{
  "best_metric": 0.8602029312288614,
  "best_model_checkpoint": "swinv2-small-patch4-window16-256-finetuned-galaxy10-decals/checkpoint-937",
  "epoch": 29.76,
  "eval_steps": 500,
  "global_step": 1860,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "learning_rate": 2.688172043010753e-06,
      "loss": 2.377,
      "step": 10
    },
    {
      "epoch": 0.32,
      "learning_rate": 5.376344086021506e-06,
      "loss": 2.2992,
      "step": 20
    },
    {
      "epoch": 0.48,
      "learning_rate": 8.064516129032258e-06,
      "loss": 2.188,
      "step": 30
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.0752688172043012e-05,
      "loss": 2.0729,
      "step": 40
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.3440860215053763e-05,
      "loss": 1.8938,
      "step": 50
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.6129032258064517e-05,
      "loss": 1.6168,
      "step": 60
    },
    {
      "epoch": 0.99,
      "eval_accuracy": 0.5005636978579482,
      "eval_f1": 0.4598783396512017,
      "eval_loss": 1.3397349119186401,
      "eval_precision": 0.4880433458740018,
      "eval_recall": 0.5005636978579482,
      "eval_runtime": 13.4986,
      "eval_samples_per_second": 131.421,
      "eval_steps_per_second": 2.074,
      "step": 62
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.881720430107527e-05,
      "loss": 1.4689,
      "step": 70
    },
    {
      "epoch": 1.28,
      "learning_rate": 2.1505376344086024e-05,
      "loss": 1.34,
      "step": 80
    },
    {
      "epoch": 1.44,
      "learning_rate": 2.4193548387096777e-05,
      "loss": 1.1924,
      "step": 90
    },
    {
      "epoch": 1.6,
      "learning_rate": 2.6881720430107527e-05,
      "loss": 1.0898,
      "step": 100
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.9569892473118284e-05,
      "loss": 0.9966,
      "step": 110
    },
    {
      "epoch": 1.92,
      "learning_rate": 3.2258064516129034e-05,
      "loss": 0.9396,
      "step": 120
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7463359639233371,
      "eval_f1": 0.7410407780959335,
      "eval_loss": 0.7823030948638916,
      "eval_precision": 0.7601555691928246,
      "eval_recall": 0.7463359639233371,
      "eval_runtime": 13.3574,
      "eval_samples_per_second": 132.81,
      "eval_steps_per_second": 2.096,
      "step": 125
    },
    {
      "epoch": 2.08,
      "learning_rate": 3.494623655913979e-05,
      "loss": 0.9146,
      "step": 130
    },
    {
      "epoch": 2.24,
      "learning_rate": 3.763440860215054e-05,
      "loss": 0.8752,
      "step": 140
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.032258064516129e-05,
      "loss": 0.8463,
      "step": 150
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.301075268817205e-05,
      "loss": 0.8719,
      "step": 160
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.56989247311828e-05,
      "loss": 0.8057,
      "step": 170
    },
    {
      "epoch": 2.88,
      "learning_rate": 4.8387096774193554e-05,
      "loss": 0.782,
      "step": 180
    },
    {
      "epoch": 2.99,
      "eval_accuracy": 0.7948139797068771,
      "eval_f1": 0.7884779801709714,
      "eval_loss": 0.5995017290115356,
      "eval_precision": 0.7936891250581956,
      "eval_recall": 0.7948139797068771,
      "eval_runtime": 13.5734,
      "eval_samples_per_second": 130.697,
      "eval_steps_per_second": 2.063,
      "step": 187
    },
    {
      "epoch": 3.04,
      "learning_rate": 4.98805256869773e-05,
      "loss": 0.8,
      "step": 190
    },
    {
      "epoch": 3.2,
      "learning_rate": 4.9581839904420555e-05,
      "loss": 0.7576,
      "step": 200
    },
    {
      "epoch": 3.36,
      "learning_rate": 4.92831541218638e-05,
      "loss": 0.7308,
      "step": 210
    },
    {
      "epoch": 3.52,
      "learning_rate": 4.898446833930705e-05,
      "loss": 0.7454,
      "step": 220
    },
    {
      "epoch": 3.68,
      "learning_rate": 4.86857825567503e-05,
      "loss": 0.6737,
      "step": 230
    },
    {
      "epoch": 3.84,
      "learning_rate": 4.8387096774193554e-05,
      "loss": 0.6978,
      "step": 240
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.80884109916368e-05,
      "loss": 0.6373,
      "step": 250
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8229988726042841,
      "eval_f1": 0.81759282913123,
      "eval_loss": 0.5226621031761169,
      "eval_precision": 0.8192338763552195,
      "eval_recall": 0.8229988726042841,
      "eval_runtime": 13.2363,
      "eval_samples_per_second": 134.025,
      "eval_steps_per_second": 2.115,
      "step": 250
    },
    {
      "epoch": 4.16,
      "learning_rate": 4.778972520908005e-05,
      "loss": 0.6896,
      "step": 260
    },
    {
      "epoch": 4.32,
      "learning_rate": 4.74910394265233e-05,
      "loss": 0.6573,
      "step": 270
    },
    {
      "epoch": 4.48,
      "learning_rate": 4.7192353643966546e-05,
      "loss": 0.6664,
      "step": 280
    },
    {
      "epoch": 4.64,
      "learning_rate": 4.6893667861409805e-05,
      "loss": 0.6308,
      "step": 290
    },
    {
      "epoch": 4.8,
      "learning_rate": 4.659498207885305e-05,
      "loss": 0.6681,
      "step": 300
    },
    {
      "epoch": 4.96,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.6047,
      "step": 310
    },
    {
      "epoch": 4.99,
      "eval_accuracy": 0.8280721533258174,
      "eval_f1": 0.8262057895473833,
      "eval_loss": 0.5238057971000671,
      "eval_precision": 0.8271529816878367,
      "eval_recall": 0.8280721533258174,
      "eval_runtime": 13.1159,
      "eval_samples_per_second": 135.255,
      "eval_steps_per_second": 2.135,
      "step": 312
    },
    {
      "epoch": 5.12,
      "learning_rate": 4.5997610513739546e-05,
      "loss": 0.6046,
      "step": 320
    },
    {
      "epoch": 5.28,
      "learning_rate": 4.56989247311828e-05,
      "loss": 0.6297,
      "step": 330
    },
    {
      "epoch": 5.44,
      "learning_rate": 4.540023894862604e-05,
      "loss": 0.6256,
      "step": 340
    },
    {
      "epoch": 5.6,
      "learning_rate": 4.51015531660693e-05,
      "loss": 0.6005,
      "step": 350
    },
    {
      "epoch": 5.76,
      "learning_rate": 4.4802867383512545e-05,
      "loss": 0.59,
      "step": 360
    },
    {
      "epoch": 5.92,
      "learning_rate": 4.4504181600955796e-05,
      "loss": 0.6143,
      "step": 370
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8348365276211951,
      "eval_f1": 0.8297590620731212,
      "eval_loss": 0.5090616345405579,
      "eval_precision": 0.842878443292543,
      "eval_recall": 0.8348365276211951,
      "eval_runtime": 13.7594,
      "eval_samples_per_second": 128.93,
      "eval_steps_per_second": 2.035,
      "step": 375
    },
    {
      "epoch": 6.08,
      "learning_rate": 4.420549581839905e-05,
      "loss": 0.5794,
      "step": 380
    },
    {
      "epoch": 6.24,
      "learning_rate": 4.390681003584229e-05,
      "loss": 0.6023,
      "step": 390
    },
    {
      "epoch": 6.4,
      "learning_rate": 4.360812425328555e-05,
      "loss": 0.5718,
      "step": 400
    },
    {
      "epoch": 6.56,
      "learning_rate": 4.3309438470728796e-05,
      "loss": 0.5889,
      "step": 410
    },
    {
      "epoch": 6.72,
      "learning_rate": 4.301075268817205e-05,
      "loss": 0.5652,
      "step": 420
    },
    {
      "epoch": 6.88,
      "learning_rate": 4.271206690561529e-05,
      "loss": 0.5805,
      "step": 430
    },
    {
      "epoch": 6.99,
      "eval_accuracy": 0.826381059751973,
      "eval_f1": 0.8253882697893267,
      "eval_loss": 0.4920646846294403,
      "eval_precision": 0.8275287235098401,
      "eval_recall": 0.826381059751973,
      "eval_runtime": 13.4404,
      "eval_samples_per_second": 131.99,
      "eval_steps_per_second": 2.083,
      "step": 437
    },
    {
      "epoch": 7.04,
      "learning_rate": 4.241338112305854e-05,
      "loss": 0.5902,
      "step": 440
    },
    {
      "epoch": 7.2,
      "learning_rate": 4.2114695340501795e-05,
      "loss": 0.5981,
      "step": 450
    },
    {
      "epoch": 7.36,
      "learning_rate": 4.1816009557945046e-05,
      "loss": 0.529,
      "step": 460
    },
    {
      "epoch": 7.52,
      "learning_rate": 4.15173237753883e-05,
      "loss": 0.5686,
      "step": 470
    },
    {
      "epoch": 7.68,
      "learning_rate": 4.121863799283154e-05,
      "loss": 0.5523,
      "step": 480
    },
    {
      "epoch": 7.84,
      "learning_rate": 4.0919952210274794e-05,
      "loss": 0.5366,
      "step": 490
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.062126642771804e-05,
      "loss": 0.5476,
      "step": 500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8320180383314544,
      "eval_f1": 0.8290681706833106,
      "eval_loss": 0.48320192098617554,
      "eval_precision": 0.8409394102308216,
      "eval_recall": 0.8320180383314544,
      "eval_runtime": 13.465,
      "eval_samples_per_second": 131.749,
      "eval_steps_per_second": 2.079,
      "step": 500
    },
    {
      "epoch": 8.16,
      "learning_rate": 4.032258064516129e-05,
      "loss": 0.516,
      "step": 510
    },
    {
      "epoch": 8.32,
      "learning_rate": 4.002389486260454e-05,
      "loss": 0.5298,
      "step": 520
    },
    {
      "epoch": 8.48,
      "learning_rate": 3.972520908004779e-05,
      "loss": 0.5817,
      "step": 530
    },
    {
      "epoch": 8.64,
      "learning_rate": 3.9426523297491045e-05,
      "loss": 0.5451,
      "step": 540
    },
    {
      "epoch": 8.8,
      "learning_rate": 3.912783751493429e-05,
      "loss": 0.5621,
      "step": 550
    },
    {
      "epoch": 8.96,
      "learning_rate": 3.882915173237754e-05,
      "loss": 0.5333,
      "step": 560
    },
    {
      "epoch": 8.99,
      "eval_accuracy": 0.8500563697857948,
      "eval_f1": 0.8477229440479351,
      "eval_loss": 0.4456086754798889,
      "eval_precision": 0.8500029957213829,
      "eval_recall": 0.8500563697857948,
      "eval_runtime": 13.7729,
      "eval_samples_per_second": 128.804,
      "eval_steps_per_second": 2.033,
      "step": 562
    },
    {
      "epoch": 9.12,
      "learning_rate": 3.8530465949820786e-05,
      "loss": 0.522,
      "step": 570
    },
    {
      "epoch": 9.28,
      "learning_rate": 3.8231780167264044e-05,
      "loss": 0.5057,
      "step": 580
    },
    {
      "epoch": 9.44,
      "learning_rate": 3.793309438470729e-05,
      "loss": 0.516,
      "step": 590
    },
    {
      "epoch": 9.6,
      "learning_rate": 3.763440860215054e-05,
      "loss": 0.518,
      "step": 600
    },
    {
      "epoch": 9.76,
      "learning_rate": 3.7335722819593785e-05,
      "loss": 0.4821,
      "step": 610
    },
    {
      "epoch": 9.92,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.5062,
      "step": 620
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.846674182638106,
      "eval_f1": 0.8457044095560543,
      "eval_loss": 0.44933807849884033,
      "eval_precision": 0.848013601343001,
      "eval_recall": 0.846674182638106,
      "eval_runtime": 13.5051,
      "eval_samples_per_second": 131.358,
      "eval_steps_per_second": 2.073,
      "step": 625
    },
    {
      "epoch": 10.08,
      "learning_rate": 3.673835125448029e-05,
      "loss": 0.5048,
      "step": 630
    },
    {
      "epoch": 10.24,
      "learning_rate": 3.643966547192354e-05,
      "loss": 0.503,
      "step": 640
    },
    {
      "epoch": 10.4,
      "learning_rate": 3.614097968936679e-05,
      "loss": 0.5178,
      "step": 650
    },
    {
      "epoch": 10.56,
      "learning_rate": 3.5842293906810036e-05,
      "loss": 0.4983,
      "step": 660
    },
    {
      "epoch": 10.72,
      "learning_rate": 3.554360812425329e-05,
      "loss": 0.4925,
      "step": 670
    },
    {
      "epoch": 10.88,
      "learning_rate": 3.524492234169653e-05,
      "loss": 0.5001,
      "step": 680
    },
    {
      "epoch": 10.99,
      "eval_accuracy": 0.8449830890642616,
      "eval_f1": 0.8448852162243315,
      "eval_loss": 0.46166303753852844,
      "eval_precision": 0.8467862659645279,
      "eval_recall": 0.8449830890642616,
      "eval_runtime": 12.895,
      "eval_samples_per_second": 137.572,
      "eval_steps_per_second": 2.171,
      "step": 687
    },
    {
      "epoch": 11.04,
      "learning_rate": 3.494623655913979e-05,
      "loss": 0.467,
      "step": 690
    },
    {
      "epoch": 11.2,
      "learning_rate": 3.4647550776583035e-05,
      "loss": 0.454,
      "step": 700
    },
    {
      "epoch": 11.36,
      "learning_rate": 3.4348864994026287e-05,
      "loss": 0.4626,
      "step": 710
    },
    {
      "epoch": 11.52,
      "learning_rate": 3.405017921146954e-05,
      "loss": 0.4641,
      "step": 720
    },
    {
      "epoch": 11.68,
      "learning_rate": 3.375149342891278e-05,
      "loss": 0.5068,
      "step": 730
    },
    {
      "epoch": 11.84,
      "learning_rate": 3.3452807646356034e-05,
      "loss": 0.4885,
      "step": 740
    },
    {
      "epoch": 12.0,
      "learning_rate": 3.3154121863799286e-05,
      "loss": 0.4572,
      "step": 750
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.846674182638106,
      "eval_f1": 0.8448982276240636,
      "eval_loss": 0.4496840536594391,
      "eval_precision": 0.8449696394145345,
      "eval_recall": 0.846674182638106,
      "eval_runtime": 13.2081,
      "eval_samples_per_second": 134.312,
      "eval_steps_per_second": 2.12,
      "step": 750
    },
    {
      "epoch": 12.16,
      "learning_rate": 3.285543608124254e-05,
      "loss": 0.473,
      "step": 760
    },
    {
      "epoch": 12.32,
      "learning_rate": 3.255675029868578e-05,
      "loss": 0.453,
      "step": 770
    },
    {
      "epoch": 12.48,
      "learning_rate": 3.2258064516129034e-05,
      "loss": 0.4644,
      "step": 780
    },
    {
      "epoch": 12.64,
      "learning_rate": 3.195937873357228e-05,
      "loss": 0.481,
      "step": 790
    },
    {
      "epoch": 12.8,
      "learning_rate": 3.1660692951015537e-05,
      "loss": 0.4653,
      "step": 800
    },
    {
      "epoch": 12.96,
      "learning_rate": 3.136200716845878e-05,
      "loss": 0.4681,
      "step": 810
    },
    {
      "epoch": 12.99,
      "eval_accuracy": 0.8489289740698985,
      "eval_f1": 0.8452026182581576,
      "eval_loss": 0.45876094698905945,
      "eval_precision": 0.8486312927031242,
      "eval_recall": 0.8489289740698985,
      "eval_runtime": 13.4808,
      "eval_samples_per_second": 131.594,
      "eval_steps_per_second": 2.077,
      "step": 812
    },
    {
      "epoch": 13.12,
      "learning_rate": 3.106332138590203e-05,
      "loss": 0.4385,
      "step": 820
    },
    {
      "epoch": 13.28,
      "learning_rate": 3.0764635603345284e-05,
      "loss": 0.4493,
      "step": 830
    },
    {
      "epoch": 13.44,
      "learning_rate": 3.046594982078853e-05,
      "loss": 0.4853,
      "step": 840
    },
    {
      "epoch": 13.6,
      "learning_rate": 3.016726403823178e-05,
      "loss": 0.4415,
      "step": 850
    },
    {
      "epoch": 13.76,
      "learning_rate": 2.9868578255675032e-05,
      "loss": 0.4727,
      "step": 860
    },
    {
      "epoch": 13.92,
      "learning_rate": 2.9569892473118284e-05,
      "loss": 0.4747,
      "step": 870
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.8528748590755355,
      "eval_f1": 0.8507531520007248,
      "eval_loss": 0.4280567765235901,
      "eval_precision": 0.8554045998853204,
      "eval_recall": 0.8528748590755355,
      "eval_runtime": 13.3113,
      "eval_samples_per_second": 133.27,
      "eval_steps_per_second": 2.103,
      "step": 875
    },
    {
      "epoch": 14.08,
      "learning_rate": 2.9271206690561532e-05,
      "loss": 0.4514,
      "step": 880
    },
    {
      "epoch": 14.24,
      "learning_rate": 2.897252090800478e-05,
      "loss": 0.4121,
      "step": 890
    },
    {
      "epoch": 14.4,
      "learning_rate": 2.8673835125448028e-05,
      "loss": 0.4297,
      "step": 900
    },
    {
      "epoch": 14.56,
      "learning_rate": 2.8375149342891276e-05,
      "loss": 0.4597,
      "step": 910
    },
    {
      "epoch": 14.72,
      "learning_rate": 2.807646356033453e-05,
      "loss": 0.4351,
      "step": 920
    },
    {
      "epoch": 14.88,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.4283,
      "step": 930
    },
    {
      "epoch": 14.99,
      "eval_accuracy": 0.8602029312288614,
      "eval_f1": 0.8584694828211913,
      "eval_loss": 0.4405820965766907,
      "eval_precision": 0.857692461827972,
      "eval_recall": 0.8602029312288614,
      "eval_runtime": 13.0636,
      "eval_samples_per_second": 135.797,
      "eval_steps_per_second": 2.143,
      "step": 937
    },
    {
      "epoch": 15.04,
      "learning_rate": 2.747909199522103e-05,
      "loss": 0.4408,
      "step": 940
    },
    {
      "epoch": 15.2,
      "learning_rate": 2.718040621266428e-05,
      "loss": 0.4163,
      "step": 950
    },
    {
      "epoch": 15.36,
      "learning_rate": 2.6881720430107527e-05,
      "loss": 0.433,
      "step": 960
    },
    {
      "epoch": 15.52,
      "learning_rate": 2.6583034647550775e-05,
      "loss": 0.4437,
      "step": 970
    },
    {
      "epoch": 15.68,
      "learning_rate": 2.628434886499403e-05,
      "loss": 0.4072,
      "step": 980
    },
    {
      "epoch": 15.84,
      "learning_rate": 2.5985663082437278e-05,
      "loss": 0.4331,
      "step": 990
    },
    {
      "epoch": 16.0,
      "learning_rate": 2.5686977299880526e-05,
      "loss": 0.4296,
      "step": 1000
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.8534385569334837,
      "eval_f1": 0.8497703469593414,
      "eval_loss": 0.44582441449165344,
      "eval_precision": 0.8511530207566267,
      "eval_recall": 0.8534385569334837,
      "eval_runtime": 13.7627,
      "eval_samples_per_second": 128.899,
      "eval_steps_per_second": 2.034,
      "step": 1000
    },
    {
      "epoch": 16.16,
      "learning_rate": 2.5388291517323774e-05,
      "loss": 0.4261,
      "step": 1010
    },
    {
      "epoch": 16.32,
      "learning_rate": 2.5089605734767026e-05,
      "loss": 0.3933,
      "step": 1020
    },
    {
      "epoch": 16.48,
      "learning_rate": 2.4790919952210277e-05,
      "loss": 0.4239,
      "step": 1030
    },
    {
      "epoch": 16.64,
      "learning_rate": 2.4492234169653525e-05,
      "loss": 0.436,
      "step": 1040
    },
    {
      "epoch": 16.8,
      "learning_rate": 2.4193548387096777e-05,
      "loss": 0.4149,
      "step": 1050
    },
    {
      "epoch": 16.96,
      "learning_rate": 2.3894862604540025e-05,
      "loss": 0.3734,
      "step": 1060
    },
    {
      "epoch": 16.99,
      "eval_accuracy": 0.8416009019165727,
      "eval_f1": 0.8386440673113804,
      "eval_loss": 0.46230921149253845,
      "eval_precision": 0.8419435509719339,
      "eval_recall": 0.8416009019165727,
      "eval_runtime": 13.6631,
      "eval_samples_per_second": 129.839,
      "eval_steps_per_second": 2.049,
      "step": 1062
    },
    {
      "epoch": 17.12,
      "learning_rate": 2.3596176821983273e-05,
      "loss": 0.42,
      "step": 1070
    },
    {
      "epoch": 17.28,
      "learning_rate": 2.3297491039426525e-05,
      "loss": 0.4075,
      "step": 1080
    },
    {
      "epoch": 17.44,
      "learning_rate": 2.2998805256869773e-05,
      "loss": 0.4467,
      "step": 1090
    },
    {
      "epoch": 17.6,
      "learning_rate": 2.270011947431302e-05,
      "loss": 0.4213,
      "step": 1100
    },
    {
      "epoch": 17.76,
      "learning_rate": 2.2401433691756272e-05,
      "loss": 0.4102,
      "step": 1110
    },
    {
      "epoch": 17.92,
      "learning_rate": 2.2102747909199524e-05,
      "loss": 0.3921,
      "step": 1120
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.8517474633596392,
      "eval_f1": 0.8495803094909621,
      "eval_loss": 0.4437521696090698,
      "eval_precision": 0.850579454526507,
      "eval_recall": 0.8517474633596392,
      "eval_runtime": 13.5077,
      "eval_samples_per_second": 131.333,
      "eval_steps_per_second": 2.073,
      "step": 1125
    },
    {
      "epoch": 18.08,
      "learning_rate": 2.1804062126642775e-05,
      "loss": 0.4117,
      "step": 1130
    },
    {
      "epoch": 18.24,
      "learning_rate": 2.1505376344086024e-05,
      "loss": 0.384,
      "step": 1140
    },
    {
      "epoch": 18.4,
      "learning_rate": 2.120669056152927e-05,
      "loss": 0.3907,
      "step": 1150
    },
    {
      "epoch": 18.56,
      "learning_rate": 2.0908004778972523e-05,
      "loss": 0.3935,
      "step": 1160
    },
    {
      "epoch": 18.72,
      "learning_rate": 2.060931899641577e-05,
      "loss": 0.4008,
      "step": 1170
    },
    {
      "epoch": 18.88,
      "learning_rate": 2.031063321385902e-05,
      "loss": 0.3954,
      "step": 1180
    },
    {
      "epoch": 18.99,
      "eval_accuracy": 0.846674182638106,
      "eval_f1": 0.8445676498803157,
      "eval_loss": 0.47115737199783325,
      "eval_precision": 0.8486690377876364,
      "eval_recall": 0.846674182638106,
      "eval_runtime": 13.148,
      "eval_samples_per_second": 134.925,
      "eval_steps_per_second": 2.13,
      "step": 1187
    },
    {
      "epoch": 19.04,
      "learning_rate": 2.001194743130227e-05,
      "loss": 0.4024,
      "step": 1190
    },
    {
      "epoch": 19.2,
      "learning_rate": 1.9713261648745522e-05,
      "loss": 0.3845,
      "step": 1200
    },
    {
      "epoch": 19.36,
      "learning_rate": 1.941457586618877e-05,
      "loss": 0.3949,
      "step": 1210
    },
    {
      "epoch": 19.52,
      "learning_rate": 1.9115890083632022e-05,
      "loss": 0.3905,
      "step": 1220
    },
    {
      "epoch": 19.68,
      "learning_rate": 1.881720430107527e-05,
      "loss": 0.364,
      "step": 1230
    },
    {
      "epoch": 19.84,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.3898,
      "step": 1240
    },
    {
      "epoch": 20.0,
      "learning_rate": 1.821983273596177e-05,
      "loss": 0.3995,
      "step": 1250
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.8483652762119503,
      "eval_f1": 0.8448086716206932,
      "eval_loss": 0.4647820293903351,
      "eval_precision": 0.8467330017782122,
      "eval_recall": 0.8483652762119503,
      "eval_runtime": 12.9665,
      "eval_samples_per_second": 136.815,
      "eval_steps_per_second": 2.159,
      "step": 1250
    },
    {
      "epoch": 20.16,
      "learning_rate": 1.7921146953405018e-05,
      "loss": 0.3894,
      "step": 1260
    },
    {
      "epoch": 20.32,
      "learning_rate": 1.7622461170848266e-05,
      "loss": 0.4036,
      "step": 1270
    },
    {
      "epoch": 20.48,
      "learning_rate": 1.7323775388291518e-05,
      "loss": 0.3695,
      "step": 1280
    },
    {
      "epoch": 20.64,
      "learning_rate": 1.702508960573477e-05,
      "loss": 0.368,
      "step": 1290
    },
    {
      "epoch": 20.8,
      "learning_rate": 1.6726403823178017e-05,
      "loss": 0.3759,
      "step": 1300
    },
    {
      "epoch": 20.96,
      "learning_rate": 1.642771804062127e-05,
      "loss": 0.3859,
      "step": 1310
    },
    {
      "epoch": 20.99,
      "eval_accuracy": 0.8494926719278467,
      "eval_f1": 0.8461868473998411,
      "eval_loss": 0.47284412384033203,
      "eval_precision": 0.848705583367713,
      "eval_recall": 0.8494926719278467,
      "eval_runtime": 13.284,
      "eval_samples_per_second": 133.544,
      "eval_steps_per_second": 2.108,
      "step": 1312
    },
    {
      "epoch": 21.12,
      "learning_rate": 1.6129032258064517e-05,
      "loss": 0.368,
      "step": 1320
    },
    {
      "epoch": 21.28,
      "learning_rate": 1.5830346475507768e-05,
      "loss": 0.3859,
      "step": 1330
    },
    {
      "epoch": 21.44,
      "learning_rate": 1.5531660692951016e-05,
      "loss": 0.3694,
      "step": 1340
    },
    {
      "epoch": 21.6,
      "learning_rate": 1.5232974910394265e-05,
      "loss": 0.3804,
      "step": 1350
    },
    {
      "epoch": 21.76,
      "learning_rate": 1.4934289127837516e-05,
      "loss": 0.3796,
      "step": 1360
    },
    {
      "epoch": 21.92,
      "learning_rate": 1.4635603345280766e-05,
      "loss": 0.4046,
      "step": 1370
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.8472378804960541,
      "eval_f1": 0.8452686187473546,
      "eval_loss": 0.4719749093055725,
      "eval_precision": 0.8466689832080643,
      "eval_recall": 0.8472378804960541,
      "eval_runtime": 13.6675,
      "eval_samples_per_second": 129.797,
      "eval_steps_per_second": 2.049,
      "step": 1375
    },
    {
      "epoch": 22.08,
      "learning_rate": 1.4336917562724014e-05,
      "loss": 0.351,
      "step": 1380
    },
    {
      "epoch": 22.24,
      "learning_rate": 1.4038231780167265e-05,
      "loss": 0.3412,
      "step": 1390
    },
    {
      "epoch": 22.4,
      "learning_rate": 1.3739545997610515e-05,
      "loss": 0.3684,
      "step": 1400
    },
    {
      "epoch": 22.56,
      "learning_rate": 1.3440860215053763e-05,
      "loss": 0.3688,
      "step": 1410
    },
    {
      "epoch": 22.72,
      "learning_rate": 1.3142174432497015e-05,
      "loss": 0.3723,
      "step": 1420
    },
    {
      "epoch": 22.88,
      "learning_rate": 1.2843488649940263e-05,
      "loss": 0.3651,
      "step": 1430
    },
    {
      "epoch": 22.99,
      "eval_accuracy": 0.8416009019165727,
      "eval_f1": 0.8395535353828757,
      "eval_loss": 0.48366203904151917,
      "eval_precision": 0.8408810842655634,
      "eval_recall": 0.8416009019165727,
      "eval_runtime": 13.7488,
      "eval_samples_per_second": 129.029,
      "eval_steps_per_second": 2.037,
      "step": 1437
    },
    {
      "epoch": 23.04,
      "learning_rate": 1.2544802867383513e-05,
      "loss": 0.3741,
      "step": 1440
    },
    {
      "epoch": 23.2,
      "learning_rate": 1.2246117084826763e-05,
      "loss": 0.36,
      "step": 1450
    },
    {
      "epoch": 23.36,
      "learning_rate": 1.1947431302270013e-05,
      "loss": 0.3561,
      "step": 1460
    },
    {
      "epoch": 23.52,
      "learning_rate": 1.1648745519713262e-05,
      "loss": 0.3561,
      "step": 1470
    },
    {
      "epoch": 23.68,
      "learning_rate": 1.135005973715651e-05,
      "loss": 0.3779,
      "step": 1480
    },
    {
      "epoch": 23.84,
      "learning_rate": 1.1051373954599762e-05,
      "loss": 0.3345,
      "step": 1490
    },
    {
      "epoch": 24.0,
      "learning_rate": 1.0752688172043012e-05,
      "loss": 0.3481,
      "step": 1500
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8540022547914318,
      "eval_f1": 0.8524362037582941,
      "eval_loss": 0.47422128915786743,
      "eval_precision": 0.8522498451390563,
      "eval_recall": 0.8540022547914318,
      "eval_runtime": 13.0453,
      "eval_samples_per_second": 135.988,
      "eval_steps_per_second": 2.146,
      "step": 1500
    },
    {
      "epoch": 24.16,
      "learning_rate": 1.0454002389486262e-05,
      "loss": 0.3735,
      "step": 1510
    },
    {
      "epoch": 24.32,
      "learning_rate": 1.015531660692951e-05,
      "loss": 0.3815,
      "step": 1520
    },
    {
      "epoch": 24.48,
      "learning_rate": 9.856630824372761e-06,
      "loss": 0.3541,
      "step": 1530
    },
    {
      "epoch": 24.64,
      "learning_rate": 9.557945041816011e-06,
      "loss": 0.3483,
      "step": 1540
    },
    {
      "epoch": 24.8,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.3491,
      "step": 1550
    },
    {
      "epoch": 24.96,
      "learning_rate": 8.960573476702509e-06,
      "loss": 0.3706,
      "step": 1560
    },
    {
      "epoch": 24.99,
      "eval_accuracy": 0.8478015783540023,
      "eval_f1": 0.845457124690547,
      "eval_loss": 0.48457956314086914,
      "eval_precision": 0.8477288496990673,
      "eval_recall": 0.8478015783540023,
      "eval_runtime": 13.5806,
      "eval_samples_per_second": 130.628,
      "eval_steps_per_second": 2.062,
      "step": 1562
    },
    {
      "epoch": 25.12,
      "learning_rate": 8.661887694145759e-06,
      "loss": 0.3271,
      "step": 1570
    },
    {
      "epoch": 25.28,
      "learning_rate": 8.363201911589009e-06,
      "loss": 0.3501,
      "step": 1580
    },
    {
      "epoch": 25.44,
      "learning_rate": 8.064516129032258e-06,
      "loss": 0.3459,
      "step": 1590
    },
    {
      "epoch": 25.6,
      "learning_rate": 7.765830346475508e-06,
      "loss": 0.3628,
      "step": 1600
    },
    {
      "epoch": 25.76,
      "learning_rate": 7.467144563918758e-06,
      "loss": 0.3368,
      "step": 1610
    },
    {
      "epoch": 25.92,
      "learning_rate": 7.168458781362007e-06,
      "loss": 0.3278,
      "step": 1620
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.850620067643743,
      "eval_f1": 0.8484355776067231,
      "eval_loss": 0.4798063635826111,
      "eval_precision": 0.8502366126091575,
      "eval_recall": 0.850620067643743,
      "eval_runtime": 13.119,
      "eval_samples_per_second": 135.224,
      "eval_steps_per_second": 2.134,
      "step": 1625
    },
    {
      "epoch": 26.08,
      "learning_rate": 6.869772998805258e-06,
      "loss": 0.3462,
      "step": 1630
    },
    {
      "epoch": 26.24,
      "learning_rate": 6.5710872162485075e-06,
      "loss": 0.3525,
      "step": 1640
    },
    {
      "epoch": 26.4,
      "learning_rate": 6.2724014336917564e-06,
      "loss": 0.3399,
      "step": 1650
    },
    {
      "epoch": 26.56,
      "learning_rate": 5.973715651135006e-06,
      "loss": 0.3261,
      "step": 1660
    },
    {
      "epoch": 26.72,
      "learning_rate": 5.675029868578255e-06,
      "loss": 0.3573,
      "step": 1670
    },
    {
      "epoch": 26.88,
      "learning_rate": 5.376344086021506e-06,
      "loss": 0.3484,
      "step": 1680
    },
    {
      "epoch": 26.99,
      "eval_accuracy": 0.8528748590755355,
      "eval_f1": 0.8520109294677986,
      "eval_loss": 0.46748560667037964,
      "eval_precision": 0.8538422987628977,
      "eval_recall": 0.8528748590755355,
      "eval_runtime": 13.8896,
      "eval_samples_per_second": 127.721,
      "eval_steps_per_second": 2.016,
      "step": 1687
    },
    {
      "epoch": 27.04,
      "learning_rate": 5.077658303464755e-06,
      "loss": 0.3456,
      "step": 1690
    },
    {
      "epoch": 27.2,
      "learning_rate": 4.7789725209080055e-06,
      "loss": 0.3374,
      "step": 1700
    },
    {
      "epoch": 27.36,
      "learning_rate": 4.4802867383512545e-06,
      "loss": 0.3692,
      "step": 1710
    },
    {
      "epoch": 27.52,
      "learning_rate": 4.181600955794504e-06,
      "loss": 0.3299,
      "step": 1720
    },
    {
      "epoch": 27.68,
      "learning_rate": 3.882915173237754e-06,
      "loss": 0.3214,
      "step": 1730
    },
    {
      "epoch": 27.84,
      "learning_rate": 3.5842293906810035e-06,
      "loss": 0.3433,
      "step": 1740
    },
    {
      "epoch": 28.0,
      "learning_rate": 3.2855436081242537e-06,
      "loss": 0.3626,
      "step": 1750
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8449830890642616,
      "eval_f1": 0.842919257929613,
      "eval_loss": 0.47684696316719055,
      "eval_precision": 0.8445947883906417,
      "eval_recall": 0.8449830890642616,
      "eval_runtime": 13.5898,
      "eval_samples_per_second": 130.539,
      "eval_steps_per_second": 2.06,
      "step": 1750
    },
    {
      "epoch": 28.16,
      "learning_rate": 2.986857825567503e-06,
      "loss": 0.3398,
      "step": 1760
    },
    {
      "epoch": 28.32,
      "learning_rate": 2.688172043010753e-06,
      "loss": 0.3417,
      "step": 1770
    },
    {
      "epoch": 28.48,
      "learning_rate": 2.3894862604540028e-06,
      "loss": 0.3317,
      "step": 1780
    },
    {
      "epoch": 28.64,
      "learning_rate": 2.090800477897252e-06,
      "loss": 0.3435,
      "step": 1790
    },
    {
      "epoch": 28.8,
      "learning_rate": 1.7921146953405017e-06,
      "loss": 0.3392,
      "step": 1800
    },
    {
      "epoch": 28.96,
      "learning_rate": 1.4934289127837516e-06,
      "loss": 0.3324,
      "step": 1810
    },
    {
      "epoch": 28.99,
      "eval_accuracy": 0.8483652762119503,
      "eval_f1": 0.8459600842032119,
      "eval_loss": 0.4725462794303894,
      "eval_precision": 0.8469506532798351,
      "eval_recall": 0.8483652762119503,
      "eval_runtime": 13.1087,
      "eval_samples_per_second": 135.33,
      "eval_steps_per_second": 2.136,
      "step": 1812
    },
    {
      "epoch": 29.12,
      "learning_rate": 1.1947431302270014e-06,
      "loss": 0.3447,
      "step": 1820
    },
    {
      "epoch": 29.28,
      "learning_rate": 8.960573476702509e-07,
      "loss": 0.3229,
      "step": 1830
    },
    {
      "epoch": 29.44,
      "learning_rate": 5.973715651135007e-07,
      "loss": 0.3269,
      "step": 1840
    },
    {
      "epoch": 29.6,
      "learning_rate": 2.9868578255675034e-07,
      "loss": 0.3277,
      "step": 1850
    },
    {
      "epoch": 29.76,
      "learning_rate": 0.0,
      "loss": 0.3462,
      "step": 1860
    },
    {
      "epoch": 29.76,
      "eval_accuracy": 0.8489289740698985,
      "eval_f1": 0.8472236091345323,
      "eval_loss": 0.47372013330459595,
      "eval_precision": 0.848634752763623,
      "eval_recall": 0.8489289740698985,
      "eval_runtime": 13.329,
      "eval_samples_per_second": 133.093,
      "eval_steps_per_second": 2.101,
      "step": 1860
    },
    {
      "epoch": 29.76,
      "step": 1860,
      "total_flos": 2.7441227042701443e+19,
      "train_loss": 0.5384533834713762,
      "train_runtime": 6912.8232,
      "train_samples_per_second": 69.271,
      "train_steps_per_second": 0.269
    }
  ],
  "logging_steps": 10,
  "max_steps": 1860,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 2.7441227042701443e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}