{
  "best_metric": 0.859075535512965,
  "best_model_checkpoint": "swinv2-large-patch4-window12to16-192to256-22kto1k-ft-finetuned-galaxy10-decals/checkpoint-1000",
  "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": 2.5279,
      "step": 10
    },
    {
      "epoch": 0.32,
      "learning_rate": 5.376344086021506e-06,
      "loss": 2.2096,
      "step": 20
    },
    {
      "epoch": 0.48,
      "learning_rate": 8.064516129032258e-06,
      "loss": 1.9674,
      "step": 30
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.0752688172043012e-05,
      "loss": 1.6277,
      "step": 40
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.3440860215053763e-05,
      "loss": 1.3725,
      "step": 50
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.6129032258064517e-05,
      "loss": 1.0846,
      "step": 60
    },
    {
      "epoch": 0.99,
      "eval_accuracy": 0.7271702367531003,
      "eval_f1": 0.710066912705258,
      "eval_loss": 0.8092365264892578,
      "eval_precision": 0.7246117503323691,
      "eval_recall": 0.7271702367531003,
      "eval_runtime": 20.4313,
      "eval_samples_per_second": 86.828,
      "eval_steps_per_second": 1.37,
      "step": 62
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.881720430107527e-05,
      "loss": 0.9825,
      "step": 70
    },
    {
      "epoch": 1.28,
      "learning_rate": 2.1505376344086024e-05,
      "loss": 0.9137,
      "step": 80
    },
    {
      "epoch": 1.44,
      "learning_rate": 2.4193548387096777e-05,
      "loss": 0.8068,
      "step": 90
    },
    {
      "epoch": 1.6,
      "learning_rate": 2.6881720430107527e-05,
      "loss": 0.804,
      "step": 100
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.9569892473118284e-05,
      "loss": 0.7886,
      "step": 110
    },
    {
      "epoch": 1.92,
      "learning_rate": 3.2258064516129034e-05,
      "loss": 0.7867,
      "step": 120
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7987598647125141,
      "eval_f1": 0.7894542704452726,
      "eval_loss": 0.6366492509841919,
      "eval_precision": 0.7995839457793248,
      "eval_recall": 0.7987598647125141,
      "eval_runtime": 18.3725,
      "eval_samples_per_second": 96.557,
      "eval_steps_per_second": 1.524,
      "step": 125
    },
    {
      "epoch": 2.08,
      "learning_rate": 3.494623655913979e-05,
      "loss": 0.7576,
      "step": 130
    },
    {
      "epoch": 2.24,
      "learning_rate": 3.763440860215054e-05,
      "loss": 0.7167,
      "step": 140
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.032258064516129e-05,
      "loss": 0.727,
      "step": 150
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.301075268817205e-05,
      "loss": 0.7155,
      "step": 160
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.56989247311828e-05,
      "loss": 0.6944,
      "step": 170
    },
    {
      "epoch": 2.88,
      "learning_rate": 4.8387096774193554e-05,
      "loss": 0.6835,
      "step": 180
    },
    {
      "epoch": 2.99,
      "eval_accuracy": 0.8207440811724915,
      "eval_f1": 0.8156594653604424,
      "eval_loss": 0.5315034985542297,
      "eval_precision": 0.8195475126860307,
      "eval_recall": 0.8207440811724915,
      "eval_runtime": 18.6425,
      "eval_samples_per_second": 95.159,
      "eval_steps_per_second": 1.502,
      "step": 187
    },
    {
      "epoch": 3.04,
      "learning_rate": 4.98805256869773e-05,
      "loss": 0.729,
      "step": 190
    },
    {
      "epoch": 3.2,
      "learning_rate": 4.9581839904420555e-05,
      "loss": 0.6381,
      "step": 200
    },
    {
      "epoch": 3.36,
      "learning_rate": 4.92831541218638e-05,
      "loss": 0.6632,
      "step": 210
    },
    {
      "epoch": 3.52,
      "learning_rate": 4.898446833930705e-05,
      "loss": 0.6732,
      "step": 220
    },
    {
      "epoch": 3.68,
      "learning_rate": 4.86857825567503e-05,
      "loss": 0.606,
      "step": 230
    },
    {
      "epoch": 3.84,
      "learning_rate": 4.8387096774193554e-05,
      "loss": 0.6185,
      "step": 240
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.80884109916368e-05,
      "loss": 0.586,
      "step": 250
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8489289740698985,
      "eval_f1": 0.8451847678792103,
      "eval_loss": 0.46113088726997375,
      "eval_precision": 0.8468054811398825,
      "eval_recall": 0.8489289740698985,
      "eval_runtime": 18.5238,
      "eval_samples_per_second": 95.769,
      "eval_steps_per_second": 1.512,
      "step": 250
    },
    {
      "epoch": 4.16,
      "learning_rate": 4.778972520908005e-05,
      "loss": 0.5921,
      "step": 260
    },
    {
      "epoch": 4.32,
      "learning_rate": 4.74910394265233e-05,
      "loss": 0.5657,
      "step": 270
    },
    {
      "epoch": 4.48,
      "learning_rate": 4.7192353643966546e-05,
      "loss": 0.5686,
      "step": 280
    },
    {
      "epoch": 4.64,
      "learning_rate": 4.6893667861409805e-05,
      "loss": 0.5679,
      "step": 290
    },
    {
      "epoch": 4.8,
      "learning_rate": 4.659498207885305e-05,
      "loss": 0.5743,
      "step": 300
    },
    {
      "epoch": 4.96,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.5263,
      "step": 310
    },
    {
      "epoch": 4.99,
      "eval_accuracy": 0.8399098083427283,
      "eval_f1": 0.8399773742510294,
      "eval_loss": 0.4752826988697052,
      "eval_precision": 0.8421160537809894,
      "eval_recall": 0.8399098083427283,
      "eval_runtime": 18.5448,
      "eval_samples_per_second": 95.66,
      "eval_steps_per_second": 1.51,
      "step": 312
    },
    {
      "epoch": 5.12,
      "learning_rate": 4.5997610513739546e-05,
      "loss": 0.5385,
      "step": 320
    },
    {
      "epoch": 5.28,
      "learning_rate": 4.56989247311828e-05,
      "loss": 0.5556,
      "step": 330
    },
    {
      "epoch": 5.44,
      "learning_rate": 4.540023894862604e-05,
      "loss": 0.5365,
      "step": 340
    },
    {
      "epoch": 5.6,
      "learning_rate": 4.51015531660693e-05,
      "loss": 0.5025,
      "step": 350
    },
    {
      "epoch": 5.76,
      "learning_rate": 4.4802867383512545e-05,
      "loss": 0.5026,
      "step": 360
    },
    {
      "epoch": 5.92,
      "learning_rate": 4.4504181600955796e-05,
      "loss": 0.5341,
      "step": 370
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.842728297632469,
      "eval_f1": 0.8385794292821692,
      "eval_loss": 0.4551289677619934,
      "eval_precision": 0.8432859973436896,
      "eval_recall": 0.842728297632469,
      "eval_runtime": 18.5183,
      "eval_samples_per_second": 95.797,
      "eval_steps_per_second": 1.512,
      "step": 375
    },
    {
      "epoch": 6.08,
      "learning_rate": 4.420549581839905e-05,
      "loss": 0.5035,
      "step": 380
    },
    {
      "epoch": 6.24,
      "learning_rate": 4.390681003584229e-05,
      "loss": 0.4956,
      "step": 390
    },
    {
      "epoch": 6.4,
      "learning_rate": 4.360812425328555e-05,
      "loss": 0.4627,
      "step": 400
    },
    {
      "epoch": 6.56,
      "learning_rate": 4.3309438470728796e-05,
      "loss": 0.475,
      "step": 410
    },
    {
      "epoch": 6.72,
      "learning_rate": 4.301075268817205e-05,
      "loss": 0.482,
      "step": 420
    },
    {
      "epoch": 6.88,
      "learning_rate": 4.271206690561529e-05,
      "loss": 0.4743,
      "step": 430
    },
    {
      "epoch": 6.99,
      "eval_accuracy": 0.8382187147688839,
      "eval_f1": 0.8391125888012241,
      "eval_loss": 0.46388497948646545,
      "eval_precision": 0.8433436286058569,
      "eval_recall": 0.8382187147688839,
      "eval_runtime": 18.5306,
      "eval_samples_per_second": 95.734,
      "eval_steps_per_second": 1.511,
      "step": 437
    },
    {
      "epoch": 7.04,
      "learning_rate": 4.241338112305854e-05,
      "loss": 0.5091,
      "step": 440
    },
    {
      "epoch": 7.2,
      "learning_rate": 4.2114695340501795e-05,
      "loss": 0.4903,
      "step": 450
    },
    {
      "epoch": 7.36,
      "learning_rate": 4.1816009557945046e-05,
      "loss": 0.4464,
      "step": 460
    },
    {
      "epoch": 7.52,
      "learning_rate": 4.15173237753883e-05,
      "loss": 0.484,
      "step": 470
    },
    {
      "epoch": 7.68,
      "learning_rate": 4.121863799283154e-05,
      "loss": 0.4381,
      "step": 480
    },
    {
      "epoch": 7.84,
      "learning_rate": 4.0919952210274794e-05,
      "loss": 0.4494,
      "step": 490
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.062126642771804e-05,
      "loss": 0.4573,
      "step": 500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8359639233370914,
      "eval_f1": 0.8345320167901458,
      "eval_loss": 0.47709470987319946,
      "eval_precision": 0.8422198206630899,
      "eval_recall": 0.8359639233370914,
      "eval_runtime": 18.5105,
      "eval_samples_per_second": 95.837,
      "eval_steps_per_second": 1.513,
      "step": 500
    },
    {
      "epoch": 8.16,
      "learning_rate": 4.032258064516129e-05,
      "loss": 0.4243,
      "step": 510
    },
    {
      "epoch": 8.32,
      "learning_rate": 4.002389486260454e-05,
      "loss": 0.4333,
      "step": 520
    },
    {
      "epoch": 8.48,
      "learning_rate": 3.972520908004779e-05,
      "loss": 0.4711,
      "step": 530
    },
    {
      "epoch": 8.64,
      "learning_rate": 3.9426523297491045e-05,
      "loss": 0.4368,
      "step": 540
    },
    {
      "epoch": 8.8,
      "learning_rate": 3.912783751493429e-05,
      "loss": 0.446,
      "step": 550
    },
    {
      "epoch": 8.96,
      "learning_rate": 3.882915173237754e-05,
      "loss": 0.4368,
      "step": 560
    },
    {
      "epoch": 8.99,
      "eval_accuracy": 0.8472378804960541,
      "eval_f1": 0.845198276036514,
      "eval_loss": 0.47313007712364197,
      "eval_precision": 0.8450218204030704,
      "eval_recall": 0.8472378804960541,
      "eval_runtime": 18.4879,
      "eval_samples_per_second": 95.955,
      "eval_steps_per_second": 1.515,
      "step": 562
    },
    {
      "epoch": 9.12,
      "learning_rate": 3.8530465949820786e-05,
      "loss": 0.4094,
      "step": 570
    },
    {
      "epoch": 9.28,
      "learning_rate": 3.8231780167264044e-05,
      "loss": 0.3983,
      "step": 580
    },
    {
      "epoch": 9.44,
      "learning_rate": 3.793309438470729e-05,
      "loss": 0.3879,
      "step": 590
    },
    {
      "epoch": 9.6,
      "learning_rate": 3.763440860215054e-05,
      "loss": 0.4079,
      "step": 600
    },
    {
      "epoch": 9.76,
      "learning_rate": 3.7335722819593785e-05,
      "loss": 0.3878,
      "step": 610
    },
    {
      "epoch": 9.92,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.4022,
      "step": 620
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8540022547914318,
      "eval_f1": 0.8515779567608943,
      "eval_loss": 0.47362062335014343,
      "eval_precision": 0.8528009311153181,
      "eval_recall": 0.8540022547914318,
      "eval_runtime": 18.5141,
      "eval_samples_per_second": 95.819,
      "eval_steps_per_second": 1.512,
      "step": 625
    },
    {
      "epoch": 10.08,
      "learning_rate": 3.673835125448029e-05,
      "loss": 0.4054,
      "step": 630
    },
    {
      "epoch": 10.24,
      "learning_rate": 3.643966547192354e-05,
      "loss": 0.3767,
      "step": 640
    },
    {
      "epoch": 10.4,
      "learning_rate": 3.614097968936679e-05,
      "loss": 0.4041,
      "step": 650
    },
    {
      "epoch": 10.56,
      "learning_rate": 3.5842293906810036e-05,
      "loss": 0.3865,
      "step": 660
    },
    {
      "epoch": 10.72,
      "learning_rate": 3.554360812425329e-05,
      "loss": 0.3695,
      "step": 670
    },
    {
      "epoch": 10.88,
      "learning_rate": 3.524492234169653e-05,
      "loss": 0.4005,
      "step": 680
    },
    {
      "epoch": 10.99,
      "eval_accuracy": 0.855129650507328,
      "eval_f1": 0.8546661245976763,
      "eval_loss": 0.4541534185409546,
      "eval_precision": 0.855380931020191,
      "eval_recall": 0.855129650507328,
      "eval_runtime": 18.5065,
      "eval_samples_per_second": 95.858,
      "eval_steps_per_second": 1.513,
      "step": 687
    },
    {
      "epoch": 11.04,
      "learning_rate": 3.494623655913979e-05,
      "loss": 0.3635,
      "step": 690
    },
    {
      "epoch": 11.2,
      "learning_rate": 3.4647550776583035e-05,
      "loss": 0.3416,
      "step": 700
    },
    {
      "epoch": 11.36,
      "learning_rate": 3.4348864994026287e-05,
      "loss": 0.3555,
      "step": 710
    },
    {
      "epoch": 11.52,
      "learning_rate": 3.405017921146954e-05,
      "loss": 0.3658,
      "step": 720
    },
    {
      "epoch": 11.68,
      "learning_rate": 3.375149342891278e-05,
      "loss": 0.378,
      "step": 730
    },
    {
      "epoch": 11.84,
      "learning_rate": 3.3452807646356034e-05,
      "loss": 0.3693,
      "step": 740
    },
    {
      "epoch": 12.0,
      "learning_rate": 3.3154121863799286e-05,
      "loss": 0.3514,
      "step": 750
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.846674182638106,
      "eval_f1": 0.8470972505067972,
      "eval_loss": 0.5542634129524231,
      "eval_precision": 0.8526708457787948,
      "eval_recall": 0.846674182638106,
      "eval_runtime": 18.417,
      "eval_samples_per_second": 96.324,
      "eval_steps_per_second": 1.52,
      "step": 750
    },
    {
      "epoch": 12.16,
      "learning_rate": 3.285543608124254e-05,
      "loss": 0.3468,
      "step": 760
    },
    {
      "epoch": 12.32,
      "learning_rate": 3.255675029868578e-05,
      "loss": 0.3509,
      "step": 770
    },
    {
      "epoch": 12.48,
      "learning_rate": 3.2258064516129034e-05,
      "loss": 0.3406,
      "step": 780
    },
    {
      "epoch": 12.64,
      "learning_rate": 3.195937873357228e-05,
      "loss": 0.3787,
      "step": 790
    },
    {
      "epoch": 12.8,
      "learning_rate": 3.1660692951015537e-05,
      "loss": 0.3402,
      "step": 800
    },
    {
      "epoch": 12.96,
      "learning_rate": 3.136200716845878e-05,
      "loss": 0.3565,
      "step": 810
    },
    {
      "epoch": 12.99,
      "eval_accuracy": 0.850620067643743,
      "eval_f1": 0.8492909741299086,
      "eval_loss": 0.5318478345870972,
      "eval_precision": 0.8535480166519224,
      "eval_recall": 0.850620067643743,
      "eval_runtime": 18.3848,
      "eval_samples_per_second": 96.493,
      "eval_steps_per_second": 1.523,
      "step": 812
    },
    {
      "epoch": 13.12,
      "learning_rate": 3.106332138590203e-05,
      "loss": 0.3157,
      "step": 820
    },
    {
      "epoch": 13.28,
      "learning_rate": 3.0764635603345284e-05,
      "loss": 0.323,
      "step": 830
    },
    {
      "epoch": 13.44,
      "learning_rate": 3.046594982078853e-05,
      "loss": 0.3523,
      "step": 840
    },
    {
      "epoch": 13.6,
      "learning_rate": 3.016726403823178e-05,
      "loss": 0.3087,
      "step": 850
    },
    {
      "epoch": 13.76,
      "learning_rate": 2.9868578255675032e-05,
      "loss": 0.359,
      "step": 860
    },
    {
      "epoch": 13.92,
      "learning_rate": 2.9569892473118284e-05,
      "loss": 0.3717,
      "step": 870
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.8579481397970687,
      "eval_f1": 0.8573519793191111,
      "eval_loss": 0.5058580636978149,
      "eval_precision": 0.8582440064627089,
      "eval_recall": 0.8579481397970687,
      "eval_runtime": 18.2348,
      "eval_samples_per_second": 97.287,
      "eval_steps_per_second": 1.536,
      "step": 875
    },
    {
      "epoch": 14.08,
      "learning_rate": 2.9271206690561532e-05,
      "loss": 0.3215,
      "step": 880
    },
    {
      "epoch": 14.24,
      "learning_rate": 2.897252090800478e-05,
      "loss": 0.2951,
      "step": 890
    },
    {
      "epoch": 14.4,
      "learning_rate": 2.8673835125448028e-05,
      "loss": 0.2976,
      "step": 900
    },
    {
      "epoch": 14.56,
      "learning_rate": 2.8375149342891276e-05,
      "loss": 0.3339,
      "step": 910
    },
    {
      "epoch": 14.72,
      "learning_rate": 2.807646356033453e-05,
      "loss": 0.3176,
      "step": 920
    },
    {
      "epoch": 14.88,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.3343,
      "step": 930
    },
    {
      "epoch": 14.99,
      "eval_accuracy": 0.8472378804960541,
      "eval_f1": 0.8474450635491448,
      "eval_loss": 0.5234676599502563,
      "eval_precision": 0.8492285303630243,
      "eval_recall": 0.8472378804960541,
      "eval_runtime": 18.253,
      "eval_samples_per_second": 97.189,
      "eval_steps_per_second": 1.534,
      "step": 937
    },
    {
      "epoch": 15.04,
      "learning_rate": 2.747909199522103e-05,
      "loss": 0.3133,
      "step": 940
    },
    {
      "epoch": 15.2,
      "learning_rate": 2.718040621266428e-05,
      "loss": 0.3006,
      "step": 950
    },
    {
      "epoch": 15.36,
      "learning_rate": 2.6881720430107527e-05,
      "loss": 0.2988,
      "step": 960
    },
    {
      "epoch": 15.52,
      "learning_rate": 2.6583034647550775e-05,
      "loss": 0.2897,
      "step": 970
    },
    {
      "epoch": 15.68,
      "learning_rate": 2.628434886499403e-05,
      "loss": 0.2989,
      "step": 980
    },
    {
      "epoch": 15.84,
      "learning_rate": 2.5985663082437278e-05,
      "loss": 0.3261,
      "step": 990
    },
    {
      "epoch": 16.0,
      "learning_rate": 2.5686977299880526e-05,
      "loss": 0.3053,
      "step": 1000
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.859075535512965,
      "eval_f1": 0.8567497740752267,
      "eval_loss": 0.5226189494132996,
      "eval_precision": 0.8570814608237626,
      "eval_recall": 0.859075535512965,
      "eval_runtime": 18.4455,
      "eval_samples_per_second": 96.175,
      "eval_steps_per_second": 1.518,
      "step": 1000
    },
    {
      "epoch": 16.16,
      "learning_rate": 2.5388291517323774e-05,
      "loss": 0.293,
      "step": 1010
    },
    {
      "epoch": 16.32,
      "learning_rate": 2.5089605734767026e-05,
      "loss": 0.2744,
      "step": 1020
    },
    {
      "epoch": 16.48,
      "learning_rate": 2.4790919952210277e-05,
      "loss": 0.2987,
      "step": 1030
    },
    {
      "epoch": 16.64,
      "learning_rate": 2.4492234169653525e-05,
      "loss": 0.3371,
      "step": 1040
    },
    {
      "epoch": 16.8,
      "learning_rate": 2.4193548387096777e-05,
      "loss": 0.3071,
      "step": 1050
    },
    {
      "epoch": 16.96,
      "learning_rate": 2.3894862604540025e-05,
      "loss": 0.2607,
      "step": 1060
    },
    {
      "epoch": 16.99,
      "eval_accuracy": 0.859075535512965,
      "eval_f1": 0.8571893102690665,
      "eval_loss": 0.5654497146606445,
      "eval_precision": 0.8578913520683663,
      "eval_recall": 0.859075535512965,
      "eval_runtime": 18.484,
      "eval_samples_per_second": 95.975,
      "eval_steps_per_second": 1.515,
      "step": 1062
    },
    {
      "epoch": 17.12,
      "learning_rate": 2.3596176821983273e-05,
      "loss": 0.2984,
      "step": 1070
    },
    {
      "epoch": 17.28,
      "learning_rate": 2.3297491039426525e-05,
      "loss": 0.2784,
      "step": 1080
    },
    {
      "epoch": 17.44,
      "learning_rate": 2.2998805256869773e-05,
      "loss": 0.29,
      "step": 1090
    },
    {
      "epoch": 17.6,
      "learning_rate": 2.270011947431302e-05,
      "loss": 0.2765,
      "step": 1100
    },
    {
      "epoch": 17.76,
      "learning_rate": 2.2401433691756272e-05,
      "loss": 0.2867,
      "step": 1110
    },
    {
      "epoch": 17.92,
      "learning_rate": 2.2102747909199524e-05,
      "loss": 0.2814,
      "step": 1120
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.85456595264938,
      "eval_f1": 0.8536618062073682,
      "eval_loss": 0.5622056722640991,
      "eval_precision": 0.854062248816498,
      "eval_recall": 0.85456595264938,
      "eval_runtime": 18.1382,
      "eval_samples_per_second": 97.804,
      "eval_steps_per_second": 1.544,
      "step": 1125
    },
    {
      "epoch": 18.08,
      "learning_rate": 2.1804062126642775e-05,
      "loss": 0.2877,
      "step": 1130
    },
    {
      "epoch": 18.24,
      "learning_rate": 2.1505376344086024e-05,
      "loss": 0.2531,
      "step": 1140
    },
    {
      "epoch": 18.4,
      "learning_rate": 2.120669056152927e-05,
      "loss": 0.2641,
      "step": 1150
    },
    {
      "epoch": 18.56,
      "learning_rate": 2.0908004778972523e-05,
      "loss": 0.2465,
      "step": 1160
    },
    {
      "epoch": 18.72,
      "learning_rate": 2.060931899641577e-05,
      "loss": 0.2643,
      "step": 1170
    },
    {
      "epoch": 18.88,
      "learning_rate": 2.031063321385902e-05,
      "loss": 0.2735,
      "step": 1180
    },
    {
      "epoch": 18.99,
      "eval_accuracy": 0.850620067643743,
      "eval_f1": 0.8507783454083078,
      "eval_loss": 0.6184728145599365,
      "eval_precision": 0.8525300741996475,
      "eval_recall": 0.850620067643743,
      "eval_runtime": 18.6812,
      "eval_samples_per_second": 94.962,
      "eval_steps_per_second": 1.499,
      "step": 1187
    },
    {
      "epoch": 19.04,
      "learning_rate": 2.001194743130227e-05,
      "loss": 0.2745,
      "step": 1190
    },
    {
      "epoch": 19.2,
      "learning_rate": 1.9713261648745522e-05,
      "loss": 0.2556,
      "step": 1200
    },
    {
      "epoch": 19.36,
      "learning_rate": 1.941457586618877e-05,
      "loss": 0.2613,
      "step": 1210
    },
    {
      "epoch": 19.52,
      "learning_rate": 1.9115890083632022e-05,
      "loss": 0.2646,
      "step": 1220
    },
    {
      "epoch": 19.68,
      "learning_rate": 1.881720430107527e-05,
      "loss": 0.2558,
      "step": 1230
    },
    {
      "epoch": 19.84,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.2516,
      "step": 1240
    },
    {
      "epoch": 20.0,
      "learning_rate": 1.821983273596177e-05,
      "loss": 0.2673,
      "step": 1250
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.8573844419391207,
      "eval_f1": 0.8549826807665455,
      "eval_loss": 0.6209840178489685,
      "eval_precision": 0.8544111607262009,
      "eval_recall": 0.8573844419391207,
      "eval_runtime": 18.4393,
      "eval_samples_per_second": 96.207,
      "eval_steps_per_second": 1.518,
      "step": 1250
    },
    {
      "epoch": 20.16,
      "learning_rate": 1.7921146953405018e-05,
      "loss": 0.2472,
      "step": 1260
    },
    {
      "epoch": 20.32,
      "learning_rate": 1.7622461170848266e-05,
      "loss": 0.2591,
      "step": 1270
    },
    {
      "epoch": 20.48,
      "learning_rate": 1.7323775388291518e-05,
      "loss": 0.247,
      "step": 1280
    },
    {
      "epoch": 20.64,
      "learning_rate": 1.702508960573477e-05,
      "loss": 0.2502,
      "step": 1290
    },
    {
      "epoch": 20.8,
      "learning_rate": 1.6726403823178017e-05,
      "loss": 0.2532,
      "step": 1300
    },
    {
      "epoch": 20.96,
      "learning_rate": 1.642771804062127e-05,
      "loss": 0.2595,
      "step": 1310
    },
    {
      "epoch": 20.99,
      "eval_accuracy": 0.8421645997745209,
      "eval_f1": 0.8399025193865248,
      "eval_loss": 0.6333963871002197,
      "eval_precision": 0.8415109723113486,
      "eval_recall": 0.8421645997745209,
      "eval_runtime": 18.5443,
      "eval_samples_per_second": 95.663,
      "eval_steps_per_second": 1.51,
      "step": 1312
    },
    {
      "epoch": 21.12,
      "learning_rate": 1.6129032258064517e-05,
      "loss": 0.2453,
      "step": 1320
    },
    {
      "epoch": 21.28,
      "learning_rate": 1.5830346475507768e-05,
      "loss": 0.2524,
      "step": 1330
    },
    {
      "epoch": 21.44,
      "learning_rate": 1.5531660692951016e-05,
      "loss": 0.2461,
      "step": 1340
    },
    {
      "epoch": 21.6,
      "learning_rate": 1.5232974910394265e-05,
      "loss": 0.2342,
      "step": 1350
    },
    {
      "epoch": 21.76,
      "learning_rate": 1.4934289127837516e-05,
      "loss": 0.261,
      "step": 1360
    },
    {
      "epoch": 21.92,
      "learning_rate": 1.4635603345280766e-05,
      "loss": 0.2583,
      "step": 1370
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.8540022547914318,
      "eval_f1": 0.8526802018683215,
      "eval_loss": 0.6565257906913757,
      "eval_precision": 0.854484958770396,
      "eval_recall": 0.8540022547914318,
      "eval_runtime": 18.4868,
      "eval_samples_per_second": 95.961,
      "eval_steps_per_second": 1.515,
      "step": 1375
    },
    {
      "epoch": 22.08,
      "learning_rate": 1.4336917562724014e-05,
      "loss": 0.2372,
      "step": 1380
    },
    {
      "epoch": 22.24,
      "learning_rate": 1.4038231780167265e-05,
      "loss": 0.2271,
      "step": 1390
    },
    {
      "epoch": 22.4,
      "learning_rate": 1.3739545997610515e-05,
      "loss": 0.2385,
      "step": 1400
    },
    {
      "epoch": 22.56,
      "learning_rate": 1.3440860215053763e-05,
      "loss": 0.2432,
      "step": 1410
    },
    {
      "epoch": 22.72,
      "learning_rate": 1.3142174432497015e-05,
      "loss": 0.2332,
      "step": 1420
    },
    {
      "epoch": 22.88,
      "learning_rate": 1.2843488649940263e-05,
      "loss": 0.239,
      "step": 1430
    },
    {
      "epoch": 22.99,
      "eval_accuracy": 0.8455467869222097,
      "eval_f1": 0.8447040379210317,
      "eval_loss": 0.685856819152832,
      "eval_precision": 0.8457553216629743,
      "eval_recall": 0.8455467869222097,
      "eval_runtime": 18.4024,
      "eval_samples_per_second": 96.4,
      "eval_steps_per_second": 1.522,
      "step": 1437
    },
    {
      "epoch": 23.04,
      "learning_rate": 1.2544802867383513e-05,
      "loss": 0.2302,
      "step": 1440
    },
    {
      "epoch": 23.2,
      "learning_rate": 1.2246117084826763e-05,
      "loss": 0.2368,
      "step": 1450
    },
    {
      "epoch": 23.36,
      "learning_rate": 1.1947431302270013e-05,
      "loss": 0.2248,
      "step": 1460
    },
    {
      "epoch": 23.52,
      "learning_rate": 1.1648745519713262e-05,
      "loss": 0.2197,
      "step": 1470
    },
    {
      "epoch": 23.68,
      "learning_rate": 1.135005973715651e-05,
      "loss": 0.2568,
      "step": 1480
    },
    {
      "epoch": 23.84,
      "learning_rate": 1.1051373954599762e-05,
      "loss": 0.2208,
      "step": 1490
    },
    {
      "epoch": 24.0,
      "learning_rate": 1.0752688172043012e-05,
      "loss": 0.2174,
      "step": 1500
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.859075535512965,
      "eval_f1": 0.858050772674415,
      "eval_loss": 0.6709402799606323,
      "eval_precision": 0.858080303770645,
      "eval_recall": 0.859075535512965,
      "eval_runtime": 18.5074,
      "eval_samples_per_second": 95.853,
      "eval_steps_per_second": 1.513,
      "step": 1500
    },
    {
      "epoch": 24.16,
      "learning_rate": 1.0454002389486262e-05,
      "loss": 0.2337,
      "step": 1510
    },
    {
      "epoch": 24.32,
      "learning_rate": 1.015531660692951e-05,
      "loss": 0.2439,
      "step": 1520
    },
    {
      "epoch": 24.48,
      "learning_rate": 9.856630824372761e-06,
      "loss": 0.2225,
      "step": 1530
    },
    {
      "epoch": 24.64,
      "learning_rate": 9.557945041816011e-06,
      "loss": 0.2082,
      "step": 1540
    },
    {
      "epoch": 24.8,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.2151,
      "step": 1550
    },
    {
      "epoch": 24.96,
      "learning_rate": 8.960573476702509e-06,
      "loss": 0.2288,
      "step": 1560
    },
    {
      "epoch": 24.99,
      "eval_accuracy": 0.8444193912063134,
      "eval_f1": 0.8418649366037473,
      "eval_loss": 0.7437374591827393,
      "eval_precision": 0.8426163594450699,
      "eval_recall": 0.8444193912063134,
      "eval_runtime": 18.5298,
      "eval_samples_per_second": 95.738,
      "eval_steps_per_second": 1.511,
      "step": 1562
    },
    {
      "epoch": 25.12,
      "learning_rate": 8.661887694145759e-06,
      "loss": 0.2229,
      "step": 1570
    },
    {
      "epoch": 25.28,
      "learning_rate": 8.363201911589009e-06,
      "loss": 0.2123,
      "step": 1580
    },
    {
      "epoch": 25.44,
      "learning_rate": 8.064516129032258e-06,
      "loss": 0.2093,
      "step": 1590
    },
    {
      "epoch": 25.6,
      "learning_rate": 7.765830346475508e-06,
      "loss": 0.2285,
      "step": 1600
    },
    {
      "epoch": 25.76,
      "learning_rate": 7.467144563918758e-06,
      "loss": 0.2035,
      "step": 1610
    },
    {
      "epoch": 25.92,
      "learning_rate": 7.168458781362007e-06,
      "loss": 0.2305,
      "step": 1620
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.8528748590755355,
      "eval_f1": 0.8504693914732183,
      "eval_loss": 0.7048392295837402,
      "eval_precision": 0.8496994067833048,
      "eval_recall": 0.8528748590755355,
      "eval_runtime": 18.4911,
      "eval_samples_per_second": 95.938,
      "eval_steps_per_second": 1.514,
      "step": 1625
    },
    {
      "epoch": 26.08,
      "learning_rate": 6.869772998805258e-06,
      "loss": 0.2193,
      "step": 1630
    },
    {
      "epoch": 26.24,
      "learning_rate": 6.5710872162485075e-06,
      "loss": 0.2269,
      "step": 1640
    },
    {
      "epoch": 26.4,
      "learning_rate": 6.2724014336917564e-06,
      "loss": 0.2146,
      "step": 1650
    },
    {
      "epoch": 26.56,
      "learning_rate": 5.973715651135006e-06,
      "loss": 0.2067,
      "step": 1660
    },
    {
      "epoch": 26.72,
      "learning_rate": 5.675029868578255e-06,
      "loss": 0.2301,
      "step": 1670
    },
    {
      "epoch": 26.88,
      "learning_rate": 5.376344086021506e-06,
      "loss": 0.2071,
      "step": 1680
    },
    {
      "epoch": 26.99,
      "eval_accuracy": 0.8540022547914318,
      "eval_f1": 0.8528799874612519,
      "eval_loss": 0.7152166962623596,
      "eval_precision": 0.8526768423519663,
      "eval_recall": 0.8540022547914318,
      "eval_runtime": 18.4564,
      "eval_samples_per_second": 96.118,
      "eval_steps_per_second": 1.517,
      "step": 1687
    },
    {
      "epoch": 27.04,
      "learning_rate": 5.077658303464755e-06,
      "loss": 0.217,
      "step": 1690
    },
    {
      "epoch": 27.2,
      "learning_rate": 4.7789725209080055e-06,
      "loss": 0.2143,
      "step": 1700
    },
    {
      "epoch": 27.36,
      "learning_rate": 4.4802867383512545e-06,
      "loss": 0.2406,
      "step": 1710
    },
    {
      "epoch": 27.52,
      "learning_rate": 4.181600955794504e-06,
      "loss": 0.2089,
      "step": 1720
    },
    {
      "epoch": 27.68,
      "learning_rate": 3.882915173237754e-06,
      "loss": 0.1949,
      "step": 1730
    },
    {
      "epoch": 27.84,
      "learning_rate": 3.5842293906810035e-06,
      "loss": 0.2189,
      "step": 1740
    },
    {
      "epoch": 28.0,
      "learning_rate": 3.2855436081242537e-06,
      "loss": 0.2282,
      "step": 1750
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8568207440811725,
      "eval_f1": 0.855387115048938,
      "eval_loss": 0.7273016571998596,
      "eval_precision": 0.8559343763907659,
      "eval_recall": 0.8568207440811725,
      "eval_runtime": 18.0723,
      "eval_samples_per_second": 98.161,
      "eval_steps_per_second": 1.549,
      "step": 1750
    },
    {
      "epoch": 28.16,
      "learning_rate": 2.986857825567503e-06,
      "loss": 0.2123,
      "step": 1760
    },
    {
      "epoch": 28.32,
      "learning_rate": 2.688172043010753e-06,
      "loss": 0.2084,
      "step": 1770
    },
    {
      "epoch": 28.48,
      "learning_rate": 2.3894862604540028e-06,
      "loss": 0.2008,
      "step": 1780
    },
    {
      "epoch": 28.64,
      "learning_rate": 2.090800477897252e-06,
      "loss": 0.2103,
      "step": 1790
    },
    {
      "epoch": 28.8,
      "learning_rate": 1.7921146953405017e-06,
      "loss": 0.208,
      "step": 1800
    },
    {
      "epoch": 28.96,
      "learning_rate": 1.4934289127837516e-06,
      "loss": 0.209,
      "step": 1810
    },
    {
      "epoch": 28.99,
      "eval_accuracy": 0.8556933483652762,
      "eval_f1": 0.8539613944120064,
      "eval_loss": 0.7212563157081604,
      "eval_precision": 0.8534141018186441,
      "eval_recall": 0.8556933483652762,
      "eval_runtime": 18.5102,
      "eval_samples_per_second": 95.839,
      "eval_steps_per_second": 1.513,
      "step": 1812
    },
    {
      "epoch": 29.12,
      "learning_rate": 1.1947431302270014e-06,
      "loss": 0.2049,
      "step": 1820
    },
    {
      "epoch": 29.28,
      "learning_rate": 8.960573476702509e-07,
      "loss": 0.1903,
      "step": 1830
    },
    {
      "epoch": 29.44,
      "learning_rate": 5.973715651135007e-07,
      "loss": 0.2082,
      "step": 1840
    },
    {
      "epoch": 29.6,
      "learning_rate": 2.9868578255675034e-07,
      "loss": 0.2124,
      "step": 1850
    },
    {
      "epoch": 29.76,
      "learning_rate": 0.0,
      "loss": 0.2078,
      "step": 1860
    },
    {
      "epoch": 29.76,
      "eval_accuracy": 0.8562570462232244,
      "eval_f1": 0.8548157706421159,
      "eval_loss": 0.72725909948349,
      "eval_precision": 0.8544032501079596,
      "eval_recall": 0.8562570462232244,
      "eval_runtime": 18.5918,
      "eval_samples_per_second": 95.418,
      "eval_steps_per_second": 1.506,
      "step": 1860
    },
    {
      "epoch": 29.76,
      "step": 1860,
      "total_flos": 1.0940057452591763e+20,
      "train_loss": 0.4096715816887476,
      "train_runtime": 12450.2224,
      "train_samples_per_second": 38.462,
      "train_steps_per_second": 0.149
    }
  ],
  "logging_steps": 10,
  "max_steps": 1860,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 1.0940057452591763e+20,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}