{
  "best_metric": 0.8461104847801578,
  "best_model_checkpoint": "convnextv2-atto-1k-224-finetuned-galaxy10-decals/checkpoint-1812",
  "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": 2.4903,
      "step": 10
    },
    {
      "epoch": 0.32,
      "learning_rate": 5.376344086021506e-06,
      "loss": 2.3411,
      "step": 20
    },
    {
      "epoch": 0.48,
      "learning_rate": 8.064516129032258e-06,
      "loss": 2.2567,
      "step": 30
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.0752688172043012e-05,
      "loss": 2.1992,
      "step": 40
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.3440860215053763e-05,
      "loss": 2.1353,
      "step": 50
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.6129032258064517e-05,
      "loss": 2.0062,
      "step": 60
    },
    {
      "epoch": 0.99,
      "eval_accuracy": 0.34498308906426156,
      "eval_f1": 0.29563135350373826,
      "eval_loss": 1.8928039073944092,
      "eval_precision": 0.3432438883226705,
      "eval_recall": 0.34498308906426156,
      "eval_runtime": 12.532,
      "eval_samples_per_second": 141.557,
      "eval_steps_per_second": 2.234,
      "step": 62
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.881720430107527e-05,
      "loss": 1.8524,
      "step": 70
    },
    {
      "epoch": 1.28,
      "learning_rate": 2.1505376344086024e-05,
      "loss": 1.6666,
      "step": 80
    },
    {
      "epoch": 1.44,
      "learning_rate": 2.4193548387096777e-05,
      "loss": 1.4856,
      "step": 90
    },
    {
      "epoch": 1.6,
      "learning_rate": 2.6881720430107527e-05,
      "loss": 1.3611,
      "step": 100
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.9569892473118284e-05,
      "loss": 1.2277,
      "step": 110
    },
    {
      "epoch": 1.92,
      "learning_rate": 3.2258064516129034e-05,
      "loss": 1.1323,
      "step": 120
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6589627959413754,
      "eval_f1": 0.6398866800037653,
      "eval_loss": 1.0025697946548462,
      "eval_precision": 0.6633569178312325,
      "eval_recall": 0.6589627959413754,
      "eval_runtime": 11.198,
      "eval_samples_per_second": 158.421,
      "eval_steps_per_second": 2.5,
      "step": 125
    },
    {
      "epoch": 2.08,
      "learning_rate": 3.494623655913979e-05,
      "loss": 1.0653,
      "step": 130
    },
    {
      "epoch": 2.24,
      "learning_rate": 3.763440860215054e-05,
      "loss": 0.9858,
      "step": 140
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.032258064516129e-05,
      "loss": 0.93,
      "step": 150
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.301075268817205e-05,
      "loss": 0.9381,
      "step": 160
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.56989247311828e-05,
      "loss": 0.9113,
      "step": 170
    },
    {
      "epoch": 2.88,
      "learning_rate": 4.8387096774193554e-05,
      "loss": 0.8977,
      "step": 180
    },
    {
      "epoch": 2.99,
      "eval_accuracy": 0.7485907553551296,
      "eval_f1": 0.7398969485570298,
      "eval_loss": 0.7348349094390869,
      "eval_precision": 0.7414739028478952,
      "eval_recall": 0.7485907553551296,
      "eval_runtime": 11.2555,
      "eval_samples_per_second": 157.612,
      "eval_steps_per_second": 2.488,
      "step": 187
    },
    {
      "epoch": 3.04,
      "learning_rate": 4.98805256869773e-05,
      "loss": 0.8998,
      "step": 190
    },
    {
      "epoch": 3.2,
      "learning_rate": 4.9581839904420555e-05,
      "loss": 0.8191,
      "step": 200
    },
    {
      "epoch": 3.36,
      "learning_rate": 4.92831541218638e-05,
      "loss": 0.8119,
      "step": 210
    },
    {
      "epoch": 3.52,
      "learning_rate": 4.898446833930705e-05,
      "loss": 0.8158,
      "step": 220
    },
    {
      "epoch": 3.68,
      "learning_rate": 4.86857825567503e-05,
      "loss": 0.7582,
      "step": 230
    },
    {
      "epoch": 3.84,
      "learning_rate": 4.8387096774193554e-05,
      "loss": 0.7741,
      "step": 240
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.80884109916368e-05,
      "loss": 0.7119,
      "step": 250
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7891770011273957,
      "eval_f1": 0.7770214066706361,
      "eval_loss": 0.6394641399383545,
      "eval_precision": 0.7878433104843552,
      "eval_recall": 0.7891770011273957,
      "eval_runtime": 11.2183,
      "eval_samples_per_second": 158.135,
      "eval_steps_per_second": 2.496,
      "step": 250
    },
    {
      "epoch": 4.16,
      "learning_rate": 4.778972520908005e-05,
      "loss": 0.7633,
      "step": 260
    },
    {
      "epoch": 4.32,
      "learning_rate": 4.74910394265233e-05,
      "loss": 0.7013,
      "step": 270
    },
    {
      "epoch": 4.48,
      "learning_rate": 4.7192353643966546e-05,
      "loss": 0.7011,
      "step": 280
    },
    {
      "epoch": 4.64,
      "learning_rate": 4.6893667861409805e-05,
      "loss": 0.6941,
      "step": 290
    },
    {
      "epoch": 4.8,
      "learning_rate": 4.659498207885305e-05,
      "loss": 0.7186,
      "step": 300
    },
    {
      "epoch": 4.96,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.6393,
      "step": 310
    },
    {
      "epoch": 4.99,
      "eval_accuracy": 0.7970687711386697,
      "eval_f1": 0.7915391017654148,
      "eval_loss": 0.5801244378089905,
      "eval_precision": 0.7916338810961826,
      "eval_recall": 0.7970687711386697,
      "eval_runtime": 10.9886,
      "eval_samples_per_second": 161.44,
      "eval_steps_per_second": 2.548,
      "step": 312
    },
    {
      "epoch": 5.12,
      "learning_rate": 4.5997610513739546e-05,
      "loss": 0.6775,
      "step": 320
    },
    {
      "epoch": 5.28,
      "learning_rate": 4.56989247311828e-05,
      "loss": 0.6901,
      "step": 330
    },
    {
      "epoch": 5.44,
      "learning_rate": 4.540023894862604e-05,
      "loss": 0.6794,
      "step": 340
    },
    {
      "epoch": 5.6,
      "learning_rate": 4.51015531660693e-05,
      "loss": 0.6857,
      "step": 350
    },
    {
      "epoch": 5.76,
      "learning_rate": 4.4802867383512545e-05,
      "loss": 0.617,
      "step": 360
    },
    {
      "epoch": 5.92,
      "learning_rate": 4.4504181600955796e-05,
      "loss": 0.6463,
      "step": 370
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7976324689966178,
      "eval_f1": 0.790881634631752,
      "eval_loss": 0.5957704782485962,
      "eval_precision": 0.8146922179391244,
      "eval_recall": 0.7976324689966178,
      "eval_runtime": 11.2198,
      "eval_samples_per_second": 158.113,
      "eval_steps_per_second": 2.496,
      "step": 375
    },
    {
      "epoch": 6.08,
      "learning_rate": 4.420549581839905e-05,
      "loss": 0.6278,
      "step": 380
    },
    {
      "epoch": 6.24,
      "learning_rate": 4.390681003584229e-05,
      "loss": 0.649,
      "step": 390
    },
    {
      "epoch": 6.4,
      "learning_rate": 4.360812425328555e-05,
      "loss": 0.6137,
      "step": 400
    },
    {
      "epoch": 6.56,
      "learning_rate": 4.3309438470728796e-05,
      "loss": 0.625,
      "step": 410
    },
    {
      "epoch": 6.72,
      "learning_rate": 4.301075268817205e-05,
      "loss": 0.6123,
      "step": 420
    },
    {
      "epoch": 6.88,
      "learning_rate": 4.271206690561529e-05,
      "loss": 0.6197,
      "step": 430
    },
    {
      "epoch": 6.99,
      "eval_accuracy": 0.8151071025930101,
      "eval_f1": 0.8112332556430505,
      "eval_loss": 0.536342203617096,
      "eval_precision": 0.8119110502125383,
      "eval_recall": 0.8151071025930101,
      "eval_runtime": 11.2233,
      "eval_samples_per_second": 158.064,
      "eval_steps_per_second": 2.495,
      "step": 437
    },
    {
      "epoch": 7.04,
      "learning_rate": 4.241338112305854e-05,
      "loss": 0.6571,
      "step": 440
    },
    {
      "epoch": 7.2,
      "learning_rate": 4.2114695340501795e-05,
      "loss": 0.6315,
      "step": 450
    },
    {
      "epoch": 7.36,
      "learning_rate": 4.1816009557945046e-05,
      "loss": 0.5868,
      "step": 460
    },
    {
      "epoch": 7.52,
      "learning_rate": 4.15173237753883e-05,
      "loss": 0.6322,
      "step": 470
    },
    {
      "epoch": 7.68,
      "learning_rate": 4.121863799283154e-05,
      "loss": 0.5916,
      "step": 480
    },
    {
      "epoch": 7.84,
      "learning_rate": 4.0919952210274794e-05,
      "loss": 0.601,
      "step": 490
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.062126642771804e-05,
      "loss": 0.5779,
      "step": 500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8207440811724915,
      "eval_f1": 0.8184693186543313,
      "eval_loss": 0.5275559425354004,
      "eval_precision": 0.8204593714710395,
      "eval_recall": 0.8207440811724915,
      "eval_runtime": 11.2254,
      "eval_samples_per_second": 158.035,
      "eval_steps_per_second": 2.494,
      "step": 500
    },
    {
      "epoch": 8.16,
      "learning_rate": 4.032258064516129e-05,
      "loss": 0.5683,
      "step": 510
    },
    {
      "epoch": 8.32,
      "learning_rate": 4.002389486260454e-05,
      "loss": 0.5751,
      "step": 520
    },
    {
      "epoch": 8.48,
      "learning_rate": 3.972520908004779e-05,
      "loss": 0.6121,
      "step": 530
    },
    {
      "epoch": 8.64,
      "learning_rate": 3.9426523297491045e-05,
      "loss": 0.5769,
      "step": 540
    },
    {
      "epoch": 8.8,
      "learning_rate": 3.912783751493429e-05,
      "loss": 0.5958,
      "step": 550
    },
    {
      "epoch": 8.96,
      "learning_rate": 3.882915173237754e-05,
      "loss": 0.5841,
      "step": 560
    },
    {
      "epoch": 8.99,
      "eval_accuracy": 0.818489289740699,
      "eval_f1": 0.8156685934925514,
      "eval_loss": 0.5197471380233765,
      "eval_precision": 0.8203151387753229,
      "eval_recall": 0.818489289740699,
      "eval_runtime": 11.2314,
      "eval_samples_per_second": 157.95,
      "eval_steps_per_second": 2.493,
      "step": 562
    },
    {
      "epoch": 9.12,
      "learning_rate": 3.8530465949820786e-05,
      "loss": 0.5818,
      "step": 570
    },
    {
      "epoch": 9.28,
      "learning_rate": 3.8231780167264044e-05,
      "loss": 0.5606,
      "step": 580
    },
    {
      "epoch": 9.44,
      "learning_rate": 3.793309438470729e-05,
      "loss": 0.5584,
      "step": 590
    },
    {
      "epoch": 9.6,
      "learning_rate": 3.763440860215054e-05,
      "loss": 0.587,
      "step": 600
    },
    {
      "epoch": 9.76,
      "learning_rate": 3.7335722819593785e-05,
      "loss": 0.5456,
      "step": 610
    },
    {
      "epoch": 9.92,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.5597,
      "step": 620
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8252536640360767,
      "eval_f1": 0.8192672895248532,
      "eval_loss": 0.5025486946105957,
      "eval_precision": 0.8192007741568054,
      "eval_recall": 0.8252536640360767,
      "eval_runtime": 11.238,
      "eval_samples_per_second": 157.857,
      "eval_steps_per_second": 2.492,
      "step": 625
    },
    {
      "epoch": 10.08,
      "learning_rate": 3.673835125448029e-05,
      "loss": 0.5442,
      "step": 630
    },
    {
      "epoch": 10.24,
      "learning_rate": 3.643966547192354e-05,
      "loss": 0.5567,
      "step": 640
    },
    {
      "epoch": 10.4,
      "learning_rate": 3.614097968936679e-05,
      "loss": 0.5823,
      "step": 650
    },
    {
      "epoch": 10.56,
      "learning_rate": 3.5842293906810036e-05,
      "loss": 0.5602,
      "step": 660
    },
    {
      "epoch": 10.72,
      "learning_rate": 3.554360812425329e-05,
      "loss": 0.5266,
      "step": 670
    },
    {
      "epoch": 10.88,
      "learning_rate": 3.524492234169653e-05,
      "loss": 0.5437,
      "step": 680
    },
    {
      "epoch": 10.99,
      "eval_accuracy": 0.830890642615558,
      "eval_f1": 0.8296494477664437,
      "eval_loss": 0.4912206530570984,
      "eval_precision": 0.829504327675249,
      "eval_recall": 0.830890642615558,
      "eval_runtime": 11.2851,
      "eval_samples_per_second": 157.198,
      "eval_steps_per_second": 2.481,
      "step": 687
    },
    {
      "epoch": 11.04,
      "learning_rate": 3.494623655913979e-05,
      "loss": 0.5224,
      "step": 690
    },
    {
      "epoch": 11.2,
      "learning_rate": 3.4647550776583035e-05,
      "loss": 0.5281,
      "step": 700
    },
    {
      "epoch": 11.36,
      "learning_rate": 3.4348864994026287e-05,
      "loss": 0.5002,
      "step": 710
    },
    {
      "epoch": 11.52,
      "learning_rate": 3.405017921146954e-05,
      "loss": 0.5224,
      "step": 720
    },
    {
      "epoch": 11.68,
      "learning_rate": 3.375149342891278e-05,
      "loss": 0.5418,
      "step": 730
    },
    {
      "epoch": 11.84,
      "learning_rate": 3.3452807646356034e-05,
      "loss": 0.5244,
      "step": 740
    },
    {
      "epoch": 12.0,
      "learning_rate": 3.3154121863799286e-05,
      "loss": 0.5242,
      "step": 750
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.8275084554678692,
      "eval_f1": 0.8244600176027497,
      "eval_loss": 0.5000930428504944,
      "eval_precision": 0.8302654360483025,
      "eval_recall": 0.8275084554678692,
      "eval_runtime": 11.2547,
      "eval_samples_per_second": 157.623,
      "eval_steps_per_second": 2.488,
      "step": 750
    },
    {
      "epoch": 12.16,
      "learning_rate": 3.285543608124254e-05,
      "loss": 0.5141,
      "step": 760
    },
    {
      "epoch": 12.32,
      "learning_rate": 3.255675029868578e-05,
      "loss": 0.5134,
      "step": 770
    },
    {
      "epoch": 12.48,
      "learning_rate": 3.2258064516129034e-05,
      "loss": 0.5177,
      "step": 780
    },
    {
      "epoch": 12.64,
      "learning_rate": 3.195937873357228e-05,
      "loss": 0.5335,
      "step": 790
    },
    {
      "epoch": 12.8,
      "learning_rate": 3.1660692951015537e-05,
      "loss": 0.5486,
      "step": 800
    },
    {
      "epoch": 12.96,
      "learning_rate": 3.136200716845878e-05,
      "loss": 0.5029,
      "step": 810
    },
    {
      "epoch": 12.99,
      "eval_accuracy": 0.8241262683201804,
      "eval_f1": 0.8208074602527222,
      "eval_loss": 0.5075457692146301,
      "eval_precision": 0.8227574273338397,
      "eval_recall": 0.8241262683201804,
      "eval_runtime": 11.2289,
      "eval_samples_per_second": 157.985,
      "eval_steps_per_second": 2.494,
      "step": 812
    },
    {
      "epoch": 13.12,
      "learning_rate": 3.106332138590203e-05,
      "loss": 0.4764,
      "step": 820
    },
    {
      "epoch": 13.28,
      "learning_rate": 3.0764635603345284e-05,
      "loss": 0.5028,
      "step": 830
    },
    {
      "epoch": 13.44,
      "learning_rate": 3.046594982078853e-05,
      "loss": 0.5302,
      "step": 840
    },
    {
      "epoch": 13.6,
      "learning_rate": 3.016726403823178e-05,
      "loss": 0.4924,
      "step": 850
    },
    {
      "epoch": 13.76,
      "learning_rate": 2.9868578255675032e-05,
      "loss": 0.5176,
      "step": 860
    },
    {
      "epoch": 13.92,
      "learning_rate": 2.9569892473118284e-05,
      "loss": 0.5396,
      "step": 870
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.8393461104847801,
      "eval_f1": 0.8370660079844187,
      "eval_loss": 0.47840070724487305,
      "eval_precision": 0.839523767512895,
      "eval_recall": 0.8393461104847801,
      "eval_runtime": 11.2787,
      "eval_samples_per_second": 157.288,
      "eval_steps_per_second": 2.483,
      "step": 875
    },
    {
      "epoch": 14.08,
      "learning_rate": 2.9271206690561532e-05,
      "loss": 0.5004,
      "step": 880
    },
    {
      "epoch": 14.24,
      "learning_rate": 2.897252090800478e-05,
      "loss": 0.4791,
      "step": 890
    },
    {
      "epoch": 14.4,
      "learning_rate": 2.8673835125448028e-05,
      "loss": 0.486,
      "step": 900
    },
    {
      "epoch": 14.56,
      "learning_rate": 2.8375149342891276e-05,
      "loss": 0.5088,
      "step": 910
    },
    {
      "epoch": 14.72,
      "learning_rate": 2.807646356033453e-05,
      "loss": 0.503,
      "step": 920
    },
    {
      "epoch": 14.88,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.4746,
      "step": 930
    },
    {
      "epoch": 14.99,
      "eval_accuracy": 0.8331454340473506,
      "eval_f1": 0.8316592303256621,
      "eval_loss": 0.47273167967796326,
      "eval_precision": 0.8317737161543765,
      "eval_recall": 0.8331454340473506,
      "eval_runtime": 10.9831,
      "eval_samples_per_second": 161.521,
      "eval_steps_per_second": 2.549,
      "step": 937
    },
    {
      "epoch": 15.04,
      "learning_rate": 2.747909199522103e-05,
      "loss": 0.4829,
      "step": 940
    },
    {
      "epoch": 15.2,
      "learning_rate": 2.718040621266428e-05,
      "loss": 0.4626,
      "step": 950
    },
    {
      "epoch": 15.36,
      "learning_rate": 2.6881720430107527e-05,
      "loss": 0.4975,
      "step": 960
    },
    {
      "epoch": 15.52,
      "learning_rate": 2.6583034647550775e-05,
      "loss": 0.497,
      "step": 970
    },
    {
      "epoch": 15.68,
      "learning_rate": 2.628434886499403e-05,
      "loss": 0.4683,
      "step": 980
    },
    {
      "epoch": 15.84,
      "learning_rate": 2.5985663082437278e-05,
      "loss": 0.5085,
      "step": 990
    },
    {
      "epoch": 16.0,
      "learning_rate": 2.5686977299880526e-05,
      "loss": 0.4786,
      "step": 1000
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.8331454340473506,
      "eval_f1": 0.8299790015697147,
      "eval_loss": 0.4856169521808624,
      "eval_precision": 0.8308129653259497,
      "eval_recall": 0.8331454340473506,
      "eval_runtime": 11.2364,
      "eval_samples_per_second": 157.879,
      "eval_steps_per_second": 2.492,
      "step": 1000
    },
    {
      "epoch": 16.16,
      "learning_rate": 2.5388291517323774e-05,
      "loss": 0.4838,
      "step": 1010
    },
    {
      "epoch": 16.32,
      "learning_rate": 2.5089605734767026e-05,
      "loss": 0.4558,
      "step": 1020
    },
    {
      "epoch": 16.48,
      "learning_rate": 2.4790919952210277e-05,
      "loss": 0.4624,
      "step": 1030
    },
    {
      "epoch": 16.64,
      "learning_rate": 2.4492234169653525e-05,
      "loss": 0.4962,
      "step": 1040
    },
    {
      "epoch": 16.8,
      "learning_rate": 2.4193548387096777e-05,
      "loss": 0.4828,
      "step": 1050
    },
    {
      "epoch": 16.96,
      "learning_rate": 2.3894862604540025e-05,
      "loss": 0.4338,
      "step": 1060
    },
    {
      "epoch": 16.99,
      "eval_accuracy": 0.8337091319052987,
      "eval_f1": 0.8309204457109294,
      "eval_loss": 0.4883783459663391,
      "eval_precision": 0.833250336485018,
      "eval_recall": 0.8337091319052987,
      "eval_runtime": 11.2851,
      "eval_samples_per_second": 157.198,
      "eval_steps_per_second": 2.481,
      "step": 1062
    },
    {
      "epoch": 17.12,
      "learning_rate": 2.3596176821983273e-05,
      "loss": 0.4508,
      "step": 1070
    },
    {
      "epoch": 17.28,
      "learning_rate": 2.3297491039426525e-05,
      "loss": 0.4789,
      "step": 1080
    },
    {
      "epoch": 17.44,
      "learning_rate": 2.2998805256869773e-05,
      "loss": 0.4916,
      "step": 1090
    },
    {
      "epoch": 17.6,
      "learning_rate": 2.270011947431302e-05,
      "loss": 0.4701,
      "step": 1100
    },
    {
      "epoch": 17.76,
      "learning_rate": 2.2401433691756272e-05,
      "loss": 0.4632,
      "step": 1110
    },
    {
      "epoch": 17.92,
      "learning_rate": 2.2102747909199524e-05,
      "loss": 0.4772,
      "step": 1120
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.8404735062006764,
      "eval_f1": 0.8377434659582726,
      "eval_loss": 0.4618487060070038,
      "eval_precision": 0.8369871107510819,
      "eval_recall": 0.8404735062006764,
      "eval_runtime": 10.9788,
      "eval_samples_per_second": 161.584,
      "eval_steps_per_second": 2.55,
      "step": 1125
    },
    {
      "epoch": 18.08,
      "learning_rate": 2.1804062126642775e-05,
      "loss": 0.4622,
      "step": 1130
    },
    {
      "epoch": 18.24,
      "learning_rate": 2.1505376344086024e-05,
      "loss": 0.4429,
      "step": 1140
    },
    {
      "epoch": 18.4,
      "learning_rate": 2.120669056152927e-05,
      "loss": 0.447,
      "step": 1150
    },
    {
      "epoch": 18.56,
      "learning_rate": 2.0908004778972523e-05,
      "loss": 0.4583,
      "step": 1160
    },
    {
      "epoch": 18.72,
      "learning_rate": 2.060931899641577e-05,
      "loss": 0.4388,
      "step": 1170
    },
    {
      "epoch": 18.88,
      "learning_rate": 2.031063321385902e-05,
      "loss": 0.4733,
      "step": 1180
    },
    {
      "epoch": 18.99,
      "eval_accuracy": 0.8393461104847801,
      "eval_f1": 0.8381283607600565,
      "eval_loss": 0.47398388385772705,
      "eval_precision": 0.8393661287056251,
      "eval_recall": 0.8393461104847801,
      "eval_runtime": 11.274,
      "eval_samples_per_second": 157.354,
      "eval_steps_per_second": 2.484,
      "step": 1187
    },
    {
      "epoch": 19.04,
      "learning_rate": 2.001194743130227e-05,
      "loss": 0.473,
      "step": 1190
    },
    {
      "epoch": 19.2,
      "learning_rate": 1.9713261648745522e-05,
      "loss": 0.4331,
      "step": 1200
    },
    {
      "epoch": 19.36,
      "learning_rate": 1.941457586618877e-05,
      "loss": 0.4499,
      "step": 1210
    },
    {
      "epoch": 19.52,
      "learning_rate": 1.9115890083632022e-05,
      "loss": 0.4455,
      "step": 1220
    },
    {
      "epoch": 19.68,
      "learning_rate": 1.881720430107527e-05,
      "loss": 0.4435,
      "step": 1230
    },
    {
      "epoch": 19.84,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.4609,
      "step": 1240
    },
    {
      "epoch": 20.0,
      "learning_rate": 1.821983273596177e-05,
      "loss": 0.4475,
      "step": 1250
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.8387824126268321,
      "eval_f1": 0.8344955896002408,
      "eval_loss": 0.46776214241981506,
      "eval_precision": 0.8349485897476967,
      "eval_recall": 0.8387824126268321,
      "eval_runtime": 11.2755,
      "eval_samples_per_second": 157.332,
      "eval_steps_per_second": 2.483,
      "step": 1250
    },
    {
      "epoch": 20.16,
      "learning_rate": 1.7921146953405018e-05,
      "loss": 0.4388,
      "step": 1260
    },
    {
      "epoch": 20.32,
      "learning_rate": 1.7622461170848266e-05,
      "loss": 0.4496,
      "step": 1270
    },
    {
      "epoch": 20.48,
      "learning_rate": 1.7323775388291518e-05,
      "loss": 0.4217,
      "step": 1280
    },
    {
      "epoch": 20.64,
      "learning_rate": 1.702508960573477e-05,
      "loss": 0.4243,
      "step": 1290
    },
    {
      "epoch": 20.8,
      "learning_rate": 1.6726403823178017e-05,
      "loss": 0.4615,
      "step": 1300
    },
    {
      "epoch": 20.96,
      "learning_rate": 1.642771804062127e-05,
      "loss": 0.4229,
      "step": 1310
    },
    {
      "epoch": 20.99,
      "eval_accuracy": 0.8331454340473506,
      "eval_f1": 0.8302971844039118,
      "eval_loss": 0.48808062076568604,
      "eval_precision": 0.8316974605125868,
      "eval_recall": 0.8331454340473506,
      "eval_runtime": 10.9738,
      "eval_samples_per_second": 161.658,
      "eval_steps_per_second": 2.552,
      "step": 1312
    },
    {
      "epoch": 21.12,
      "learning_rate": 1.6129032258064517e-05,
      "loss": 0.4139,
      "step": 1320
    },
    {
      "epoch": 21.28,
      "learning_rate": 1.5830346475507768e-05,
      "loss": 0.423,
      "step": 1330
    },
    {
      "epoch": 21.44,
      "learning_rate": 1.5531660692951016e-05,
      "loss": 0.4173,
      "step": 1340
    },
    {
      "epoch": 21.6,
      "learning_rate": 1.5232974910394265e-05,
      "loss": 0.4253,
      "step": 1350
    },
    {
      "epoch": 21.76,
      "learning_rate": 1.4934289127837516e-05,
      "loss": 0.4358,
      "step": 1360
    },
    {
      "epoch": 21.92,
      "learning_rate": 1.4635603345280766e-05,
      "loss": 0.46,
      "step": 1370
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.8410372040586246,
      "eval_f1": 0.8371405752441734,
      "eval_loss": 0.47283288836479187,
      "eval_precision": 0.8382006013726397,
      "eval_recall": 0.8410372040586246,
      "eval_runtime": 11.2376,
      "eval_samples_per_second": 157.862,
      "eval_steps_per_second": 2.492,
      "step": 1375
    },
    {
      "epoch": 22.08,
      "learning_rate": 1.4336917562724014e-05,
      "loss": 0.4146,
      "step": 1380
    },
    {
      "epoch": 22.24,
      "learning_rate": 1.4038231780167265e-05,
      "loss": 0.3897,
      "step": 1390
    },
    {
      "epoch": 22.4,
      "learning_rate": 1.3739545997610515e-05,
      "loss": 0.4239,
      "step": 1400
    },
    {
      "epoch": 22.56,
      "learning_rate": 1.3440860215053763e-05,
      "loss": 0.4007,
      "step": 1410
    },
    {
      "epoch": 22.72,
      "learning_rate": 1.3142174432497015e-05,
      "loss": 0.4565,
      "step": 1420
    },
    {
      "epoch": 22.88,
      "learning_rate": 1.2843488649940263e-05,
      "loss": 0.4298,
      "step": 1430
    },
    {
      "epoch": 22.99,
      "eval_accuracy": 0.8359639233370914,
      "eval_f1": 0.8345442670314628,
      "eval_loss": 0.46421530842781067,
      "eval_precision": 0.8347929327915043,
      "eval_recall": 0.8359639233370914,
      "eval_runtime": 11.1996,
      "eval_samples_per_second": 158.399,
      "eval_steps_per_second": 2.5,
      "step": 1437
    },
    {
      "epoch": 23.04,
      "learning_rate": 1.2544802867383513e-05,
      "loss": 0.425,
      "step": 1440
    },
    {
      "epoch": 23.2,
      "learning_rate": 1.2246117084826763e-05,
      "loss": 0.4097,
      "step": 1450
    },
    {
      "epoch": 23.36,
      "learning_rate": 1.1947431302270013e-05,
      "loss": 0.4252,
      "step": 1460
    },
    {
      "epoch": 23.52,
      "learning_rate": 1.1648745519713262e-05,
      "loss": 0.4078,
      "step": 1470
    },
    {
      "epoch": 23.68,
      "learning_rate": 1.135005973715651e-05,
      "loss": 0.4533,
      "step": 1480
    },
    {
      "epoch": 23.84,
      "learning_rate": 1.1051373954599762e-05,
      "loss": 0.3894,
      "step": 1490
    },
    {
      "epoch": 24.0,
      "learning_rate": 1.0752688172043012e-05,
      "loss": 0.4225,
      "step": 1500
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8370913190529876,
      "eval_f1": 0.8358885146590483,
      "eval_loss": 0.47057658433914185,
      "eval_precision": 0.8368213446322319,
      "eval_recall": 0.8370913190529876,
      "eval_runtime": 11.2333,
      "eval_samples_per_second": 157.924,
      "eval_steps_per_second": 2.493,
      "step": 1500
    },
    {
      "epoch": 24.16,
      "learning_rate": 1.0454002389486262e-05,
      "loss": 0.4465,
      "step": 1510
    },
    {
      "epoch": 24.32,
      "learning_rate": 1.015531660692951e-05,
      "loss": 0.4318,
      "step": 1520
    },
    {
      "epoch": 24.48,
      "learning_rate": 9.856630824372761e-06,
      "loss": 0.4186,
      "step": 1530
    },
    {
      "epoch": 24.64,
      "learning_rate": 9.557945041816011e-06,
      "loss": 0.4053,
      "step": 1540
    },
    {
      "epoch": 24.8,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.4061,
      "step": 1550
    },
    {
      "epoch": 24.96,
      "learning_rate": 8.960573476702509e-06,
      "loss": 0.426,
      "step": 1560
    },
    {
      "epoch": 24.99,
      "eval_accuracy": 0.8399098083427283,
      "eval_f1": 0.8370590582637727,
      "eval_loss": 0.4733020067214966,
      "eval_precision": 0.83670314130695,
      "eval_recall": 0.8399098083427283,
      "eval_runtime": 11.0092,
      "eval_samples_per_second": 161.138,
      "eval_steps_per_second": 2.543,
      "step": 1562
    },
    {
      "epoch": 25.12,
      "learning_rate": 8.661887694145759e-06,
      "loss": 0.3868,
      "step": 1570
    },
    {
      "epoch": 25.28,
      "learning_rate": 8.363201911589009e-06,
      "loss": 0.4024,
      "step": 1580
    },
    {
      "epoch": 25.44,
      "learning_rate": 8.064516129032258e-06,
      "loss": 0.3977,
      "step": 1590
    },
    {
      "epoch": 25.6,
      "learning_rate": 7.765830346475508e-06,
      "loss": 0.4163,
      "step": 1600
    },
    {
      "epoch": 25.76,
      "learning_rate": 7.467144563918758e-06,
      "loss": 0.3909,
      "step": 1610
    },
    {
      "epoch": 25.92,
      "learning_rate": 7.168458781362007e-06,
      "loss": 0.3839,
      "step": 1620
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.8444193912063134,
      "eval_f1": 0.842237312876964,
      "eval_loss": 0.46824416518211365,
      "eval_precision": 0.8423076362461374,
      "eval_recall": 0.8444193912063134,
      "eval_runtime": 11.185,
      "eval_samples_per_second": 158.605,
      "eval_steps_per_second": 2.503,
      "step": 1625
    },
    {
      "epoch": 26.08,
      "learning_rate": 6.869772998805258e-06,
      "loss": 0.4023,
      "step": 1630
    },
    {
      "epoch": 26.24,
      "learning_rate": 6.5710872162485075e-06,
      "loss": 0.4116,
      "step": 1640
    },
    {
      "epoch": 26.4,
      "learning_rate": 6.2724014336917564e-06,
      "loss": 0.4061,
      "step": 1650
    },
    {
      "epoch": 26.56,
      "learning_rate": 5.973715651135006e-06,
      "loss": 0.3971,
      "step": 1660
    },
    {
      "epoch": 26.72,
      "learning_rate": 5.675029868578255e-06,
      "loss": 0.4146,
      "step": 1670
    },
    {
      "epoch": 26.88,
      "learning_rate": 5.376344086021506e-06,
      "loss": 0.4007,
      "step": 1680
    },
    {
      "epoch": 26.99,
      "eval_accuracy": 0.8382187147688839,
      "eval_f1": 0.8367328280236085,
      "eval_loss": 0.4664928615093231,
      "eval_precision": 0.8370510941185879,
      "eval_recall": 0.8382187147688839,
      "eval_runtime": 11.3183,
      "eval_samples_per_second": 156.738,
      "eval_steps_per_second": 2.474,
      "step": 1687
    },
    {
      "epoch": 27.04,
      "learning_rate": 5.077658303464755e-06,
      "loss": 0.4063,
      "step": 1690
    },
    {
      "epoch": 27.2,
      "learning_rate": 4.7789725209080055e-06,
      "loss": 0.4003,
      "step": 1700
    },
    {
      "epoch": 27.36,
      "learning_rate": 4.4802867383512545e-06,
      "loss": 0.4098,
      "step": 1710
    },
    {
      "epoch": 27.52,
      "learning_rate": 4.181600955794504e-06,
      "loss": 0.3874,
      "step": 1720
    },
    {
      "epoch": 27.68,
      "learning_rate": 3.882915173237754e-06,
      "loss": 0.3759,
      "step": 1730
    },
    {
      "epoch": 27.84,
      "learning_rate": 3.5842293906810035e-06,
      "loss": 0.4116,
      "step": 1740
    },
    {
      "epoch": 28.0,
      "learning_rate": 3.2855436081242537e-06,
      "loss": 0.4245,
      "step": 1750
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8387824126268321,
      "eval_f1": 0.8357726500514471,
      "eval_loss": 0.4694750905036926,
      "eval_precision": 0.8357086584512703,
      "eval_recall": 0.8387824126268321,
      "eval_runtime": 11.2579,
      "eval_samples_per_second": 157.578,
      "eval_steps_per_second": 2.487,
      "step": 1750
    },
    {
      "epoch": 28.16,
      "learning_rate": 2.986857825567503e-06,
      "loss": 0.3863,
      "step": 1760
    },
    {
      "epoch": 28.32,
      "learning_rate": 2.688172043010753e-06,
      "loss": 0.3928,
      "step": 1770
    },
    {
      "epoch": 28.48,
      "learning_rate": 2.3894862604540028e-06,
      "loss": 0.3847,
      "step": 1780
    },
    {
      "epoch": 28.64,
      "learning_rate": 2.090800477897252e-06,
      "loss": 0.4003,
      "step": 1790
    },
    {
      "epoch": 28.8,
      "learning_rate": 1.7921146953405017e-06,
      "loss": 0.4045,
      "step": 1800
    },
    {
      "epoch": 28.96,
      "learning_rate": 1.4934289127837516e-06,
      "loss": 0.3868,
      "step": 1810
    },
    {
      "epoch": 28.99,
      "eval_accuracy": 0.8461104847801578,
      "eval_f1": 0.844168671781978,
      "eval_loss": 0.46682804822921753,
      "eval_precision": 0.8443857819400172,
      "eval_recall": 0.8461104847801578,
      "eval_runtime": 10.9567,
      "eval_samples_per_second": 161.91,
      "eval_steps_per_second": 2.556,
      "step": 1812
    },
    {
      "epoch": 29.12,
      "learning_rate": 1.1947431302270014e-06,
      "loss": 0.397,
      "step": 1820
    },
    {
      "epoch": 29.28,
      "learning_rate": 8.960573476702509e-07,
      "loss": 0.3823,
      "step": 1830
    },
    {
      "epoch": 29.44,
      "learning_rate": 5.973715651135007e-07,
      "loss": 0.388,
      "step": 1840
    },
    {
      "epoch": 29.6,
      "learning_rate": 2.9868578255675034e-07,
      "loss": 0.3766,
      "step": 1850
    },
    {
      "epoch": 29.76,
      "learning_rate": 0.0,
      "loss": 0.3933,
      "step": 1860
    },
    {
      "epoch": 29.76,
      "eval_accuracy": 0.8461104847801578,
      "eval_f1": 0.8439938560926744,
      "eval_loss": 0.4657307267189026,
      "eval_precision": 0.8442394616991024,
      "eval_recall": 0.8461104847801578,
      "eval_runtime": 11.2751,
      "eval_samples_per_second": 157.338,
      "eval_steps_per_second": 2.483,
      "step": 1860
    },
    {
      "epoch": 29.76,
      "step": 1860,
      "total_flos": 1.4547655777742438e+18,
      "train_loss": 0.6061808514338668,
      "train_runtime": 3537.1813,
      "train_samples_per_second": 135.379,
      "train_steps_per_second": 0.526
    }
  ],
  "logging_steps": 10,
  "max_steps": 1860,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 1.4547655777742438e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}