{
  "best_metric": 0.8540022547914318,
  "best_model_checkpoint": "swinv2-base-patch4-window8-256-finetuned-galaxy10-decals/checkpoint-1687",
  "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": 2.2882,
      "step": 10
    },
    {
      "epoch": 0.32,
      "learning_rate": 5.376344086021506e-06,
      "loss": 2.2435,
      "step": 20
    },
    {
      "epoch": 0.48,
      "learning_rate": 8.064516129032258e-06,
      "loss": 2.1749,
      "step": 30
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.0752688172043012e-05,
      "loss": 2.057,
      "step": 40
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.3440860215053763e-05,
      "loss": 1.8835,
      "step": 50
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.6129032258064517e-05,
      "loss": 1.6195,
      "step": 60
    },
    {
      "epoch": 0.99,
      "eval_accuracy": 0.5101465614430665,
      "eval_f1": 0.478159592811226,
      "eval_loss": 1.4006441831588745,
      "eval_precision": 0.49101191687459667,
      "eval_recall": 0.5101465614430665,
      "eval_runtime": 14.9921,
      "eval_samples_per_second": 118.329,
      "eval_steps_per_second": 1.868,
      "step": 62
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.881720430107527e-05,
      "loss": 1.5199,
      "step": 70
    },
    {
      "epoch": 1.28,
      "learning_rate": 2.1505376344086024e-05,
      "loss": 1.3485,
      "step": 80
    },
    {
      "epoch": 1.44,
      "learning_rate": 2.4193548387096777e-05,
      "loss": 1.1999,
      "step": 90
    },
    {
      "epoch": 1.6,
      "learning_rate": 2.6881720430107527e-05,
      "loss": 1.1126,
      "step": 100
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.9569892473118284e-05,
      "loss": 0.9786,
      "step": 110
    },
    {
      "epoch": 1.92,
      "learning_rate": 3.2258064516129034e-05,
      "loss": 0.9423,
      "step": 120
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7615558060879368,
      "eval_f1": 0.7531462877179919,
      "eval_loss": 0.720885694026947,
      "eval_precision": 0.7617080220354211,
      "eval_recall": 0.7615558060879368,
      "eval_runtime": 14.5976,
      "eval_samples_per_second": 121.527,
      "eval_steps_per_second": 1.918,
      "step": 125
    },
    {
      "epoch": 2.08,
      "learning_rate": 3.494623655913979e-05,
      "loss": 0.9023,
      "step": 130
    },
    {
      "epoch": 2.24,
      "learning_rate": 3.763440860215054e-05,
      "loss": 0.8517,
      "step": 140
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.032258064516129e-05,
      "loss": 0.8301,
      "step": 150
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.301075268817205e-05,
      "loss": 0.8547,
      "step": 160
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.56989247311828e-05,
      "loss": 0.8414,
      "step": 170
    },
    {
      "epoch": 2.88,
      "learning_rate": 4.8387096774193554e-05,
      "loss": 0.8171,
      "step": 180
    },
    {
      "epoch": 2.99,
      "eval_accuracy": 0.8010146561443067,
      "eval_f1": 0.7938025504400419,
      "eval_loss": 0.5841514468193054,
      "eval_precision": 0.7950174478591531,
      "eval_recall": 0.8010146561443067,
      "eval_runtime": 14.6825,
      "eval_samples_per_second": 120.824,
      "eval_steps_per_second": 1.907,
      "step": 187
    },
    {
      "epoch": 3.04,
      "learning_rate": 4.98805256869773e-05,
      "loss": 0.8232,
      "step": 190
    },
    {
      "epoch": 3.2,
      "learning_rate": 4.9581839904420555e-05,
      "loss": 0.731,
      "step": 200
    },
    {
      "epoch": 3.36,
      "learning_rate": 4.92831541218638e-05,
      "loss": 0.7349,
      "step": 210
    },
    {
      "epoch": 3.52,
      "learning_rate": 4.898446833930705e-05,
      "loss": 0.7389,
      "step": 220
    },
    {
      "epoch": 3.68,
      "learning_rate": 4.86857825567503e-05,
      "loss": 0.6817,
      "step": 230
    },
    {
      "epoch": 3.84,
      "learning_rate": 4.8387096774193554e-05,
      "loss": 0.7092,
      "step": 240
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.80884109916368e-05,
      "loss": 0.6609,
      "step": 250
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.822435174746336,
      "eval_f1": 0.8142783009598019,
      "eval_loss": 0.5000367164611816,
      "eval_precision": 0.8158639775281362,
      "eval_recall": 0.822435174746336,
      "eval_runtime": 14.7332,
      "eval_samples_per_second": 120.408,
      "eval_steps_per_second": 1.9,
      "step": 250
    },
    {
      "epoch": 4.16,
      "learning_rate": 4.778972520908005e-05,
      "loss": 0.7047,
      "step": 260
    },
    {
      "epoch": 4.32,
      "learning_rate": 4.74910394265233e-05,
      "loss": 0.6669,
      "step": 270
    },
    {
      "epoch": 4.48,
      "learning_rate": 4.7192353643966546e-05,
      "loss": 0.658,
      "step": 280
    },
    {
      "epoch": 4.64,
      "learning_rate": 4.6893667861409805e-05,
      "loss": 0.6375,
      "step": 290
    },
    {
      "epoch": 4.8,
      "learning_rate": 4.659498207885305e-05,
      "loss": 0.6794,
      "step": 300
    },
    {
      "epoch": 4.96,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.5927,
      "step": 310
    },
    {
      "epoch": 4.99,
      "eval_accuracy": 0.8190529875986471,
      "eval_f1": 0.8184444482553919,
      "eval_loss": 0.5367255210876465,
      "eval_precision": 0.8210887386708854,
      "eval_recall": 0.8190529875986471,
      "eval_runtime": 14.5072,
      "eval_samples_per_second": 122.284,
      "eval_steps_per_second": 1.93,
      "step": 312
    },
    {
      "epoch": 5.12,
      "learning_rate": 4.5997610513739546e-05,
      "loss": 0.6292,
      "step": 320
    },
    {
      "epoch": 5.28,
      "learning_rate": 4.56989247311828e-05,
      "loss": 0.6518,
      "step": 330
    },
    {
      "epoch": 5.44,
      "learning_rate": 4.540023894862604e-05,
      "loss": 0.6207,
      "step": 340
    },
    {
      "epoch": 5.6,
      "learning_rate": 4.51015531660693e-05,
      "loss": 0.6184,
      "step": 350
    },
    {
      "epoch": 5.76,
      "learning_rate": 4.4802867383512545e-05,
      "loss": 0.5947,
      "step": 360
    },
    {
      "epoch": 5.92,
      "learning_rate": 4.4504181600955796e-05,
      "loss": 0.624,
      "step": 370
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8286358511837655,
      "eval_f1": 0.8211779080603236,
      "eval_loss": 0.49461066722869873,
      "eval_precision": 0.829528856549966,
      "eval_recall": 0.8286358511837655,
      "eval_runtime": 14.9381,
      "eval_samples_per_second": 118.757,
      "eval_steps_per_second": 1.874,
      "step": 375
    },
    {
      "epoch": 6.08,
      "learning_rate": 4.420549581839905e-05,
      "loss": 0.5834,
      "step": 380
    },
    {
      "epoch": 6.24,
      "learning_rate": 4.390681003584229e-05,
      "loss": 0.5863,
      "step": 390
    },
    {
      "epoch": 6.4,
      "learning_rate": 4.360812425328555e-05,
      "loss": 0.5658,
      "step": 400
    },
    {
      "epoch": 6.56,
      "learning_rate": 4.3309438470728796e-05,
      "loss": 0.5821,
      "step": 410
    },
    {
      "epoch": 6.72,
      "learning_rate": 4.301075268817205e-05,
      "loss": 0.5777,
      "step": 420
    },
    {
      "epoch": 6.88,
      "learning_rate": 4.271206690561529e-05,
      "loss": 0.5891,
      "step": 430
    },
    {
      "epoch": 6.99,
      "eval_accuracy": 0.8218714768883878,
      "eval_f1": 0.8201216062794426,
      "eval_loss": 0.5067989826202393,
      "eval_precision": 0.8243751790259868,
      "eval_recall": 0.8218714768883878,
      "eval_runtime": 14.8573,
      "eval_samples_per_second": 119.403,
      "eval_steps_per_second": 1.885,
      "step": 437
    },
    {
      "epoch": 7.04,
      "learning_rate": 4.241338112305854e-05,
      "loss": 0.6164,
      "step": 440
    },
    {
      "epoch": 7.2,
      "learning_rate": 4.2114695340501795e-05,
      "loss": 0.5936,
      "step": 450
    },
    {
      "epoch": 7.36,
      "learning_rate": 4.1816009557945046e-05,
      "loss": 0.5529,
      "step": 460
    },
    {
      "epoch": 7.52,
      "learning_rate": 4.15173237753883e-05,
      "loss": 0.5683,
      "step": 470
    },
    {
      "epoch": 7.68,
      "learning_rate": 4.121863799283154e-05,
      "loss": 0.5434,
      "step": 480
    },
    {
      "epoch": 7.84,
      "learning_rate": 4.0919952210274794e-05,
      "loss": 0.5501,
      "step": 490
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.062126642771804e-05,
      "loss": 0.5597,
      "step": 500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8229988726042841,
      "eval_f1": 0.819820354133241,
      "eval_loss": 0.5071399211883545,
      "eval_precision": 0.8381743468025983,
      "eval_recall": 0.8229988726042841,
      "eval_runtime": 14.8276,
      "eval_samples_per_second": 119.641,
      "eval_steps_per_second": 1.888,
      "step": 500
    },
    {
      "epoch": 8.16,
      "learning_rate": 4.032258064516129e-05,
      "loss": 0.5365,
      "step": 510
    },
    {
      "epoch": 8.32,
      "learning_rate": 4.002389486260454e-05,
      "loss": 0.5339,
      "step": 520
    },
    {
      "epoch": 8.48,
      "learning_rate": 3.972520908004779e-05,
      "loss": 0.5937,
      "step": 530
    },
    {
      "epoch": 8.64,
      "learning_rate": 3.9426523297491045e-05,
      "loss": 0.5316,
      "step": 540
    },
    {
      "epoch": 8.8,
      "learning_rate": 3.912783751493429e-05,
      "loss": 0.5565,
      "step": 550
    },
    {
      "epoch": 8.96,
      "learning_rate": 3.882915173237754e-05,
      "loss": 0.5292,
      "step": 560
    },
    {
      "epoch": 8.99,
      "eval_accuracy": 0.8444193912063134,
      "eval_f1": 0.8425840733336722,
      "eval_loss": 0.4463561177253723,
      "eval_precision": 0.8461878988130385,
      "eval_recall": 0.8444193912063134,
      "eval_runtime": 14.5832,
      "eval_samples_per_second": 121.647,
      "eval_steps_per_second": 1.92,
      "step": 562
    },
    {
      "epoch": 9.12,
      "learning_rate": 3.8530465949820786e-05,
      "loss": 0.5276,
      "step": 570
    },
    {
      "epoch": 9.28,
      "learning_rate": 3.8231780167264044e-05,
      "loss": 0.5219,
      "step": 580
    },
    {
      "epoch": 9.44,
      "learning_rate": 3.793309438470729e-05,
      "loss": 0.5171,
      "step": 590
    },
    {
      "epoch": 9.6,
      "learning_rate": 3.763440860215054e-05,
      "loss": 0.5229,
      "step": 600
    },
    {
      "epoch": 9.76,
      "learning_rate": 3.7335722819593785e-05,
      "loss": 0.4924,
      "step": 610
    },
    {
      "epoch": 9.92,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.5143,
      "step": 620
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8370913190529876,
      "eval_f1": 0.834952706810825,
      "eval_loss": 0.45558929443359375,
      "eval_precision": 0.8420278125941518,
      "eval_recall": 0.8370913190529876,
      "eval_runtime": 14.6031,
      "eval_samples_per_second": 121.481,
      "eval_steps_per_second": 1.917,
      "step": 625
    },
    {
      "epoch": 10.08,
      "learning_rate": 3.673835125448029e-05,
      "loss": 0.5206,
      "step": 630
    },
    {
      "epoch": 10.24,
      "learning_rate": 3.643966547192354e-05,
      "loss": 0.5084,
      "step": 640
    },
    {
      "epoch": 10.4,
      "learning_rate": 3.614097968936679e-05,
      "loss": 0.5251,
      "step": 650
    },
    {
      "epoch": 10.56,
      "learning_rate": 3.5842293906810036e-05,
      "loss": 0.4951,
      "step": 660
    },
    {
      "epoch": 10.72,
      "learning_rate": 3.554360812425329e-05,
      "loss": 0.4857,
      "step": 670
    },
    {
      "epoch": 10.88,
      "learning_rate": 3.524492234169653e-05,
      "loss": 0.5122,
      "step": 680
    },
    {
      "epoch": 10.99,
      "eval_accuracy": 0.8382187147688839,
      "eval_f1": 0.8368656096018949,
      "eval_loss": 0.4764641225337982,
      "eval_precision": 0.8432532753176901,
      "eval_recall": 0.8382187147688839,
      "eval_runtime": 14.5688,
      "eval_samples_per_second": 121.767,
      "eval_steps_per_second": 1.922,
      "step": 687
    },
    {
      "epoch": 11.04,
      "learning_rate": 3.494623655913979e-05,
      "loss": 0.4707,
      "step": 690
    },
    {
      "epoch": 11.2,
      "learning_rate": 3.4647550776583035e-05,
      "loss": 0.4726,
      "step": 700
    },
    {
      "epoch": 11.36,
      "learning_rate": 3.4348864994026287e-05,
      "loss": 0.4691,
      "step": 710
    },
    {
      "epoch": 11.52,
      "learning_rate": 3.405017921146954e-05,
      "loss": 0.4643,
      "step": 720
    },
    {
      "epoch": 11.68,
      "learning_rate": 3.375149342891278e-05,
      "loss": 0.4981,
      "step": 730
    },
    {
      "epoch": 11.84,
      "learning_rate": 3.3452807646356034e-05,
      "loss": 0.4818,
      "step": 740
    },
    {
      "epoch": 12.0,
      "learning_rate": 3.3154121863799286e-05,
      "loss": 0.4647,
      "step": 750
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.8365276211950394,
      "eval_f1": 0.834792825674035,
      "eval_loss": 0.4899607002735138,
      "eval_precision": 0.8443027256326996,
      "eval_recall": 0.8365276211950394,
      "eval_runtime": 14.6881,
      "eval_samples_per_second": 120.778,
      "eval_steps_per_second": 1.906,
      "step": 750
    },
    {
      "epoch": 12.16,
      "learning_rate": 3.285543608124254e-05,
      "loss": 0.4725,
      "step": 760
    },
    {
      "epoch": 12.32,
      "learning_rate": 3.255675029868578e-05,
      "loss": 0.4607,
      "step": 770
    },
    {
      "epoch": 12.48,
      "learning_rate": 3.2258064516129034e-05,
      "loss": 0.4621,
      "step": 780
    },
    {
      "epoch": 12.64,
      "learning_rate": 3.195937873357228e-05,
      "loss": 0.4861,
      "step": 790
    },
    {
      "epoch": 12.8,
      "learning_rate": 3.1660692951015537e-05,
      "loss": 0.4823,
      "step": 800
    },
    {
      "epoch": 12.96,
      "learning_rate": 3.136200716845878e-05,
      "loss": 0.4769,
      "step": 810
    },
    {
      "epoch": 12.99,
      "eval_accuracy": 0.842728297632469,
      "eval_f1": 0.8396494048281634,
      "eval_loss": 0.4639113247394562,
      "eval_precision": 0.8474916361770126,
      "eval_recall": 0.842728297632469,
      "eval_runtime": 14.4496,
      "eval_samples_per_second": 122.771,
      "eval_steps_per_second": 1.938,
      "step": 812
    },
    {
      "epoch": 13.12,
      "learning_rate": 3.106332138590203e-05,
      "loss": 0.4404,
      "step": 820
    },
    {
      "epoch": 13.28,
      "learning_rate": 3.0764635603345284e-05,
      "loss": 0.4541,
      "step": 830
    },
    {
      "epoch": 13.44,
      "learning_rate": 3.046594982078853e-05,
      "loss": 0.4797,
      "step": 840
    },
    {
      "epoch": 13.6,
      "learning_rate": 3.016726403823178e-05,
      "loss": 0.4431,
      "step": 850
    },
    {
      "epoch": 13.76,
      "learning_rate": 2.9868578255675032e-05,
      "loss": 0.4817,
      "step": 860
    },
    {
      "epoch": 13.92,
      "learning_rate": 2.9569892473118284e-05,
      "loss": 0.4804,
      "step": 870
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.8483652762119503,
      "eval_f1": 0.8461038057067607,
      "eval_loss": 0.4468301236629486,
      "eval_precision": 0.8498752163163035,
      "eval_recall": 0.8483652762119503,
      "eval_runtime": 14.522,
      "eval_samples_per_second": 122.16,
      "eval_steps_per_second": 1.928,
      "step": 875
    },
    {
      "epoch": 14.08,
      "learning_rate": 2.9271206690561532e-05,
      "loss": 0.4529,
      "step": 880
    },
    {
      "epoch": 14.24,
      "learning_rate": 2.897252090800478e-05,
      "loss": 0.4171,
      "step": 890
    },
    {
      "epoch": 14.4,
      "learning_rate": 2.8673835125448028e-05,
      "loss": 0.4416,
      "step": 900
    },
    {
      "epoch": 14.56,
      "learning_rate": 2.8375149342891276e-05,
      "loss": 0.4598,
      "step": 910
    },
    {
      "epoch": 14.72,
      "learning_rate": 2.807646356033453e-05,
      "loss": 0.4378,
      "step": 920
    },
    {
      "epoch": 14.88,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.4452,
      "step": 930
    },
    {
      "epoch": 14.99,
      "eval_accuracy": 0.8511837655016911,
      "eval_f1": 0.8504889104242102,
      "eval_loss": 0.44922059774398804,
      "eval_precision": 0.8521721769723261,
      "eval_recall": 0.8511837655016911,
      "eval_runtime": 14.5402,
      "eval_samples_per_second": 122.006,
      "eval_steps_per_second": 1.926,
      "step": 937
    },
    {
      "epoch": 15.04,
      "learning_rate": 2.747909199522103e-05,
      "loss": 0.432,
      "step": 940
    },
    {
      "epoch": 15.2,
      "learning_rate": 2.718040621266428e-05,
      "loss": 0.4106,
      "step": 950
    },
    {
      "epoch": 15.36,
      "learning_rate": 2.6881720430107527e-05,
      "loss": 0.4408,
      "step": 960
    },
    {
      "epoch": 15.52,
      "learning_rate": 2.6583034647550775e-05,
      "loss": 0.4328,
      "step": 970
    },
    {
      "epoch": 15.68,
      "learning_rate": 2.628434886499403e-05,
      "loss": 0.4125,
      "step": 980
    },
    {
      "epoch": 15.84,
      "learning_rate": 2.5985663082437278e-05,
      "loss": 0.4314,
      "step": 990
    },
    {
      "epoch": 16.0,
      "learning_rate": 2.5686977299880526e-05,
      "loss": 0.4283,
      "step": 1000
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.8432919954904171,
      "eval_f1": 0.8400560735292193,
      "eval_loss": 0.4659724235534668,
      "eval_precision": 0.8445731988875808,
      "eval_recall": 0.8432919954904171,
      "eval_runtime": 14.3743,
      "eval_samples_per_second": 123.414,
      "eval_steps_per_second": 1.948,
      "step": 1000
    },
    {
      "epoch": 16.16,
      "learning_rate": 2.5388291517323774e-05,
      "loss": 0.4255,
      "step": 1010
    },
    {
      "epoch": 16.32,
      "learning_rate": 2.5089605734767026e-05,
      "loss": 0.4008,
      "step": 1020
    },
    {
      "epoch": 16.48,
      "learning_rate": 2.4790919952210277e-05,
      "loss": 0.4192,
      "step": 1030
    },
    {
      "epoch": 16.64,
      "learning_rate": 2.4492234169653525e-05,
      "loss": 0.4387,
      "step": 1040
    },
    {
      "epoch": 16.8,
      "learning_rate": 2.4193548387096777e-05,
      "loss": 0.4191,
      "step": 1050
    },
    {
      "epoch": 16.96,
      "learning_rate": 2.3894862604540025e-05,
      "loss": 0.3788,
      "step": 1060
    },
    {
      "epoch": 16.99,
      "eval_accuracy": 0.8478015783540023,
      "eval_f1": 0.8443825980969434,
      "eval_loss": 0.46885567903518677,
      "eval_precision": 0.845414319624598,
      "eval_recall": 0.8478015783540023,
      "eval_runtime": 14.0279,
      "eval_samples_per_second": 126.462,
      "eval_steps_per_second": 1.996,
      "step": 1062
    },
    {
      "epoch": 17.12,
      "learning_rate": 2.3596176821983273e-05,
      "loss": 0.4113,
      "step": 1070
    },
    {
      "epoch": 17.28,
      "learning_rate": 2.3297491039426525e-05,
      "loss": 0.421,
      "step": 1080
    },
    {
      "epoch": 17.44,
      "learning_rate": 2.2998805256869773e-05,
      "loss": 0.4404,
      "step": 1090
    },
    {
      "epoch": 17.6,
      "learning_rate": 2.270011947431302e-05,
      "loss": 0.429,
      "step": 1100
    },
    {
      "epoch": 17.76,
      "learning_rate": 2.2401433691756272e-05,
      "loss": 0.4113,
      "step": 1110
    },
    {
      "epoch": 17.92,
      "learning_rate": 2.2102747909199524e-05,
      "loss": 0.41,
      "step": 1120
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.850620067643743,
      "eval_f1": 0.8479598566035595,
      "eval_loss": 0.45433101058006287,
      "eval_precision": 0.8501923040015001,
      "eval_recall": 0.850620067643743,
      "eval_runtime": 14.8781,
      "eval_samples_per_second": 119.235,
      "eval_steps_per_second": 1.882,
      "step": 1125
    },
    {
      "epoch": 18.08,
      "learning_rate": 2.1804062126642775e-05,
      "loss": 0.4038,
      "step": 1130
    },
    {
      "epoch": 18.24,
      "learning_rate": 2.1505376344086024e-05,
      "loss": 0.3832,
      "step": 1140
    },
    {
      "epoch": 18.4,
      "learning_rate": 2.120669056152927e-05,
      "loss": 0.3845,
      "step": 1150
    },
    {
      "epoch": 18.56,
      "learning_rate": 2.0908004778972523e-05,
      "loss": 0.3929,
      "step": 1160
    },
    {
      "epoch": 18.72,
      "learning_rate": 2.060931899641577e-05,
      "loss": 0.3769,
      "step": 1170
    },
    {
      "epoch": 18.88,
      "learning_rate": 2.031063321385902e-05,
      "loss": 0.4007,
      "step": 1180
    },
    {
      "epoch": 18.99,
      "eval_accuracy": 0.8478015783540023,
      "eval_f1": 0.8454868202047309,
      "eval_loss": 0.47658494114875793,
      "eval_precision": 0.851146998899857,
      "eval_recall": 0.8478015783540023,
      "eval_runtime": 14.833,
      "eval_samples_per_second": 119.598,
      "eval_steps_per_second": 1.888,
      "step": 1187
    },
    {
      "epoch": 19.04,
      "learning_rate": 2.001194743130227e-05,
      "loss": 0.403,
      "step": 1190
    },
    {
      "epoch": 19.2,
      "learning_rate": 1.9713261648745522e-05,
      "loss": 0.3899,
      "step": 1200
    },
    {
      "epoch": 19.36,
      "learning_rate": 1.941457586618877e-05,
      "loss": 0.3839,
      "step": 1210
    },
    {
      "epoch": 19.52,
      "learning_rate": 1.9115890083632022e-05,
      "loss": 0.3703,
      "step": 1220
    },
    {
      "epoch": 19.68,
      "learning_rate": 1.881720430107527e-05,
      "loss": 0.3747,
      "step": 1230
    },
    {
      "epoch": 19.84,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.3995,
      "step": 1240
    },
    {
      "epoch": 20.0,
      "learning_rate": 1.821983273596177e-05,
      "loss": 0.406,
      "step": 1250
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.8478015783540023,
      "eval_f1": 0.8444369977344951,
      "eval_loss": 0.47164490818977356,
      "eval_precision": 0.8474409018042954,
      "eval_recall": 0.8478015783540023,
      "eval_runtime": 14.4569,
      "eval_samples_per_second": 122.709,
      "eval_steps_per_second": 1.937,
      "step": 1250
    },
    {
      "epoch": 20.16,
      "learning_rate": 1.7921146953405018e-05,
      "loss": 0.3808,
      "step": 1260
    },
    {
      "epoch": 20.32,
      "learning_rate": 1.7622461170848266e-05,
      "loss": 0.4002,
      "step": 1270
    },
    {
      "epoch": 20.48,
      "learning_rate": 1.7323775388291518e-05,
      "loss": 0.3679,
      "step": 1280
    },
    {
      "epoch": 20.64,
      "learning_rate": 1.702508960573477e-05,
      "loss": 0.369,
      "step": 1290
    },
    {
      "epoch": 20.8,
      "learning_rate": 1.6726403823178017e-05,
      "loss": 0.3875,
      "step": 1300
    },
    {
      "epoch": 20.96,
      "learning_rate": 1.642771804062127e-05,
      "loss": 0.3777,
      "step": 1310
    },
    {
      "epoch": 20.99,
      "eval_accuracy": 0.8455467869222097,
      "eval_f1": 0.8429662578800389,
      "eval_loss": 0.5026230812072754,
      "eval_precision": 0.8453980146551628,
      "eval_recall": 0.8455467869222097,
      "eval_runtime": 14.6724,
      "eval_samples_per_second": 120.907,
      "eval_steps_per_second": 1.908,
      "step": 1312
    },
    {
      "epoch": 21.12,
      "learning_rate": 1.6129032258064517e-05,
      "loss": 0.3551,
      "step": 1320
    },
    {
      "epoch": 21.28,
      "learning_rate": 1.5830346475507768e-05,
      "loss": 0.3637,
      "step": 1330
    },
    {
      "epoch": 21.44,
      "learning_rate": 1.5531660692951016e-05,
      "loss": 0.3754,
      "step": 1340
    },
    {
      "epoch": 21.6,
      "learning_rate": 1.5232974910394265e-05,
      "loss": 0.3745,
      "step": 1350
    },
    {
      "epoch": 21.76,
      "learning_rate": 1.4934289127837516e-05,
      "loss": 0.3839,
      "step": 1360
    },
    {
      "epoch": 21.92,
      "learning_rate": 1.4635603345280766e-05,
      "loss": 0.3972,
      "step": 1370
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.8393461104847801,
      "eval_f1": 0.8371232795500526,
      "eval_loss": 0.5107545852661133,
      "eval_precision": 0.8402037485528469,
      "eval_recall": 0.8393461104847801,
      "eval_runtime": 14.7507,
      "eval_samples_per_second": 120.265,
      "eval_steps_per_second": 1.898,
      "step": 1375
    },
    {
      "epoch": 22.08,
      "learning_rate": 1.4336917562724014e-05,
      "loss": 0.3666,
      "step": 1380
    },
    {
      "epoch": 22.24,
      "learning_rate": 1.4038231780167265e-05,
      "loss": 0.3452,
      "step": 1390
    },
    {
      "epoch": 22.4,
      "learning_rate": 1.3739545997610515e-05,
      "loss": 0.3539,
      "step": 1400
    },
    {
      "epoch": 22.56,
      "learning_rate": 1.3440860215053763e-05,
      "loss": 0.3687,
      "step": 1410
    },
    {
      "epoch": 22.72,
      "learning_rate": 1.3142174432497015e-05,
      "loss": 0.3703,
      "step": 1420
    },
    {
      "epoch": 22.88,
      "learning_rate": 1.2843488649940263e-05,
      "loss": 0.3665,
      "step": 1430
    },
    {
      "epoch": 22.99,
      "eval_accuracy": 0.8489289740698985,
      "eval_f1": 0.8474228146343941,
      "eval_loss": 0.49335137009620667,
      "eval_precision": 0.8498174041997181,
      "eval_recall": 0.8489289740698985,
      "eval_runtime": 14.7164,
      "eval_samples_per_second": 120.546,
      "eval_steps_per_second": 1.903,
      "step": 1437
    },
    {
      "epoch": 23.04,
      "learning_rate": 1.2544802867383513e-05,
      "loss": 0.3688,
      "step": 1440
    },
    {
      "epoch": 23.2,
      "learning_rate": 1.2246117084826763e-05,
      "loss": 0.3567,
      "step": 1450
    },
    {
      "epoch": 23.36,
      "learning_rate": 1.1947431302270013e-05,
      "loss": 0.3551,
      "step": 1460
    },
    {
      "epoch": 23.52,
      "learning_rate": 1.1648745519713262e-05,
      "loss": 0.355,
      "step": 1470
    },
    {
      "epoch": 23.68,
      "learning_rate": 1.135005973715651e-05,
      "loss": 0.3761,
      "step": 1480
    },
    {
      "epoch": 23.84,
      "learning_rate": 1.1051373954599762e-05,
      "loss": 0.3217,
      "step": 1490
    },
    {
      "epoch": 24.0,
      "learning_rate": 1.0752688172043012e-05,
      "loss": 0.3569,
      "step": 1500
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8494926719278467,
      "eval_f1": 0.8477787837150598,
      "eval_loss": 0.4988711178302765,
      "eval_precision": 0.8494857087385842,
      "eval_recall": 0.8494926719278467,
      "eval_runtime": 14.565,
      "eval_samples_per_second": 121.798,
      "eval_steps_per_second": 1.922,
      "step": 1500
    },
    {
      "epoch": 24.16,
      "learning_rate": 1.0454002389486262e-05,
      "loss": 0.3674,
      "step": 1510
    },
    {
      "epoch": 24.32,
      "learning_rate": 1.015531660692951e-05,
      "loss": 0.3662,
      "step": 1520
    },
    {
      "epoch": 24.48,
      "learning_rate": 9.856630824372761e-06,
      "loss": 0.3618,
      "step": 1530
    },
    {
      "epoch": 24.64,
      "learning_rate": 9.557945041816011e-06,
      "loss": 0.3528,
      "step": 1540
    },
    {
      "epoch": 24.8,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.3554,
      "step": 1550
    },
    {
      "epoch": 24.96,
      "learning_rate": 8.960573476702509e-06,
      "loss": 0.3735,
      "step": 1560
    },
    {
      "epoch": 24.99,
      "eval_accuracy": 0.8494926719278467,
      "eval_f1": 0.846781722291396,
      "eval_loss": 0.4918430745601654,
      "eval_precision": 0.8468491527163158,
      "eval_recall": 0.8494926719278467,
      "eval_runtime": 14.7111,
      "eval_samples_per_second": 120.589,
      "eval_steps_per_second": 1.903,
      "step": 1562
    },
    {
      "epoch": 25.12,
      "learning_rate": 8.661887694145759e-06,
      "loss": 0.3357,
      "step": 1570
    },
    {
      "epoch": 25.28,
      "learning_rate": 8.363201911589009e-06,
      "loss": 0.331,
      "step": 1580
    },
    {
      "epoch": 25.44,
      "learning_rate": 8.064516129032258e-06,
      "loss": 0.3452,
      "step": 1590
    },
    {
      "epoch": 25.6,
      "learning_rate": 7.765830346475508e-06,
      "loss": 0.3573,
      "step": 1600
    },
    {
      "epoch": 25.76,
      "learning_rate": 7.467144563918758e-06,
      "loss": 0.338,
      "step": 1610
    },
    {
      "epoch": 25.92,
      "learning_rate": 7.168458781362007e-06,
      "loss": 0.3301,
      "step": 1620
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.8511837655016911,
      "eval_f1": 0.8488476738406706,
      "eval_loss": 0.49265769124031067,
      "eval_precision": 0.8512031142203579,
      "eval_recall": 0.8511837655016911,
      "eval_runtime": 14.4849,
      "eval_samples_per_second": 122.473,
      "eval_steps_per_second": 1.933,
      "step": 1625
    },
    {
      "epoch": 26.08,
      "learning_rate": 6.869772998805258e-06,
      "loss": 0.3509,
      "step": 1630
    },
    {
      "epoch": 26.24,
      "learning_rate": 6.5710872162485075e-06,
      "loss": 0.3469,
      "step": 1640
    },
    {
      "epoch": 26.4,
      "learning_rate": 6.2724014336917564e-06,
      "loss": 0.3472,
      "step": 1650
    },
    {
      "epoch": 26.56,
      "learning_rate": 5.973715651135006e-06,
      "loss": 0.3277,
      "step": 1660
    },
    {
      "epoch": 26.72,
      "learning_rate": 5.675029868578255e-06,
      "loss": 0.3484,
      "step": 1670
    },
    {
      "epoch": 26.88,
      "learning_rate": 5.376344086021506e-06,
      "loss": 0.3438,
      "step": 1680
    },
    {
      "epoch": 26.99,
      "eval_accuracy": 0.8540022547914318,
      "eval_f1": 0.8520059264195492,
      "eval_loss": 0.48286083340644836,
      "eval_precision": 0.852942098455949,
      "eval_recall": 0.8540022547914318,
      "eval_runtime": 14.8521,
      "eval_samples_per_second": 119.444,
      "eval_steps_per_second": 1.885,
      "step": 1687
    },
    {
      "epoch": 27.04,
      "learning_rate": 5.077658303464755e-06,
      "loss": 0.335,
      "step": 1690
    },
    {
      "epoch": 27.2,
      "learning_rate": 4.7789725209080055e-06,
      "loss": 0.3292,
      "step": 1700
    },
    {
      "epoch": 27.36,
      "learning_rate": 4.4802867383512545e-06,
      "loss": 0.3691,
      "step": 1710
    },
    {
      "epoch": 27.52,
      "learning_rate": 4.181600955794504e-06,
      "loss": 0.3279,
      "step": 1720
    },
    {
      "epoch": 27.68,
      "learning_rate": 3.882915173237754e-06,
      "loss": 0.3102,
      "step": 1730
    },
    {
      "epoch": 27.84,
      "learning_rate": 3.5842293906810035e-06,
      "loss": 0.3523,
      "step": 1740
    },
    {
      "epoch": 28.0,
      "learning_rate": 3.2855436081242537e-06,
      "loss": 0.3553,
      "step": 1750
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8540022547914318,
      "eval_f1": 0.8512198800317611,
      "eval_loss": 0.4934614896774292,
      "eval_precision": 0.8529596044039405,
      "eval_recall": 0.8540022547914318,
      "eval_runtime": 14.3586,
      "eval_samples_per_second": 123.55,
      "eval_steps_per_second": 1.95,
      "step": 1750
    },
    {
      "epoch": 28.16,
      "learning_rate": 2.986857825567503e-06,
      "loss": 0.3589,
      "step": 1760
    },
    {
      "epoch": 28.32,
      "learning_rate": 2.688172043010753e-06,
      "loss": 0.3442,
      "step": 1770
    },
    {
      "epoch": 28.48,
      "learning_rate": 2.3894862604540028e-06,
      "loss": 0.3237,
      "step": 1780
    },
    {
      "epoch": 28.64,
      "learning_rate": 2.090800477897252e-06,
      "loss": 0.3341,
      "step": 1790
    },
    {
      "epoch": 28.8,
      "learning_rate": 1.7921146953405017e-06,
      "loss": 0.334,
      "step": 1800
    },
    {
      "epoch": 28.96,
      "learning_rate": 1.4934289127837516e-06,
      "loss": 0.3312,
      "step": 1810
    },
    {
      "epoch": 28.99,
      "eval_accuracy": 0.8517474633596392,
      "eval_f1": 0.8491345535989786,
      "eval_loss": 0.4881838858127594,
      "eval_precision": 0.8509432522202501,
      "eval_recall": 0.8517474633596392,
      "eval_runtime": 14.6798,
      "eval_samples_per_second": 120.846,
      "eval_steps_per_second": 1.907,
      "step": 1812
    },
    {
      "epoch": 29.12,
      "learning_rate": 1.1947431302270014e-06,
      "loss": 0.3403,
      "step": 1820
    },
    {
      "epoch": 29.28,
      "learning_rate": 8.960573476702509e-07,
      "loss": 0.3236,
      "step": 1830
    },
    {
      "epoch": 29.44,
      "learning_rate": 5.973715651135007e-07,
      "loss": 0.3188,
      "step": 1840
    },
    {
      "epoch": 29.6,
      "learning_rate": 2.9868578255675034e-07,
      "loss": 0.3273,
      "step": 1850
    },
    {
      "epoch": 29.76,
      "learning_rate": 0.0,
      "loss": 0.3319,
      "step": 1860
    },
    {
      "epoch": 29.76,
      "eval_accuracy": 0.8517474633596392,
      "eval_f1": 0.8496502752285078,
      "eval_loss": 0.4875507056713104,
      "eval_precision": 0.8515975357479637,
      "eval_recall": 0.8517474633596392,
      "eval_runtime": 14.3702,
      "eval_samples_per_second": 123.45,
      "eval_steps_per_second": 1.948,
      "step": 1860
    },
    {
      "epoch": 29.76,
      "step": 1860,
      "total_flos": 4.870114457320843e+19,
      "train_loss": 0.5392252865658012,
      "train_runtime": 7880.761,
      "train_samples_per_second": 60.763,
      "train_steps_per_second": 0.236
    }
  ],
  "logging_steps": 10,
  "max_steps": 1860,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 4.870114457320843e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}