norm": 58382.01171875,
      "learning_rate": 7.196833393306946e-05,
      "loss": 0.082,
      "step": 1780
    },
    {
      "epoch": 4.508816120906801,
      "grad_norm": 13096.5205078125,
      "learning_rate": 7.124865059373876e-05,
      "loss": 0.0754,
      "step": 1790
    },
    {
      "epoch": 4.534005037783375,
      "grad_norm": 21618.96875,
      "learning_rate": 7.052896725440807e-05,
      "loss": 0.1825,
      "step": 1800
    },
    {
      "epoch": 4.534005037783375,
      "eval_accuracy": 0.53125,
      "eval_f1_macro": 0.4663662196170286,
      "eval_loss": 3.1456074714660645,
      "eval_runtime": 11.9035,
      "eval_samples_per_second": 86.025,
      "eval_steps_per_second": 5.377,
      "step": 1800
    },
    {
      "epoch": 4.55919395465995,
      "grad_norm": 97136.0546875,
      "learning_rate": 6.980928391507737e-05,
      "loss": 0.0821,
      "step": 1810
    },
    {
      "epoch": 4.584382871536524,
      "grad_norm": 14197.41796875,
      "learning_rate": 6.908960057574667e-05,
      "loss": 0.1245,
      "step": 1820
    },
    {
      "epoch": 4.609571788413098,
      "grad_norm": 792662.5,
      "learning_rate": 6.836991723641598e-05,
      "loss": 0.0678,
      "step": 1830
    },
    {
      "epoch": 4.634760705289673,
      "grad_norm": 70302.1484375,
      "learning_rate": 6.765023389708528e-05,
      "loss": 0.0369,
      "step": 1840
    },
    {
      "epoch": 4.659949622166247,
      "grad_norm": 315541.75,
      "learning_rate": 6.693055055775459e-05,
      "loss": 0.086,
      "step": 1850
    },
    {
      "epoch": 4.685138539042821,
      "grad_norm": 426322.5625,
      "learning_rate": 6.62108672184239e-05,
      "loss": 0.1367,
      "step": 1860
    },
    {
      "epoch": 4.710327455919396,
      "grad_norm": 5201.7265625,
      "learning_rate": 6.54911838790932e-05,
      "loss": 0.0518,
      "step": 1870
    },
    {
      "epoch": 4.73551637279597,
      "grad_norm": 16552.916015625,
      "learning_rate": 6.47715005397625e-05,
      "loss": 0.0857,
      "step": 1880
    },
    {
      "epoch": 4.760705289672544,
      "grad_norm": 74746.0234375,
      "learning_rate": 6.405181720043182e-05,
      "loss": 0.178,
      "step": 1890
    },
    {
      "epoch": 4.785894206549118,
      "grad_norm": 30216.1328125,
      "learning_rate": 6.333213386110111e-05,
      "loss": 0.0758,
      "step": 1900
    },
    {
      "epoch": 4.785894206549118,
      "eval_accuracy": 0.5546875,
      "eval_f1_macro": 0.4578418057053371,
      "eval_loss": 3.2782468795776367,
      "eval_runtime": 11.4304,
      "eval_samples_per_second": 89.586,
      "eval_steps_per_second": 5.599,
      "step": 1900
    },
    {
      "epoch": 4.811083123425693,
      "grad_norm": 122862.0703125,
      "learning_rate": 6.261245052177043e-05,
      "loss": 0.1008,
      "step": 1910
    },
    {
      "epoch": 4.836272040302267,
      "grad_norm": 355039.96875,
      "learning_rate": 6.189276718243972e-05,
      "loss": 0.0619,
      "step": 1920
    },
    {
      "epoch": 4.861460957178841,
      "grad_norm": 180546.546875,
      "learning_rate": 6.117308384310904e-05,
      "loss": 0.212,
      "step": 1930
    },
    {
      "epoch": 4.886649874055416,
      "grad_norm": 1149145.875,
      "learning_rate": 6.045340050377835e-05,
      "loss": 0.1821,
      "step": 1940
    },
    {
      "epoch": 4.91183879093199,
      "grad_norm": 197082.046875,
      "learning_rate": 5.9733717164447645e-05,
      "loss": 0.0539,
      "step": 1950
    },
    {
      "epoch": 4.937027707808564,
      "grad_norm": 361.24676513671875,
      "learning_rate": 5.901403382511696e-05,
      "loss": 0.1526,
      "step": 1960
    },
    {
      "epoch": 4.962216624685139,
      "grad_norm": 362799.34375,
      "learning_rate": 5.8294350485786255e-05,
      "loss": 0.0838,
      "step": 1970
    },
    {
      "epoch": 4.987405541561713,
      "grad_norm": 4683.50732421875,
      "learning_rate": 5.757466714645556e-05,
      "loss": 0.0995,
      "step": 1980
    },
    {
      "epoch": 5.012594458438287,
      "grad_norm": 5118.85986328125,
      "learning_rate": 5.685498380712487e-05,
      "loss": 0.0488,
      "step": 1990
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 6831.2880859375,
      "learning_rate": 5.613530046779417e-05,
      "loss": 0.0471,
      "step": 2000
    },
    {
      "epoch": 5.037783375314861,
      "eval_accuracy": 0.5517578125,
      "eval_f1_macro": 0.4725336026660133,
      "eval_loss": 3.334784507751465,
      "eval_runtime": 11.4554,
      "eval_samples_per_second": 89.39,
      "eval_steps_per_second": 5.587,
      "step": 2000
    },
    {
      "epoch": 5.062972292191436,
      "grad_norm": 88275.8359375,
      "learning_rate": 5.541561712846348e-05,
      "loss": 0.0503,
      "step": 2010
    },
    {
      "epoch": 5.08816120906801,
      "grad_norm": 1161589.25,
      "learning_rate": 5.469593378913278e-05,
      "loss": 0.0326,
      "step": 2020
    },
    {
      "epoch": 5.113350125944584,
      "grad_norm": 263008.125,
      "learning_rate": 5.397625044980209e-05,
      "loss": 0.103,
      "step": 2030
    },
    {
      "epoch": 5.138539042821159,
      "grad_norm": 2268.762939453125,
      "learning_rate": 5.32565671104714e-05,
      "loss": 0.0093,
      "step": 2040
    },
    {
      "epoch": 5.163727959697733,
      "grad_norm": 530844.6875,
      "learning_rate": 5.25368837711407e-05,
      "loss": 0.0919,
      "step": 2050
    },
    {
      "epoch": 5.188916876574307,
      "grad_norm": 181022.703125,
      "learning_rate": 5.1817200431810006e-05,
      "loss": 0.049,
      "step": 2060
    },
    {
      "epoch": 5.214105793450882,
      "grad_norm": 1360.029541015625,
      "learning_rate": 5.109751709247931e-05,
      "loss": 0.0656,
      "step": 2070
    },
    {
      "epoch": 5.239294710327456,
      "grad_norm": 541578.0625,
      "learning_rate": 5.0377833753148616e-05,
      "loss": 0.0287,
      "step": 2080
    },
    {
      "epoch": 5.26448362720403,
      "grad_norm": 8820.35546875,
      "learning_rate": 4.965815041381792e-05,
      "loss": 0.0252,
      "step": 2090
    },
    {
      "epoch": 5.289672544080605,
      "grad_norm": 50239.13671875,
      "learning_rate": 4.8938467074487226e-05,
      "loss": 0.0512,
      "step": 2100
    },
    {
      "epoch": 5.289672544080605,
      "eval_accuracy": 0.5283203125,
      "eval_f1_macro": 0.45143425422893263,
      "eval_loss": 3.718236207962036,
      "eval_runtime": 11.3333,
      "eval_samples_per_second": 90.353,
      "eval_steps_per_second": 5.647,
      "step": 2100
    },
    {
      "epoch": 5.314861460957179,
      "grad_norm": 1060627.125,
      "learning_rate": 4.821878373515653e-05,
      "loss": 0.0412,
      "step": 2110
    },
    {
      "epoch": 5.340050377833753,
      "grad_norm": 258190.28125,
      "learning_rate": 4.749910039582584e-05,
      "loss": 0.1177,
      "step": 2120
    },
    {
      "epoch": 5.365239294710327,
      "grad_norm": 126000.2265625,
      "learning_rate": 4.677941705649515e-05,
      "loss": 0.0049,
      "step": 2130
    },
    {
      "epoch": 5.390428211586902,
      "grad_norm": 2938.684814453125,
      "learning_rate": 4.605973371716445e-05,
      "loss": 0.1016,
      "step": 2140
    },
    {
      "epoch": 5.415617128463476,
      "grad_norm": 191.32302856445312,
      "learning_rate": 4.534005037783376e-05,
      "loss": 0.0051,
      "step": 2150
    },
    {
      "epoch": 5.44080604534005,
      "grad_norm": 70.82958984375,
      "learning_rate": 4.4620367038503055e-05,
      "loss": 0.0335,
      "step": 2160
    },
    {
      "epoch": 5.465994962216625,
      "grad_norm": 2224.07568359375,
      "learning_rate": 4.390068369917237e-05,
      "loss": 0.0019,
      "step": 2170
    },
    {
      "epoch": 5.491183879093199,
      "grad_norm": 7065.70703125,
      "learning_rate": 4.318100035984167e-05,
      "loss": 0.0489,
      "step": 2180
    },
    {
      "epoch": 5.516372795969773,
      "grad_norm": 218.8759765625,
      "learning_rate": 4.246131702051098e-05,
      "loss": 0.0761,
      "step": 2190
    },
    {
      "epoch": 5.541561712846348,
      "grad_norm": 2789.82763671875,
      "learning_rate": 4.174163368118028e-05,
      "loss": 0.0095,
      "step": 2200
    },
    {
      "epoch": 5.541561712846348,
      "eval_accuracy": 0.5341796875,
      "eval_f1_macro": 0.478510457034419,
      "eval_loss": 3.902801036834717,
      "eval_runtime": 11.7659,
      "eval_samples_per_second": 87.031,
      "eval_steps_per_second": 5.439,
      "step": 2200
    },
    {
      "epoch": 5.566750629722922,
      "grad_norm": 33521.52734375,
      "learning_rate": 4.1021950341849587e-05,
      "loss": 0.0445,
      "step": 2210
    },
    {
      "epoch": 5.591939546599496,
      "grad_norm": 339091.78125,
      "learning_rate": 4.03022670025189e-05,
      "loss": 0.0717,
      "step": 2220
    },
    {
      "epoch": 5.617128463476071,
      "grad_norm": 570797.6875,
      "learning_rate": 3.9582583663188196e-05,
      "loss": 0.0664,
      "step": 2230
    },
    {
      "epoch": 5.642317380352645,
      "grad_norm": 638298.25,
      "learning_rate": 3.88629003238575e-05,
      "loss": 0.0152,
      "step": 2240
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 1308.215087890625,
      "learning_rate": 3.8143216984526806e-05,
      "loss": 0.0029,
      "step": 2250
    },
    {
      "epoch": 5.692695214105793,
      "grad_norm": 47.38713836669922,
      "learning_rate": 3.742353364519612e-05,
      "loss": 0.0199,
      "step": 2260
    },
    {
      "epoch": 5.717884130982368,
      "grad_norm": 56138.89453125,
      "learning_rate": 3.670385030586542e-05,
      "loss": 0.0748,
      "step": 2270
    },
    {
      "epoch": 5.7430730478589425,
      "grad_norm": 49.50386047363281,
      "learning_rate": 3.598416696653473e-05,
      "loss": 0.0174,
      "step": 2280
    },
    {
      "epoch": 5.768261964735516,
      "grad_norm": 7627.2783203125,
      "learning_rate": 3.526448362720403e-05,
      "loss": 0.002,
      "step": 2290
    },
    {
      "epoch": 5.793450881612091,
      "grad_norm": 123.52435302734375,
      "learning_rate": 3.454480028787334e-05,
      "loss": 0.0247,
      "step": 2300
    },
    {
      "epoch": 5.793450881612091,
      "eval_accuracy": 0.5712890625,
      "eval_f1_macro": 0.48791698609028533,
      "eval_loss": 3.960580348968506,
      "eval_runtime": 11.8204,
      "eval_samples_per_second": 86.63,
      "eval_steps_per_second": 5.414,
      "step": 2300
    },
    {
      "epoch": 5.818639798488665,
      "grad_norm": 534.7195434570312,
      "learning_rate": 3.382511694854264e-05,
      "loss": 0.0496,
      "step": 2310
    },
    {
      "epoch": 5.843828715365239,
      "grad_norm": 17011.57421875,
      "learning_rate": 3.310543360921195e-05,
      "loss": 0.1476,
      "step": 2320
    },
    {
      "epoch": 5.869017632241814,
      "grad_norm": 386354.28125,
      "learning_rate": 3.238575026988125e-05,
      "loss": 0.0593,
      "step": 2330
    },
    {
      "epoch": 5.894206549118388,
      "grad_norm": 102287.3828125,
      "learning_rate": 3.166606693055056e-05,
      "loss": 0.0117,
      "step": 2340
    },
    {
      "epoch": 5.919395465994962,
      "grad_norm": 29.17066764831543,
      "learning_rate": 3.094638359121986e-05,
      "loss": 0.0702,
      "step": 2350
    },
    {
      "epoch": 5.944584382871536,
      "grad_norm": 1611.7230224609375,
      "learning_rate": 3.0226700251889174e-05,
      "loss": 0.0687,
      "step": 2360
    },
    {
      "epoch": 5.969773299748111,
      "grad_norm": 168.02320861816406,
      "learning_rate": 2.950701691255848e-05,
      "loss": 0.0872,
      "step": 2370
    },
    {
      "epoch": 5.994962216624685,
      "grad_norm": 1058.51904296875,
      "learning_rate": 2.878733357322778e-05,
      "loss": 0.0008,
      "step": 2380
    },
    {
      "epoch": 6.020151133501259,
      "grad_norm": 23.99247932434082,
      "learning_rate": 2.8067650233897085e-05,
      "loss": 0.0526,
      "step": 2390
    },
    {
      "epoch": 6.045340050377834,
      "grad_norm": 2630.893798828125,
      "learning_rate": 2.734796689456639e-05,
      "loss": 0.0008,
      "step": 2400
    },
    {
      "epoch": 6.045340050377834,
      "eval_accuracy": 0.5654296875,
      "eval_f1_macro": 0.49182304765519874,
      "eval_loss": 4.129029750823975,
      "eval_runtime": 11.7471,
      "eval_samples_per_second": 87.17,
      "eval_steps_per_second": 5.448,
      "step": 2400
    },
    {
      "epoch": 6.0705289672544085,
      "grad_norm": 3671.589111328125,
      "learning_rate": 2.66282835552357e-05,
      "loss": 0.0009,
      "step": 2410
    },
    {
      "epoch": 6.095717884130982,
      "grad_norm": 246.3783721923828,
      "learning_rate": 2.5908600215905003e-05,
      "loss": 0.0279,
      "step": 2420
    },
    {
      "epoch": 6.120906801007557,
      "grad_norm": 250481.90625,
      "learning_rate": 2.5188916876574308e-05,
      "loss": 0.0367,
      "step": 2430
    },
    {
      "epoch": 6.146095717884131,
      "grad_norm": 5997.31396484375,
      "learning_rate": 2.4469233537243613e-05,
      "loss": 0.0002,
      "step": 2440
    },
    {
      "epoch": 6.171284634760705,
      "grad_norm": 294020.6875,
      "learning_rate": 2.374955019791292e-05,
      "loss": 0.0447,
      "step": 2450
    },
    {
      "epoch": 6.19647355163728,
      "grad_norm": 5173.7607421875,
      "learning_rate": 2.3029866858582226e-05,
      "loss": 0.0006,
      "step": 2460
    },
    {
      "epoch": 6.221662468513854,
      "grad_norm": 4.966667175292969,
      "learning_rate": 2.2310183519251528e-05,
      "loss": 0.0394,
      "step": 2470
    },
    {
      "epoch": 6.246851385390428,
      "grad_norm": 328896.6875,
      "learning_rate": 2.1590500179920836e-05,
      "loss": 0.0725,
      "step": 2480
    },
    {
      "epoch": 6.272040302267002,
      "grad_norm": 855.643798828125,
      "learning_rate": 2.087081684059014e-05,
      "loss": 0.0001,
      "step": 2490
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 18.617643356323242,
      "learning_rate": 2.015113350125945e-05,
      "loss": 0.0024,
      "step": 2500
    },
    {
      "epoch": 6.297229219143577,
      "eval_accuracy": 0.5654296875,
      "eval_f1_macro": 0.4862593826724424,
      "eval_loss": 4.414713382720947,
      "eval_runtime": 11.3987,
      "eval_samples_per_second": 89.835,
      "eval_steps_per_second": 5.615,
      "step": 2500
    },
    {
      "epoch": 6.3224181360201515,
      "grad_norm": 43.19185256958008,
      "learning_rate": 1.943145016192875e-05,
      "loss": 0.005,
      "step": 2510
    },
    {
      "epoch": 6.347607052896725,
      "grad_norm": 636742.8125,
      "learning_rate": 1.871176682259806e-05,
      "loss": 0.0187,
      "step": 2520
    },
    {
      "epoch": 6.3727959697733,
      "grad_norm": 14.576433181762695,
      "learning_rate": 1.7992083483267364e-05,
      "loss": 0.0008,
      "step": 2530
    },
    {
      "epoch": 6.3979848866498745,
      "grad_norm": 22.115917205810547,
      "learning_rate": 1.727240014393667e-05,
      "loss": 0.0038,
      "step": 2540
    },
    {
      "epoch": 6.423173803526448,
      "grad_norm": 618.1704711914062,
      "learning_rate": 1.6552716804605974e-05,
      "loss": 0.0685,
      "step": 2550
    },
    {
      "epoch": 6.448362720403023,
      "grad_norm": 279531.71875,
      "learning_rate": 1.583303346527528e-05,
      "loss": 0.0035,
      "step": 2560
    },
    {
      "epoch": 6.473551637279597,
      "grad_norm": 265.0247802734375,
      "learning_rate": 1.5113350125944587e-05,
      "loss": 0.0245,
      "step": 2570
    },
    {
      "epoch": 6.498740554156171,
      "grad_norm": 635805.875,
      "learning_rate": 1.439366678661389e-05,
      "loss": 0.0052,
      "step": 2580
    },
    {
      "epoch": 6.523929471032746,
      "grad_norm": 8.843326568603516,
      "learning_rate": 1.3673983447283195e-05,
      "loss": 0.0,
      "step": 2590
    },
    {
      "epoch": 6.54911838790932,
      "grad_norm": 87.60523986816406,
      "learning_rate": 1.2954300107952502e-05,
      "loss": 0.0002,
      "step": 2600
    },
    {
      "epoch": 6.54911838790932,
      "eval_accuracy": 0.5654296875,
      "eval_f1_macro": 0.4913330578351924,
      "eval_loss": 4.520939826965332,
      "eval_runtime": 11.8845,
      "eval_samples_per_second": 86.163,
      "eval_steps_per_second": 5.385,
      "step": 2600
    },
    {
      "epoch": 6.574307304785894,
      "grad_norm": 8.957763671875,
      "learning_rate": 1.2234616768621806e-05,
      "loss": 0.0,
      "step": 2610
    },
    {
      "epoch": 6.599496221662468,
      "grad_norm": 6.1030144691467285,
      "learning_rate": 1.1514933429291113e-05,
      "loss": 0.0009,
      "step": 2620
    },
    {
      "epoch": 6.624685138539043,
      "grad_norm": 4.862893581390381,
      "learning_rate": 1.0795250089960418e-05,
      "loss": 0.1231,
      "step": 2630
    },
    {
      "epoch": 6.6498740554156175,
      "grad_norm": 598.0420532226562,
      "learning_rate": 1.0075566750629725e-05,
      "loss": 0.0001,
      "step": 2640
    },
    {
      "epoch": 6.675062972292191,
      "grad_norm": 5.682479381561279,
      "learning_rate": 9.35588341129903e-06,
      "loss": 0.0365,
      "step": 2650
    },
    {
      "epoch": 6.700251889168766,
      "grad_norm": 1.800890564918518,
      "learning_rate": 8.636200071968334e-06,
      "loss": 0.133,
      "step": 2660
    },
    {
      "epoch": 6.72544080604534,
      "grad_norm": 718.6954345703125,
      "learning_rate": 7.91651673263764e-06,
      "loss": 0.0,
      "step": 2670
    },
    {
      "epoch": 6.750629722921914,
      "grad_norm": 338166.46875,
      "learning_rate": 7.196833393306945e-06,
      "loss": 0.034,
      "step": 2680
    },
    {
      "epoch": 6.775818639798489,
      "grad_norm": 471.62005615234375,
      "learning_rate": 6.477150053976251e-06,
      "loss": 0.0599,
      "step": 2690
    },
    {
      "epoch": 6.801007556675063,
      "grad_norm": 308.09417724609375,
      "learning_rate": 5.7574667146455565e-06,
      "loss": 0.0055,
      "step": 2700
    },
    {
      "epoch": 6.801007556675063,
      "eval_accuracy": 0.58203125,
      "eval_f1_macro": 0.5067161880167751,
      "eval_loss": 4.515384197235107,
      "eval_runtime": 11.8883,
      "eval_samples_per_second": 86.135,
      "eval_steps_per_second": 5.383,
      "step": 2700
    },
    {
      "epoch": 6.826196473551637,
      "grad_norm": 407.3280334472656,
      "learning_rate": 5.037783375314862e-06,
      "loss": 0.0,
      "step": 2710
    },
    {
      "epoch": 6.851385390428211,
      "grad_norm": 396.4019470214844,
      "learning_rate": 4.318100035984167e-06,
      "loss": 0.0015,
      "step": 2720
    },
    {
      "epoch": 6.876574307304786,
      "grad_norm": 2.2462317943573,
      "learning_rate": 3.5984166966534725e-06,
      "loss": 0.0007,
      "step": 2730
    },
    {
      "epoch": 6.9017632241813605,
      "grad_norm": 40.78224182128906,
      "learning_rate": 2.8787333573227783e-06,
      "loss": 0.0001,
      "step": 2740
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 11996.0771484375,
      "learning_rate": 2.1590500179920836e-06,
      "loss": 0.0002,
      "step": 2750
    },
    {
      "epoch": 6.952141057934509,
      "grad_norm": 2.2435834407806396,
      "learning_rate": 1.4393666786613891e-06,
      "loss": 0.0001,
      "step": 2760
    },
    {
      "epoch": 6.977329974811083,
      "grad_norm": 2109.901123046875,
      "learning_rate": 7.196833393306946e-07,
      "loss": 0.0832,
      "step": 2770
    },
    {
      "epoch": 7.0,
      "step": 2779,
      "total_flos": 2.5585840915697664e+18,
      "train_loss": 0.4836694428009911,
      "train_runtime": 1538.4023,
      "train_samples_per_second": 28.894,
      "train_steps_per_second": 1.806
    }
  ],
  "logging_steps": 10,
  "max_steps": 2779,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.5585840915697664e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}