{
  "best_metric": 0.8737316798196166,
  "best_model_checkpoint": "dinov2-large-finetuned-galaxy10-decals/checkpoint-1860",
  "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": 2.7977,
      "step": 10
    },
    {
      "epoch": 0.32,
      "learning_rate": 5.376344086021506e-06,
      "loss": 1.8679,
      "step": 20
    },
    {
      "epoch": 0.48,
      "learning_rate": 8.064516129032258e-06,
      "loss": 1.3021,
      "step": 30
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.0752688172043012e-05,
      "loss": 1.004,
      "step": 40
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.3440860215053763e-05,
      "loss": 0.9077,
      "step": 50
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.6129032258064517e-05,
      "loss": 0.7564,
      "step": 60
    },
    {
      "epoch": 0.99,
      "eval_accuracy": 0.7976324689966178,
      "eval_f1": 0.79902483317454,
      "eval_loss": 0.6187198758125305,
      "eval_precision": 0.8170887985737173,
      "eval_recall": 0.7976324689966178,
      "eval_runtime": 24.6164,
      "eval_samples_per_second": 72.066,
      "eval_steps_per_second": 1.137,
      "step": 62
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.881720430107527e-05,
      "loss": 0.8046,
      "step": 70
    },
    {
      "epoch": 1.28,
      "learning_rate": 2.1505376344086024e-05,
      "loss": 0.7356,
      "step": 80
    },
    {
      "epoch": 1.44,
      "learning_rate": 2.4193548387096777e-05,
      "loss": 0.7607,
      "step": 90
    },
    {
      "epoch": 1.6,
      "learning_rate": 2.6881720430107527e-05,
      "loss": 0.755,
      "step": 100
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.9569892473118284e-05,
      "loss": 0.6772,
      "step": 110
    },
    {
      "epoch": 1.92,
      "learning_rate": 3.2258064516129034e-05,
      "loss": 0.7766,
      "step": 120
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7852311161217588,
      "eval_f1": 0.7781753717594276,
      "eval_loss": 0.61021488904953,
      "eval_precision": 0.8052476467823985,
      "eval_recall": 0.7852311161217588,
      "eval_runtime": 22.5165,
      "eval_samples_per_second": 78.787,
      "eval_steps_per_second": 1.244,
      "step": 125
    },
    {
      "epoch": 2.08,
      "learning_rate": 3.494623655913979e-05,
      "loss": 0.7641,
      "step": 130
    },
    {
      "epoch": 2.24,
      "learning_rate": 3.763440860215054e-05,
      "loss": 0.721,
      "step": 140
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.032258064516129e-05,
      "loss": 0.7,
      "step": 150
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.301075268817205e-05,
      "loss": 0.7117,
      "step": 160
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.56989247311828e-05,
      "loss": 0.8347,
      "step": 170
    },
    {
      "epoch": 2.88,
      "learning_rate": 4.8387096774193554e-05,
      "loss": 0.7103,
      "step": 180
    },
    {
      "epoch": 2.99,
      "eval_accuracy": 0.8089064261555806,
      "eval_f1": 0.803152455190108,
      "eval_loss": 0.5743524432182312,
      "eval_precision": 0.8140192681408788,
      "eval_recall": 0.8089064261555806,
      "eval_runtime": 22.4141,
      "eval_samples_per_second": 79.147,
      "eval_steps_per_second": 1.249,
      "step": 187
    },
    {
      "epoch": 3.04,
      "learning_rate": 4.98805256869773e-05,
      "loss": 0.739,
      "step": 190
    },
    {
      "epoch": 3.2,
      "learning_rate": 4.9581839904420555e-05,
      "loss": 0.7019,
      "step": 200
    },
    {
      "epoch": 3.36,
      "learning_rate": 4.92831541218638e-05,
      "loss": 0.6871,
      "step": 210
    },
    {
      "epoch": 3.52,
      "learning_rate": 4.898446833930705e-05,
      "loss": 0.6587,
      "step": 220
    },
    {
      "epoch": 3.68,
      "learning_rate": 4.86857825567503e-05,
      "loss": 0.6707,
      "step": 230
    },
    {
      "epoch": 3.84,
      "learning_rate": 4.8387096774193554e-05,
      "loss": 0.6668,
      "step": 240
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.80884109916368e-05,
      "loss": 0.6704,
      "step": 250
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7745208568207441,
      "eval_f1": 0.7663257051391182,
      "eval_loss": 0.685870885848999,
      "eval_precision": 0.7899083294864989,
      "eval_recall": 0.7745208568207441,
      "eval_runtime": 22.6111,
      "eval_samples_per_second": 78.457,
      "eval_steps_per_second": 1.238,
      "step": 250
    },
    {
      "epoch": 4.16,
      "learning_rate": 4.778972520908005e-05,
      "loss": 0.7294,
      "step": 260
    },
    {
      "epoch": 4.32,
      "learning_rate": 4.74910394265233e-05,
      "loss": 0.6412,
      "step": 270
    },
    {
      "epoch": 4.48,
      "learning_rate": 4.7192353643966546e-05,
      "loss": 0.5795,
      "step": 280
    },
    {
      "epoch": 4.64,
      "learning_rate": 4.6893667861409805e-05,
      "loss": 0.5931,
      "step": 290
    },
    {
      "epoch": 4.8,
      "learning_rate": 4.659498207885305e-05,
      "loss": 0.6334,
      "step": 300
    },
    {
      "epoch": 4.96,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.599,
      "step": 310
    },
    {
      "epoch": 4.99,
      "eval_accuracy": 0.8376550169109357,
      "eval_f1": 0.8359122308498544,
      "eval_loss": 0.47291457653045654,
      "eval_precision": 0.8411656033270575,
      "eval_recall": 0.8376550169109357,
      "eval_runtime": 22.5019,
      "eval_samples_per_second": 78.838,
      "eval_steps_per_second": 1.244,
      "step": 312
    },
    {
      "epoch": 5.12,
      "learning_rate": 4.5997610513739546e-05,
      "loss": 0.6053,
      "step": 320
    },
    {
      "epoch": 5.28,
      "learning_rate": 4.56989247311828e-05,
      "loss": 0.6048,
      "step": 330
    },
    {
      "epoch": 5.44,
      "learning_rate": 4.540023894862604e-05,
      "loss": 0.5873,
      "step": 340
    },
    {
      "epoch": 5.6,
      "learning_rate": 4.51015531660693e-05,
      "loss": 0.5687,
      "step": 350
    },
    {
      "epoch": 5.76,
      "learning_rate": 4.4802867383512545e-05,
      "loss": 0.5872,
      "step": 360
    },
    {
      "epoch": 5.92,
      "learning_rate": 4.4504181600955796e-05,
      "loss": 0.565,
      "step": 370
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8517474633596392,
      "eval_f1": 0.8507041486022975,
      "eval_loss": 0.4464806914329529,
      "eval_precision": 0.854164515473097,
      "eval_recall": 0.8517474633596392,
      "eval_runtime": 22.5499,
      "eval_samples_per_second": 78.67,
      "eval_steps_per_second": 1.242,
      "step": 375
    },
    {
      "epoch": 6.08,
      "learning_rate": 4.420549581839905e-05,
      "loss": 0.5211,
      "step": 380
    },
    {
      "epoch": 6.24,
      "learning_rate": 4.390681003584229e-05,
      "loss": 0.5786,
      "step": 390
    },
    {
      "epoch": 6.4,
      "learning_rate": 4.360812425328555e-05,
      "loss": 0.5157,
      "step": 400
    },
    {
      "epoch": 6.56,
      "learning_rate": 4.3309438470728796e-05,
      "loss": 0.548,
      "step": 410
    },
    {
      "epoch": 6.72,
      "learning_rate": 4.301075268817205e-05,
      "loss": 0.552,
      "step": 420
    },
    {
      "epoch": 6.88,
      "learning_rate": 4.271206690561529e-05,
      "loss": 0.5576,
      "step": 430
    },
    {
      "epoch": 6.99,
      "eval_accuracy": 0.8483652762119503,
      "eval_f1": 0.8451673044812301,
      "eval_loss": 0.44785261154174805,
      "eval_precision": 0.8565147367639446,
      "eval_recall": 0.8483652762119503,
      "eval_runtime": 22.4755,
      "eval_samples_per_second": 78.93,
      "eval_steps_per_second": 1.246,
      "step": 437
    },
    {
      "epoch": 7.04,
      "learning_rate": 4.241338112305854e-05,
      "loss": 0.572,
      "step": 440
    },
    {
      "epoch": 7.2,
      "learning_rate": 4.2114695340501795e-05,
      "loss": 0.524,
      "step": 450
    },
    {
      "epoch": 7.36,
      "learning_rate": 4.1816009557945046e-05,
      "loss": 0.5119,
      "step": 460
    },
    {
      "epoch": 7.52,
      "learning_rate": 4.15173237753883e-05,
      "loss": 0.4995,
      "step": 470
    },
    {
      "epoch": 7.68,
      "learning_rate": 4.121863799283154e-05,
      "loss": 0.4813,
      "step": 480
    },
    {
      "epoch": 7.84,
      "learning_rate": 4.0919952210274794e-05,
      "loss": 0.4991,
      "step": 490
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.062126642771804e-05,
      "loss": 0.4966,
      "step": 500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8387824126268321,
      "eval_f1": 0.8363143589644442,
      "eval_loss": 0.48700129985809326,
      "eval_precision": 0.8399418787633016,
      "eval_recall": 0.8387824126268321,
      "eval_runtime": 22.5233,
      "eval_samples_per_second": 78.763,
      "eval_steps_per_second": 1.243,
      "step": 500
    },
    {
      "epoch": 8.16,
      "learning_rate": 4.032258064516129e-05,
      "loss": 0.4157,
      "step": 510
    },
    {
      "epoch": 8.32,
      "learning_rate": 4.002389486260454e-05,
      "loss": 0.4344,
      "step": 520
    },
    {
      "epoch": 8.48,
      "learning_rate": 3.972520908004779e-05,
      "loss": 0.4766,
      "step": 530
    },
    {
      "epoch": 8.64,
      "learning_rate": 3.9426523297491045e-05,
      "loss": 0.5018,
      "step": 540
    },
    {
      "epoch": 8.8,
      "learning_rate": 3.912783751493429e-05,
      "loss": 0.5045,
      "step": 550
    },
    {
      "epoch": 8.96,
      "learning_rate": 3.882915173237754e-05,
      "loss": 0.4667,
      "step": 560
    },
    {
      "epoch": 8.99,
      "eval_accuracy": 0.8444193912063134,
      "eval_f1": 0.8442637097518081,
      "eval_loss": 0.4763273298740387,
      "eval_precision": 0.8495616108537656,
      "eval_recall": 0.8444193912063134,
      "eval_runtime": 22.5654,
      "eval_samples_per_second": 78.616,
      "eval_steps_per_second": 1.241,
      "step": 562
    },
    {
      "epoch": 9.12,
      "learning_rate": 3.8530465949820786e-05,
      "loss": 0.4291,
      "step": 570
    },
    {
      "epoch": 9.28,
      "learning_rate": 3.8231780167264044e-05,
      "loss": 0.4551,
      "step": 580
    },
    {
      "epoch": 9.44,
      "learning_rate": 3.793309438470729e-05,
      "loss": 0.4519,
      "step": 590
    },
    {
      "epoch": 9.6,
      "learning_rate": 3.763440860215054e-05,
      "loss": 0.4531,
      "step": 600
    },
    {
      "epoch": 9.76,
      "learning_rate": 3.7335722819593785e-05,
      "loss": 0.4401,
      "step": 610
    },
    {
      "epoch": 9.92,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.4264,
      "step": 620
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8376550169109357,
      "eval_f1": 0.832389205615137,
      "eval_loss": 0.4801747500896454,
      "eval_precision": 0.8377843163688572,
      "eval_recall": 0.8376550169109357,
      "eval_runtime": 22.469,
      "eval_samples_per_second": 78.953,
      "eval_steps_per_second": 1.246,
      "step": 625
    },
    {
      "epoch": 10.08,
      "learning_rate": 3.673835125448029e-05,
      "loss": 0.4258,
      "step": 630
    },
    {
      "epoch": 10.24,
      "learning_rate": 3.643966547192354e-05,
      "loss": 0.4114,
      "step": 640
    },
    {
      "epoch": 10.4,
      "learning_rate": 3.614097968936679e-05,
      "loss": 0.4501,
      "step": 650
    },
    {
      "epoch": 10.56,
      "learning_rate": 3.5842293906810036e-05,
      "loss": 0.4344,
      "step": 660
    },
    {
      "epoch": 10.72,
      "learning_rate": 3.554360812425329e-05,
      "loss": 0.4555,
      "step": 670
    },
    {
      "epoch": 10.88,
      "learning_rate": 3.524492234169653e-05,
      "loss": 0.445,
      "step": 680
    },
    {
      "epoch": 10.99,
      "eval_accuracy": 0.8376550169109357,
      "eval_f1": 0.8343152148299824,
      "eval_loss": 0.5245540738105774,
      "eval_precision": 0.8383432199805607,
      "eval_recall": 0.8376550169109357,
      "eval_runtime": 22.4048,
      "eval_samples_per_second": 79.179,
      "eval_steps_per_second": 1.25,
      "step": 687
    },
    {
      "epoch": 11.04,
      "learning_rate": 3.494623655913979e-05,
      "loss": 0.4096,
      "step": 690
    },
    {
      "epoch": 11.2,
      "learning_rate": 3.4647550776583035e-05,
      "loss": 0.3893,
      "step": 700
    },
    {
      "epoch": 11.36,
      "learning_rate": 3.4348864994026287e-05,
      "loss": 0.3837,
      "step": 710
    },
    {
      "epoch": 11.52,
      "learning_rate": 3.405017921146954e-05,
      "loss": 0.4279,
      "step": 720
    },
    {
      "epoch": 11.68,
      "learning_rate": 3.375149342891278e-05,
      "loss": 0.4347,
      "step": 730
    },
    {
      "epoch": 11.84,
      "learning_rate": 3.3452807646356034e-05,
      "loss": 0.3979,
      "step": 740
    },
    {
      "epoch": 12.0,
      "learning_rate": 3.3154121863799286e-05,
      "loss": 0.3935,
      "step": 750
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.8438556933483653,
      "eval_f1": 0.8434227278877919,
      "eval_loss": 0.48825180530548096,
      "eval_precision": 0.8519248121654718,
      "eval_recall": 0.8438556933483653,
      "eval_runtime": 22.4171,
      "eval_samples_per_second": 79.136,
      "eval_steps_per_second": 1.249,
      "step": 750
    },
    {
      "epoch": 12.16,
      "learning_rate": 3.285543608124254e-05,
      "loss": 0.3493,
      "step": 760
    },
    {
      "epoch": 12.32,
      "learning_rate": 3.255675029868578e-05,
      "loss": 0.3344,
      "step": 770
    },
    {
      "epoch": 12.48,
      "learning_rate": 3.2258064516129034e-05,
      "loss": 0.3653,
      "step": 780
    },
    {
      "epoch": 12.64,
      "learning_rate": 3.195937873357228e-05,
      "loss": 0.4012,
      "step": 790
    },
    {
      "epoch": 12.8,
      "learning_rate": 3.1660692951015537e-05,
      "loss": 0.3749,
      "step": 800
    },
    {
      "epoch": 12.96,
      "learning_rate": 3.136200716845878e-05,
      "loss": 0.374,
      "step": 810
    },
    {
      "epoch": 12.99,
      "eval_accuracy": 0.8568207440811725,
      "eval_f1": 0.8568588686911519,
      "eval_loss": 0.4510846436023712,
      "eval_precision": 0.8602543269576464,
      "eval_recall": 0.8568207440811725,
      "eval_runtime": 22.4634,
      "eval_samples_per_second": 78.973,
      "eval_steps_per_second": 1.246,
      "step": 812
    },
    {
      "epoch": 13.12,
      "learning_rate": 3.106332138590203e-05,
      "loss": 0.346,
      "step": 820
    },
    {
      "epoch": 13.28,
      "learning_rate": 3.0764635603345284e-05,
      "loss": 0.3608,
      "step": 830
    },
    {
      "epoch": 13.44,
      "learning_rate": 3.046594982078853e-05,
      "loss": 0.3706,
      "step": 840
    },
    {
      "epoch": 13.6,
      "learning_rate": 3.016726403823178e-05,
      "loss": 0.3548,
      "step": 850
    },
    {
      "epoch": 13.76,
      "learning_rate": 2.9868578255675032e-05,
      "loss": 0.3618,
      "step": 860
    },
    {
      "epoch": 13.92,
      "learning_rate": 2.9569892473118284e-05,
      "loss": 0.3551,
      "step": 870
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.85456595264938,
      "eval_f1": 0.8495500045026548,
      "eval_loss": 0.5152960419654846,
      "eval_precision": 0.8517389435426439,
      "eval_recall": 0.85456595264938,
      "eval_runtime": 22.3914,
      "eval_samples_per_second": 79.227,
      "eval_steps_per_second": 1.25,
      "step": 875
    },
    {
      "epoch": 14.08,
      "learning_rate": 2.9271206690561532e-05,
      "loss": 0.3527,
      "step": 880
    },
    {
      "epoch": 14.24,
      "learning_rate": 2.897252090800478e-05,
      "loss": 0.3111,
      "step": 890
    },
    {
      "epoch": 14.4,
      "learning_rate": 2.8673835125448028e-05,
      "loss": 0.316,
      "step": 900
    },
    {
      "epoch": 14.56,
      "learning_rate": 2.8375149342891276e-05,
      "loss": 0.3448,
      "step": 910
    },
    {
      "epoch": 14.72,
      "learning_rate": 2.807646356033453e-05,
      "loss": 0.341,
      "step": 920
    },
    {
      "epoch": 14.88,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.3573,
      "step": 930
    },
    {
      "epoch": 14.99,
      "eval_accuracy": 0.8579481397970687,
      "eval_f1": 0.8559499586233215,
      "eval_loss": 0.47047653794288635,
      "eval_precision": 0.8554345481699459,
      "eval_recall": 0.8579481397970687,
      "eval_runtime": 22.4353,
      "eval_samples_per_second": 79.072,
      "eval_steps_per_second": 1.248,
      "step": 937
    },
    {
      "epoch": 15.04,
      "learning_rate": 2.747909199522103e-05,
      "loss": 0.3155,
      "step": 940
    },
    {
      "epoch": 15.2,
      "learning_rate": 2.718040621266428e-05,
      "loss": 0.296,
      "step": 950
    },
    {
      "epoch": 15.36,
      "learning_rate": 2.6881720430107527e-05,
      "loss": 0.2889,
      "step": 960
    },
    {
      "epoch": 15.52,
      "learning_rate": 2.6583034647550775e-05,
      "loss": 0.3255,
      "step": 970
    },
    {
      "epoch": 15.68,
      "learning_rate": 2.628434886499403e-05,
      "loss": 0.3096,
      "step": 980
    },
    {
      "epoch": 15.84,
      "learning_rate": 2.5985663082437278e-05,
      "loss": 0.32,
      "step": 990
    },
    {
      "epoch": 16.0,
      "learning_rate": 2.5686977299880526e-05,
      "loss": 0.3385,
      "step": 1000
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.8517474633596392,
      "eval_f1": 0.8517446619627265,
      "eval_loss": 0.45465072989463806,
      "eval_precision": 0.8534550844421815,
      "eval_recall": 0.8517474633596392,
      "eval_runtime": 22.4407,
      "eval_samples_per_second": 79.053,
      "eval_steps_per_second": 1.248,
      "step": 1000
    },
    {
      "epoch": 16.16,
      "learning_rate": 2.5388291517323774e-05,
      "loss": 0.3419,
      "step": 1010
    },
    {
      "epoch": 16.32,
      "learning_rate": 2.5089605734767026e-05,
      "loss": 0.2798,
      "step": 1020
    },
    {
      "epoch": 16.48,
      "learning_rate": 2.4790919952210277e-05,
      "loss": 0.2843,
      "step": 1030
    },
    {
      "epoch": 16.64,
      "learning_rate": 2.4492234169653525e-05,
      "loss": 0.3334,
      "step": 1040
    },
    {
      "epoch": 16.8,
      "learning_rate": 2.4193548387096777e-05,
      "loss": 0.2984,
      "step": 1050
    },
    {
      "epoch": 16.96,
      "learning_rate": 2.3894862604540025e-05,
      "loss": 0.2764,
      "step": 1060
    },
    {
      "epoch": 16.99,
      "eval_accuracy": 0.8528748590755355,
      "eval_f1": 0.851324755468265,
      "eval_loss": 0.5189043879508972,
      "eval_precision": 0.8543914754699317,
      "eval_recall": 0.8528748590755355,
      "eval_runtime": 22.43,
      "eval_samples_per_second": 79.091,
      "eval_steps_per_second": 1.248,
      "step": 1062
    },
    {
      "epoch": 17.12,
      "learning_rate": 2.3596176821983273e-05,
      "loss": 0.3212,
      "step": 1070
    },
    {
      "epoch": 17.28,
      "learning_rate": 2.3297491039426525e-05,
      "loss": 0.2831,
      "step": 1080
    },
    {
      "epoch": 17.44,
      "learning_rate": 2.2998805256869773e-05,
      "loss": 0.3095,
      "step": 1090
    },
    {
      "epoch": 17.6,
      "learning_rate": 2.270011947431302e-05,
      "loss": 0.2903,
      "step": 1100
    },
    {
      "epoch": 17.76,
      "learning_rate": 2.2401433691756272e-05,
      "loss": 0.2934,
      "step": 1110
    },
    {
      "epoch": 17.92,
      "learning_rate": 2.2102747909199524e-05,
      "loss": 0.2895,
      "step": 1120
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.8602029312288614,
      "eval_f1": 0.8585833513511913,
      "eval_loss": 0.5393418669700623,
      "eval_precision": 0.8586924111201653,
      "eval_recall": 0.8602029312288614,
      "eval_runtime": 22.347,
      "eval_samples_per_second": 79.384,
      "eval_steps_per_second": 1.253,
      "step": 1125
    },
    {
      "epoch": 18.08,
      "learning_rate": 2.1804062126642775e-05,
      "loss": 0.2808,
      "step": 1130
    },
    {
      "epoch": 18.24,
      "learning_rate": 2.1505376344086024e-05,
      "loss": 0.262,
      "step": 1140
    },
    {
      "epoch": 18.4,
      "learning_rate": 2.120669056152927e-05,
      "loss": 0.2419,
      "step": 1150
    },
    {
      "epoch": 18.56,
      "learning_rate": 2.0908004778972523e-05,
      "loss": 0.2588,
      "step": 1160
    },
    {
      "epoch": 18.72,
      "learning_rate": 2.060931899641577e-05,
      "loss": 0.2648,
      "step": 1170
    },
    {
      "epoch": 18.88,
      "learning_rate": 2.031063321385902e-05,
      "loss": 0.2738,
      "step": 1180
    },
    {
      "epoch": 18.99,
      "eval_accuracy": 0.8404735062006764,
      "eval_f1": 0.8380755280956992,
      "eval_loss": 0.5554386973381042,
      "eval_precision": 0.8435879241848919,
      "eval_recall": 0.8404735062006764,
      "eval_runtime": 22.5392,
      "eval_samples_per_second": 78.707,
      "eval_steps_per_second": 1.242,
      "step": 1187
    },
    {
      "epoch": 19.04,
      "learning_rate": 2.001194743130227e-05,
      "loss": 0.269,
      "step": 1190
    },
    {
      "epoch": 19.2,
      "learning_rate": 1.9713261648745522e-05,
      "loss": 0.2484,
      "step": 1200
    },
    {
      "epoch": 19.36,
      "learning_rate": 1.941457586618877e-05,
      "loss": 0.2571,
      "step": 1210
    },
    {
      "epoch": 19.52,
      "learning_rate": 1.9115890083632022e-05,
      "loss": 0.2537,
      "step": 1220
    },
    {
      "epoch": 19.68,
      "learning_rate": 1.881720430107527e-05,
      "loss": 0.2556,
      "step": 1230
    },
    {
      "epoch": 19.84,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.2732,
      "step": 1240
    },
    {
      "epoch": 20.0,
      "learning_rate": 1.821983273596177e-05,
      "loss": 0.2563,
      "step": 1250
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.8607666290868095,
      "eval_f1": 0.8574201811479036,
      "eval_loss": 0.5478461980819702,
      "eval_precision": 0.8573038878183122,
      "eval_recall": 0.8607666290868095,
      "eval_runtime": 22.5894,
      "eval_samples_per_second": 78.532,
      "eval_steps_per_second": 1.24,
      "step": 1250
    },
    {
      "epoch": 20.16,
      "learning_rate": 1.7921146953405018e-05,
      "loss": 0.2353,
      "step": 1260
    },
    {
      "epoch": 20.32,
      "learning_rate": 1.7622461170848266e-05,
      "loss": 0.233,
      "step": 1270
    },
    {
      "epoch": 20.48,
      "learning_rate": 1.7323775388291518e-05,
      "loss": 0.2367,
      "step": 1280
    },
    {
      "epoch": 20.64,
      "learning_rate": 1.702508960573477e-05,
      "loss": 0.249,
      "step": 1290
    },
    {
      "epoch": 20.8,
      "learning_rate": 1.6726403823178017e-05,
      "loss": 0.2451,
      "step": 1300
    },
    {
      "epoch": 20.96,
      "learning_rate": 1.642771804062127e-05,
      "loss": 0.2375,
      "step": 1310
    },
    {
      "epoch": 20.99,
      "eval_accuracy": 0.8664036076662909,
      "eval_f1": 0.8621553424312518,
      "eval_loss": 0.5512430667877197,
      "eval_precision": 0.8651236987001303,
      "eval_recall": 0.8664036076662909,
      "eval_runtime": 22.506,
      "eval_samples_per_second": 78.823,
      "eval_steps_per_second": 1.244,
      "step": 1312
    },
    {
      "epoch": 21.12,
      "learning_rate": 1.6129032258064517e-05,
      "loss": 0.2158,
      "step": 1320
    },
    {
      "epoch": 21.28,
      "learning_rate": 1.5830346475507768e-05,
      "loss": 0.2346,
      "step": 1330
    },
    {
      "epoch": 21.44,
      "learning_rate": 1.5531660692951016e-05,
      "loss": 0.2307,
      "step": 1340
    },
    {
      "epoch": 21.6,
      "learning_rate": 1.5232974910394265e-05,
      "loss": 0.2246,
      "step": 1350
    },
    {
      "epoch": 21.76,
      "learning_rate": 1.4934289127837516e-05,
      "loss": 0.2456,
      "step": 1360
    },
    {
      "epoch": 21.92,
      "learning_rate": 1.4635603345280766e-05,
      "loss": 0.2599,
      "step": 1370
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.8624577226606539,
      "eval_f1": 0.8599244363345828,
      "eval_loss": 0.5316722393035889,
      "eval_precision": 0.8606629640313184,
      "eval_recall": 0.8624577226606539,
      "eval_runtime": 22.543,
      "eval_samples_per_second": 78.694,
      "eval_steps_per_second": 1.242,
      "step": 1375
    },
    {
      "epoch": 22.08,
      "learning_rate": 1.4336917562724014e-05,
      "loss": 0.2305,
      "step": 1380
    },
    {
      "epoch": 22.24,
      "learning_rate": 1.4038231780167265e-05,
      "loss": 0.1953,
      "step": 1390
    },
    {
      "epoch": 22.4,
      "learning_rate": 1.3739545997610515e-05,
      "loss": 0.2223,
      "step": 1400
    },
    {
      "epoch": 22.56,
      "learning_rate": 1.3440860215053763e-05,
      "loss": 0.2179,
      "step": 1410
    },
    {
      "epoch": 22.72,
      "learning_rate": 1.3142174432497015e-05,
      "loss": 0.2149,
      "step": 1420
    },
    {
      "epoch": 22.88,
      "learning_rate": 1.2843488649940263e-05,
      "loss": 0.2146,
      "step": 1430
    },
    {
      "epoch": 22.99,
      "eval_accuracy": 0.8568207440811725,
      "eval_f1": 0.855887173333211,
      "eval_loss": 0.5971649885177612,
      "eval_precision": 0.8566915988066761,
      "eval_recall": 0.8568207440811725,
      "eval_runtime": 22.5854,
      "eval_samples_per_second": 78.546,
      "eval_steps_per_second": 1.24,
      "step": 1437
    },
    {
      "epoch": 23.04,
      "learning_rate": 1.2544802867383513e-05,
      "loss": 0.2063,
      "step": 1440
    },
    {
      "epoch": 23.2,
      "learning_rate": 1.2246117084826763e-05,
      "loss": 0.209,
      "step": 1450
    },
    {
      "epoch": 23.36,
      "learning_rate": 1.1947431302270013e-05,
      "loss": 0.2039,
      "step": 1460
    },
    {
      "epoch": 23.52,
      "learning_rate": 1.1648745519713262e-05,
      "loss": 0.22,
      "step": 1470
    },
    {
      "epoch": 23.68,
      "learning_rate": 1.135005973715651e-05,
      "loss": 0.2463,
      "step": 1480
    },
    {
      "epoch": 23.84,
      "learning_rate": 1.1051373954599762e-05,
      "loss": 0.1984,
      "step": 1490
    },
    {
      "epoch": 24.0,
      "learning_rate": 1.0752688172043012e-05,
      "loss": 0.2132,
      "step": 1500
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8635851183765502,
      "eval_f1": 0.8606211950492338,
      "eval_loss": 0.5933902859687805,
      "eval_precision": 0.8617445208407427,
      "eval_recall": 0.8635851183765502,
      "eval_runtime": 22.5856,
      "eval_samples_per_second": 78.546,
      "eval_steps_per_second": 1.24,
      "step": 1500
    },
    {
      "epoch": 24.16,
      "learning_rate": 1.0454002389486262e-05,
      "loss": 0.2098,
      "step": 1510
    },
    {
      "epoch": 24.32,
      "learning_rate": 1.015531660692951e-05,
      "loss": 0.22,
      "step": 1520
    },
    {
      "epoch": 24.48,
      "learning_rate": 9.856630824372761e-06,
      "loss": 0.2177,
      "step": 1530
    },
    {
      "epoch": 24.64,
      "learning_rate": 9.557945041816011e-06,
      "loss": 0.192,
      "step": 1540
    },
    {
      "epoch": 24.8,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.2001,
      "step": 1550
    },
    {
      "epoch": 24.96,
      "learning_rate": 8.960573476702509e-06,
      "loss": 0.2036,
      "step": 1560
    },
    {
      "epoch": 24.99,
      "eval_accuracy": 0.8664036076662909,
      "eval_f1": 0.8657672120824895,
      "eval_loss": 0.5923020839691162,
      "eval_precision": 0.8662484245822287,
      "eval_recall": 0.8664036076662909,
      "eval_runtime": 22.5989,
      "eval_samples_per_second": 78.499,
      "eval_steps_per_second": 1.239,
      "step": 1562
    },
    {
      "epoch": 25.12,
      "learning_rate": 8.661887694145759e-06,
      "loss": 0.2018,
      "step": 1570
    },
    {
      "epoch": 25.28,
      "learning_rate": 8.363201911589009e-06,
      "loss": 0.1692,
      "step": 1580
    },
    {
      "epoch": 25.44,
      "learning_rate": 8.064516129032258e-06,
      "loss": 0.181,
      "step": 1590
    },
    {
      "epoch": 25.6,
      "learning_rate": 7.765830346475508e-06,
      "loss": 0.1927,
      "step": 1600
    },
    {
      "epoch": 25.76,
      "learning_rate": 7.467144563918758e-06,
      "loss": 0.1873,
      "step": 1610
    },
    {
      "epoch": 25.92,
      "learning_rate": 7.168458781362007e-06,
      "loss": 0.1971,
      "step": 1620
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.863021420518602,
      "eval_f1": 0.8621322077660976,
      "eval_loss": 0.5839389562606812,
      "eval_precision": 0.8621026992886056,
      "eval_recall": 0.863021420518602,
      "eval_runtime": 22.442,
      "eval_samples_per_second": 79.048,
      "eval_steps_per_second": 1.248,
      "step": 1625
    },
    {
      "epoch": 26.08,
      "learning_rate": 6.869772998805258e-06,
      "loss": 0.1971,
      "step": 1630
    },
    {
      "epoch": 26.24,
      "learning_rate": 6.5710872162485075e-06,
      "loss": 0.1948,
      "step": 1640
    },
    {
      "epoch": 26.4,
      "learning_rate": 6.2724014336917564e-06,
      "loss": 0.1886,
      "step": 1650
    },
    {
      "epoch": 26.56,
      "learning_rate": 5.973715651135006e-06,
      "loss": 0.1928,
      "step": 1660
    },
    {
      "epoch": 26.72,
      "learning_rate": 5.675029868578255e-06,
      "loss": 0.1966,
      "step": 1670
    },
    {
      "epoch": 26.88,
      "learning_rate": 5.376344086021506e-06,
      "loss": 0.1878,
      "step": 1680
    },
    {
      "epoch": 26.99,
      "eval_accuracy": 0.8624577226606539,
      "eval_f1": 0.8640203850514281,
      "eval_loss": 0.5907111763954163,
      "eval_precision": 0.8669146424665295,
      "eval_recall": 0.8624577226606539,
      "eval_runtime": 22.4258,
      "eval_samples_per_second": 79.105,
      "eval_steps_per_second": 1.249,
      "step": 1687
    },
    {
      "epoch": 27.04,
      "learning_rate": 5.077658303464755e-06,
      "loss": 0.1874,
      "step": 1690
    },
    {
      "epoch": 27.2,
      "learning_rate": 4.7789725209080055e-06,
      "loss": 0.1776,
      "step": 1700
    },
    {
      "epoch": 27.36,
      "learning_rate": 4.4802867383512545e-06,
      "loss": 0.2073,
      "step": 1710
    },
    {
      "epoch": 27.52,
      "learning_rate": 4.181600955794504e-06,
      "loss": 0.1797,
      "step": 1720
    },
    {
      "epoch": 27.68,
      "learning_rate": 3.882915173237754e-06,
      "loss": 0.1631,
      "step": 1730
    },
    {
      "epoch": 27.84,
      "learning_rate": 3.5842293906810035e-06,
      "loss": 0.178,
      "step": 1740
    },
    {
      "epoch": 28.0,
      "learning_rate": 3.2855436081242537e-06,
      "loss": 0.1922,
      "step": 1750
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8692220969560316,
      "eval_f1": 0.8680166874932149,
      "eval_loss": 0.6058236956596375,
      "eval_precision": 0.8684018429035691,
      "eval_recall": 0.8692220969560316,
      "eval_runtime": 22.5447,
      "eval_samples_per_second": 78.688,
      "eval_steps_per_second": 1.242,
      "step": 1750
    },
    {
      "epoch": 28.16,
      "learning_rate": 2.986857825567503e-06,
      "loss": 0.1773,
      "step": 1760
    },
    {
      "epoch": 28.32,
      "learning_rate": 2.688172043010753e-06,
      "loss": 0.1778,
      "step": 1770
    },
    {
      "epoch": 28.48,
      "learning_rate": 2.3894862604540028e-06,
      "loss": 0.1726,
      "step": 1780
    },
    {
      "epoch": 28.64,
      "learning_rate": 2.090800477897252e-06,
      "loss": 0.1653,
      "step": 1790
    },
    {
      "epoch": 28.8,
      "learning_rate": 1.7921146953405017e-06,
      "loss": 0.1761,
      "step": 1800
    },
    {
      "epoch": 28.96,
      "learning_rate": 1.4934289127837516e-06,
      "loss": 0.1854,
      "step": 1810
    },
    {
      "epoch": 28.99,
      "eval_accuracy": 0.8669673055242391,
      "eval_f1": 0.8654906122723444,
      "eval_loss": 0.6014190912246704,
      "eval_precision": 0.8653037200870781,
      "eval_recall": 0.8669673055242391,
      "eval_runtime": 22.397,
      "eval_samples_per_second": 79.207,
      "eval_steps_per_second": 1.25,
      "step": 1812
    },
    {
      "epoch": 29.12,
      "learning_rate": 1.1947431302270014e-06,
      "loss": 0.1739,
      "step": 1820
    },
    {
      "epoch": 29.28,
      "learning_rate": 8.960573476702509e-07,
      "loss": 0.1633,
      "step": 1830
    },
    {
      "epoch": 29.44,
      "learning_rate": 5.973715651135007e-07,
      "loss": 0.1753,
      "step": 1840
    },
    {
      "epoch": 29.6,
      "learning_rate": 2.9868578255675034e-07,
      "loss": 0.1757,
      "step": 1850
    },
    {
      "epoch": 29.76,
      "learning_rate": 0.0,
      "loss": 0.1688,
      "step": 1860
    },
    {
      "epoch": 29.76,
      "eval_accuracy": 0.8737316798196166,
      "eval_f1": 0.872156389323959,
      "eval_loss": 0.586895227432251,
      "eval_precision": 0.8722327180586174,
      "eval_recall": 0.8737316798196166,
      "eval_runtime": 22.5331,
      "eval_samples_per_second": 78.728,
      "eval_steps_per_second": 1.243,
      "step": 1860
    },
    {
      "epoch": 29.76,
      "step": 1860,
      "total_flos": 1.7058003968011272e+20,
      "train_loss": 0.40524982412656146,
      "train_runtime": 15843.7635,
      "train_samples_per_second": 30.224,
      "train_steps_per_second": 0.117
    }
  ],
  "logging_steps": 10,
  "max_steps": 1860,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 1.7058003968011272e+20,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}