{
  "best_metric": 0.9769230769230769,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned_ASL_Isolated_Swin_dataset2/checkpoint-1700",
  "epoch": 20.0,
  "eval_steps": 100,
  "global_step": 1840,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11,
      "learning_rate": 0.0001991304347826087,
      "loss": 3.3413,
      "step": 10
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00019804347826086956,
      "loss": 3.1778,
      "step": 20
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019695652173913044,
      "loss": 3.1311,
      "step": 30
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001958695652173913,
      "loss": 2.9107,
      "step": 40
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00019478260869565218,
      "loss": 2.3763,
      "step": 50
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00019369565217391307,
      "loss": 2.0376,
      "step": 60
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00019260869565217392,
      "loss": 1.9444,
      "step": 70
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00019152173913043478,
      "loss": 1.8087,
      "step": 80
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00019043478260869566,
      "loss": 1.6835,
      "step": 90
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00018934782608695655,
      "loss": 1.5439,
      "step": 100
    },
    {
      "epoch": 1.09,
      "eval_accuracy": 0.5538461538461539,
      "eval_loss": 1.4188454151153564,
      "eval_runtime": 1.8531,
      "eval_samples_per_second": 140.308,
      "eval_steps_per_second": 17.808,
      "step": 100
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001882608695652174,
      "loss": 1.3483,
      "step": 110
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00018717391304347826,
      "loss": 1.3837,
      "step": 120
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00018608695652173914,
      "loss": 1.2886,
      "step": 130
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00018500000000000002,
      "loss": 1.1915,
      "step": 140
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00018391304347826088,
      "loss": 1.1142,
      "step": 150
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00018282608695652174,
      "loss": 0.9908,
      "step": 160
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00018173913043478262,
      "loss": 0.9563,
      "step": 170
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00018065217391304348,
      "loss": 0.9761,
      "step": 180
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.00017956521739130436,
      "loss": 1.0636,
      "step": 190
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.00017847826086956522,
      "loss": 0.8646,
      "step": 200
    },
    {
      "epoch": 2.17,
      "eval_accuracy": 0.8884615384615384,
      "eval_loss": 0.45421722531318665,
      "eval_runtime": 2.2052,
      "eval_samples_per_second": 117.904,
      "eval_steps_per_second": 14.965,
      "step": 200
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0001773913043478261,
      "loss": 0.7407,
      "step": 210
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.00017630434782608695,
      "loss": 0.8695,
      "step": 220
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.00017521739130434784,
      "loss": 0.7102,
      "step": 230
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.00017413043478260872,
      "loss": 0.9596,
      "step": 240
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.00017304347826086958,
      "loss": 0.8462,
      "step": 250
    },
    {
      "epoch": 2.83,
      "learning_rate": 0.00017195652173913043,
      "loss": 0.8556,
      "step": 260
    },
    {
      "epoch": 2.93,
      "learning_rate": 0.00017086956521739132,
      "loss": 0.664,
      "step": 270
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0001697826086956522,
      "loss": 0.8349,
      "step": 280
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.00016869565217391306,
      "loss": 0.6526,
      "step": 290
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0001676086956521739,
      "loss": 0.5485,
      "step": 300
    },
    {
      "epoch": 3.26,
      "eval_accuracy": 0.8538461538461538,
      "eval_loss": 0.4102769196033478,
      "eval_runtime": 1.7696,
      "eval_samples_per_second": 146.93,
      "eval_steps_per_second": 18.649,
      "step": 300
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0001665217391304348,
      "loss": 0.701,
      "step": 310
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.00016543478260869568,
      "loss": 0.725,
      "step": 320
    },
    {
      "epoch": 3.59,
      "learning_rate": 0.00016434782608695653,
      "loss": 0.7088,
      "step": 330
    },
    {
      "epoch": 3.7,
      "learning_rate": 0.0001632608695652174,
      "loss": 0.5813,
      "step": 340
    },
    {
      "epoch": 3.8,
      "learning_rate": 0.00016217391304347827,
      "loss": 0.7197,
      "step": 350
    },
    {
      "epoch": 3.91,
      "learning_rate": 0.00016108695652173913,
      "loss": 0.7112,
      "step": 360
    },
    {
      "epoch": 4.02,
      "learning_rate": 0.00016,
      "loss": 0.6083,
      "step": 370
    },
    {
      "epoch": 4.13,
      "learning_rate": 0.00015891304347826087,
      "loss": 0.6653,
      "step": 380
    },
    {
      "epoch": 4.24,
      "learning_rate": 0.00015782608695652175,
      "loss": 0.6319,
      "step": 390
    },
    {
      "epoch": 4.35,
      "learning_rate": 0.0001567391304347826,
      "loss": 0.5082,
      "step": 400
    },
    {
      "epoch": 4.35,
      "eval_accuracy": 0.8961538461538462,
      "eval_loss": 0.29254817962646484,
      "eval_runtime": 2.5312,
      "eval_samples_per_second": 102.718,
      "eval_steps_per_second": 13.037,
      "step": 400
    },
    {
      "epoch": 4.46,
      "learning_rate": 0.00015565217391304346,
      "loss": 0.6872,
      "step": 410
    },
    {
      "epoch": 4.57,
      "learning_rate": 0.00015456521739130437,
      "loss": 0.5159,
      "step": 420
    },
    {
      "epoch": 4.67,
      "learning_rate": 0.00015347826086956523,
      "loss": 0.4834,
      "step": 430
    },
    {
      "epoch": 4.78,
      "learning_rate": 0.0001523913043478261,
      "loss": 0.4467,
      "step": 440
    },
    {
      "epoch": 4.89,
      "learning_rate": 0.00015130434782608694,
      "loss": 0.5091,
      "step": 450
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.00015021739130434785,
      "loss": 0.701,
      "step": 460
    },
    {
      "epoch": 5.11,
      "learning_rate": 0.0001491304347826087,
      "loss": 0.5314,
      "step": 470
    },
    {
      "epoch": 5.22,
      "learning_rate": 0.00014804347826086957,
      "loss": 0.5484,
      "step": 480
    },
    {
      "epoch": 5.33,
      "learning_rate": 0.00014695652173913045,
      "loss": 0.4856,
      "step": 490
    },
    {
      "epoch": 5.43,
      "learning_rate": 0.0001458695652173913,
      "loss": 0.5302,
      "step": 500
    },
    {
      "epoch": 5.43,
      "eval_accuracy": 0.926923076923077,
      "eval_loss": 0.24707049131393433,
      "eval_runtime": 1.8365,
      "eval_samples_per_second": 141.57,
      "eval_steps_per_second": 17.969,
      "step": 500
    },
    {
      "epoch": 5.54,
      "learning_rate": 0.0001447826086956522,
      "loss": 0.4876,
      "step": 510
    },
    {
      "epoch": 5.65,
      "learning_rate": 0.00014369565217391304,
      "loss": 0.6267,
      "step": 520
    },
    {
      "epoch": 5.76,
      "learning_rate": 0.00014260869565217393,
      "loss": 0.5022,
      "step": 530
    },
    {
      "epoch": 5.87,
      "learning_rate": 0.00014152173913043478,
      "loss": 0.6259,
      "step": 540
    },
    {
      "epoch": 5.98,
      "learning_rate": 0.00014043478260869567,
      "loss": 0.588,
      "step": 550
    },
    {
      "epoch": 6.09,
      "learning_rate": 0.00013934782608695652,
      "loss": 0.6801,
      "step": 560
    },
    {
      "epoch": 6.2,
      "learning_rate": 0.0001382608695652174,
      "loss": 0.6062,
      "step": 570
    },
    {
      "epoch": 6.3,
      "learning_rate": 0.00013717391304347826,
      "loss": 0.3015,
      "step": 580
    },
    {
      "epoch": 6.41,
      "learning_rate": 0.00013608695652173912,
      "loss": 0.5071,
      "step": 590
    },
    {
      "epoch": 6.52,
      "learning_rate": 0.00013500000000000003,
      "loss": 0.4072,
      "step": 600
    },
    {
      "epoch": 6.52,
      "eval_accuracy": 0.9230769230769231,
      "eval_loss": 0.2676173746585846,
      "eval_runtime": 2.5959,
      "eval_samples_per_second": 100.157,
      "eval_steps_per_second": 12.712,
      "step": 600
    },
    {
      "epoch": 6.63,
      "learning_rate": 0.00013391304347826088,
      "loss": 0.451,
      "step": 610
    },
    {
      "epoch": 6.74,
      "learning_rate": 0.00013282608695652174,
      "loss": 0.3797,
      "step": 620
    },
    {
      "epoch": 6.85,
      "learning_rate": 0.0001317391304347826,
      "loss": 0.415,
      "step": 630
    },
    {
      "epoch": 6.96,
      "learning_rate": 0.00013065217391304348,
      "loss": 0.4253,
      "step": 640
    },
    {
      "epoch": 7.07,
      "learning_rate": 0.00012956521739130436,
      "loss": 0.4366,
      "step": 650
    },
    {
      "epoch": 7.17,
      "learning_rate": 0.00012847826086956522,
      "loss": 0.3499,
      "step": 660
    },
    {
      "epoch": 7.28,
      "learning_rate": 0.0001273913043478261,
      "loss": 0.4012,
      "step": 670
    },
    {
      "epoch": 7.39,
      "learning_rate": 0.00012630434782608696,
      "loss": 0.414,
      "step": 680
    },
    {
      "epoch": 7.5,
      "learning_rate": 0.00012521739130434784,
      "loss": 0.5052,
      "step": 690
    },
    {
      "epoch": 7.61,
      "learning_rate": 0.0001241304347826087,
      "loss": 0.4424,
      "step": 700
    },
    {
      "epoch": 7.61,
      "eval_accuracy": 0.9038461538461539,
      "eval_loss": 0.4149817228317261,
      "eval_runtime": 1.8169,
      "eval_samples_per_second": 143.1,
      "eval_steps_per_second": 18.163,
      "step": 700
    },
    {
      "epoch": 7.72,
      "learning_rate": 0.00012304347826086958,
      "loss": 0.5183,
      "step": 710
    },
    {
      "epoch": 7.83,
      "learning_rate": 0.00012195652173913044,
      "loss": 0.2976,
      "step": 720
    },
    {
      "epoch": 7.93,
      "learning_rate": 0.00012086956521739131,
      "loss": 0.5184,
      "step": 730
    },
    {
      "epoch": 8.04,
      "learning_rate": 0.00011978260869565219,
      "loss": 0.4194,
      "step": 740
    },
    {
      "epoch": 8.15,
      "learning_rate": 0.00011869565217391305,
      "loss": 0.3032,
      "step": 750
    },
    {
      "epoch": 8.26,
      "learning_rate": 0.00011760869565217392,
      "loss": 0.5633,
      "step": 760
    },
    {
      "epoch": 8.37,
      "learning_rate": 0.00011652173913043479,
      "loss": 0.3037,
      "step": 770
    },
    {
      "epoch": 8.48,
      "learning_rate": 0.00011543478260869567,
      "loss": 0.3617,
      "step": 780
    },
    {
      "epoch": 8.59,
      "learning_rate": 0.00011434782608695654,
      "loss": 0.4318,
      "step": 790
    },
    {
      "epoch": 8.7,
      "learning_rate": 0.0001132608695652174,
      "loss": 0.3409,
      "step": 800
    },
    {
      "epoch": 8.7,
      "eval_accuracy": 0.9538461538461539,
      "eval_loss": 0.19218704104423523,
      "eval_runtime": 2.6334,
      "eval_samples_per_second": 98.732,
      "eval_steps_per_second": 12.531,
      "step": 800
    },
    {
      "epoch": 8.8,
      "learning_rate": 0.00011217391304347826,
      "loss": 0.2645,
      "step": 810
    },
    {
      "epoch": 8.91,
      "learning_rate": 0.00011108695652173912,
      "loss": 0.2775,
      "step": 820
    },
    {
      "epoch": 9.02,
      "learning_rate": 0.00011000000000000002,
      "loss": 0.4254,
      "step": 830
    },
    {
      "epoch": 9.13,
      "learning_rate": 0.00010891304347826087,
      "loss": 0.3383,
      "step": 840
    },
    {
      "epoch": 9.24,
      "learning_rate": 0.00010782608695652174,
      "loss": 0.3443,
      "step": 850
    },
    {
      "epoch": 9.35,
      "learning_rate": 0.00010673913043478261,
      "loss": 0.2455,
      "step": 860
    },
    {
      "epoch": 9.46,
      "learning_rate": 0.00010565217391304347,
      "loss": 0.2475,
      "step": 870
    },
    {
      "epoch": 9.57,
      "learning_rate": 0.00010456521739130437,
      "loss": 0.4743,
      "step": 880
    },
    {
      "epoch": 9.67,
      "learning_rate": 0.00010347826086956522,
      "loss": 0.4935,
      "step": 890
    },
    {
      "epoch": 9.78,
      "learning_rate": 0.00010239130434782609,
      "loss": 0.3046,
      "step": 900
    },
    {
      "epoch": 9.78,
      "eval_accuracy": 0.9461538461538461,
      "eval_loss": 0.1916845738887787,
      "eval_runtime": 1.7809,
      "eval_samples_per_second": 145.992,
      "eval_steps_per_second": 18.53,
      "step": 900
    },
    {
      "epoch": 9.89,
      "learning_rate": 0.00010130434782608695,
      "loss": 0.2834,
      "step": 910
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.00010021739130434784,
      "loss": 0.4026,
      "step": 920
    },
    {
      "epoch": 10.11,
      "learning_rate": 9.91304347826087e-05,
      "loss": 0.3004,
      "step": 930
    },
    {
      "epoch": 10.22,
      "learning_rate": 9.804347826086957e-05,
      "loss": 0.2161,
      "step": 940
    },
    {
      "epoch": 10.33,
      "learning_rate": 9.695652173913044e-05,
      "loss": 0.3859,
      "step": 950
    },
    {
      "epoch": 10.43,
      "learning_rate": 9.586956521739131e-05,
      "loss": 0.2911,
      "step": 960
    },
    {
      "epoch": 10.54,
      "learning_rate": 9.478260869565218e-05,
      "loss": 0.2693,
      "step": 970
    },
    {
      "epoch": 10.65,
      "learning_rate": 9.369565217391305e-05,
      "loss": 0.2366,
      "step": 980
    },
    {
      "epoch": 10.76,
      "learning_rate": 9.260869565217392e-05,
      "loss": 0.3161,
      "step": 990
    },
    {
      "epoch": 10.87,
      "learning_rate": 9.152173913043479e-05,
      "loss": 0.2911,
      "step": 1000
    },
    {
      "epoch": 10.87,
      "eval_accuracy": 0.9423076923076923,
      "eval_loss": 0.2271503061056137,
      "eval_runtime": 2.5147,
      "eval_samples_per_second": 103.391,
      "eval_steps_per_second": 13.123,
      "step": 1000
    },
    {
      "epoch": 10.98,
      "learning_rate": 9.043478260869566e-05,
      "loss": 0.4276,
      "step": 1010
    },
    {
      "epoch": 11.09,
      "learning_rate": 8.934782608695653e-05,
      "loss": 0.245,
      "step": 1020
    },
    {
      "epoch": 11.2,
      "learning_rate": 8.82608695652174e-05,
      "loss": 0.2376,
      "step": 1030
    },
    {
      "epoch": 11.3,
      "learning_rate": 8.717391304347827e-05,
      "loss": 0.1843,
      "step": 1040
    },
    {
      "epoch": 11.41,
      "learning_rate": 8.608695652173914e-05,
      "loss": 0.3411,
      "step": 1050
    },
    {
      "epoch": 11.52,
      "learning_rate": 8.5e-05,
      "loss": 0.5081,
      "step": 1060
    },
    {
      "epoch": 11.63,
      "learning_rate": 8.391304347826088e-05,
      "loss": 0.2889,
      "step": 1070
    },
    {
      "epoch": 11.74,
      "learning_rate": 8.282608695652175e-05,
      "loss": 0.2588,
      "step": 1080
    },
    {
      "epoch": 11.85,
      "learning_rate": 8.173913043478262e-05,
      "loss": 0.1457,
      "step": 1090
    },
    {
      "epoch": 11.96,
      "learning_rate": 8.076086956521739e-05,
      "loss": 0.269,
      "step": 1100
    },
    {
      "epoch": 11.96,
      "eval_accuracy": 0.9692307692307692,
      "eval_loss": 0.07221826165914536,
      "eval_runtime": 1.7687,
      "eval_samples_per_second": 146.998,
      "eval_steps_per_second": 18.657,
      "step": 1100
    },
    {
      "epoch": 12.07,
      "learning_rate": 7.967391304347826e-05,
      "loss": 0.3124,
      "step": 1110
    },
    {
      "epoch": 12.17,
      "learning_rate": 7.858695652173913e-05,
      "loss": 0.1558,
      "step": 1120
    },
    {
      "epoch": 12.28,
      "learning_rate": 7.75e-05,
      "loss": 0.3082,
      "step": 1130
    },
    {
      "epoch": 12.39,
      "learning_rate": 7.641304347826087e-05,
      "loss": 0.2492,
      "step": 1140
    },
    {
      "epoch": 12.5,
      "learning_rate": 7.532608695652175e-05,
      "loss": 0.3847,
      "step": 1150
    },
    {
      "epoch": 12.61,
      "learning_rate": 7.423913043478261e-05,
      "loss": 0.3452,
      "step": 1160
    },
    {
      "epoch": 12.72,
      "learning_rate": 7.315217391304349e-05,
      "loss": 0.3171,
      "step": 1170
    },
    {
      "epoch": 12.83,
      "learning_rate": 7.206521739130435e-05,
      "loss": 0.3295,
      "step": 1180
    },
    {
      "epoch": 12.93,
      "learning_rate": 7.097826086956522e-05,
      "loss": 0.3387,
      "step": 1190
    },
    {
      "epoch": 13.04,
      "learning_rate": 6.989130434782609e-05,
      "loss": 0.3709,
      "step": 1200
    },
    {
      "epoch": 13.04,
      "eval_accuracy": 0.9653846153846154,
      "eval_loss": 0.14726483821868896,
      "eval_runtime": 2.5212,
      "eval_samples_per_second": 103.126,
      "eval_steps_per_second": 13.089,
      "step": 1200
    },
    {
      "epoch": 13.15,
      "learning_rate": 6.880434782608696e-05,
      "loss": 0.3647,
      "step": 1210
    },
    {
      "epoch": 13.26,
      "learning_rate": 6.771739130434783e-05,
      "loss": 0.137,
      "step": 1220
    },
    {
      "epoch": 13.37,
      "learning_rate": 6.66304347826087e-05,
      "loss": 0.3562,
      "step": 1230
    },
    {
      "epoch": 13.48,
      "learning_rate": 6.554347826086957e-05,
      "loss": 0.323,
      "step": 1240
    },
    {
      "epoch": 13.59,
      "learning_rate": 6.445652173913044e-05,
      "loss": 0.2173,
      "step": 1250
    },
    {
      "epoch": 13.7,
      "learning_rate": 6.33695652173913e-05,
      "loss": 0.315,
      "step": 1260
    },
    {
      "epoch": 13.8,
      "learning_rate": 6.228260869565218e-05,
      "loss": 0.274,
      "step": 1270
    },
    {
      "epoch": 13.91,
      "learning_rate": 6.119565217391304e-05,
      "loss": 0.1757,
      "step": 1280
    },
    {
      "epoch": 14.02,
      "learning_rate": 6.010869565217392e-05,
      "loss": 0.1968,
      "step": 1290
    },
    {
      "epoch": 14.13,
      "learning_rate": 5.9021739130434784e-05,
      "loss": 0.3443,
      "step": 1300
    },
    {
      "epoch": 14.13,
      "eval_accuracy": 0.9615384615384616,
      "eval_loss": 0.15447035431861877,
      "eval_runtime": 1.7932,
      "eval_samples_per_second": 144.988,
      "eval_steps_per_second": 18.402,
      "step": 1300
    },
    {
      "epoch": 14.24,
      "learning_rate": 5.793478260869566e-05,
      "loss": 0.225,
      "step": 1310
    },
    {
      "epoch": 14.35,
      "learning_rate": 5.6847826086956524e-05,
      "loss": 0.336,
      "step": 1320
    },
    {
      "epoch": 14.46,
      "learning_rate": 5.5760869565217386e-05,
      "loss": 0.2073,
      "step": 1330
    },
    {
      "epoch": 14.57,
      "learning_rate": 5.467391304347826e-05,
      "loss": 0.2144,
      "step": 1340
    },
    {
      "epoch": 14.67,
      "learning_rate": 5.3586956521739126e-05,
      "loss": 0.3399,
      "step": 1350
    },
    {
      "epoch": 14.78,
      "learning_rate": 5.25e-05,
      "loss": 0.2516,
      "step": 1360
    },
    {
      "epoch": 14.89,
      "learning_rate": 5.141304347826087e-05,
      "loss": 0.2494,
      "step": 1370
    },
    {
      "epoch": 15.0,
      "learning_rate": 5.032608695652175e-05,
      "loss": 0.2213,
      "step": 1380
    },
    {
      "epoch": 15.11,
      "learning_rate": 4.923913043478261e-05,
      "loss": 0.1834,
      "step": 1390
    },
    {
      "epoch": 15.22,
      "learning_rate": 4.815217391304348e-05,
      "loss": 0.187,
      "step": 1400
    },
    {
      "epoch": 15.22,
      "eval_accuracy": 0.9730769230769231,
      "eval_loss": 0.10598226636648178,
      "eval_runtime": 2.4502,
      "eval_samples_per_second": 106.116,
      "eval_steps_per_second": 13.469,
      "step": 1400
    },
    {
      "epoch": 15.33,
      "learning_rate": 4.706521739130435e-05,
      "loss": 0.1387,
      "step": 1410
    },
    {
      "epoch": 15.43,
      "learning_rate": 4.597826086956522e-05,
      "loss": 0.1653,
      "step": 1420
    },
    {
      "epoch": 15.54,
      "learning_rate": 4.489130434782609e-05,
      "loss": 0.2012,
      "step": 1430
    },
    {
      "epoch": 15.65,
      "learning_rate": 4.380434782608696e-05,
      "loss": 0.224,
      "step": 1440
    },
    {
      "epoch": 15.76,
      "learning_rate": 4.271739130434783e-05,
      "loss": 0.3341,
      "step": 1450
    },
    {
      "epoch": 15.87,
      "learning_rate": 4.16304347826087e-05,
      "loss": 0.3279,
      "step": 1460
    },
    {
      "epoch": 15.98,
      "learning_rate": 4.054347826086957e-05,
      "loss": 0.2218,
      "step": 1470
    },
    {
      "epoch": 16.09,
      "learning_rate": 3.945652173913044e-05,
      "loss": 0.2184,
      "step": 1480
    },
    {
      "epoch": 16.2,
      "learning_rate": 3.836956521739131e-05,
      "loss": 0.2069,
      "step": 1490
    },
    {
      "epoch": 16.3,
      "learning_rate": 3.728260869565218e-05,
      "loss": 0.1879,
      "step": 1500
    },
    {
      "epoch": 16.3,
      "eval_accuracy": 0.9692307692307692,
      "eval_loss": 0.11237060278654099,
      "eval_runtime": 1.8389,
      "eval_samples_per_second": 141.393,
      "eval_steps_per_second": 17.946,
      "step": 1500
    },
    {
      "epoch": 16.41,
      "learning_rate": 3.619565217391305e-05,
      "loss": 0.1226,
      "step": 1510
    },
    {
      "epoch": 16.52,
      "learning_rate": 3.510869565217392e-05,
      "loss": 0.1394,
      "step": 1520
    },
    {
      "epoch": 16.63,
      "learning_rate": 3.402173913043478e-05,
      "loss": 0.2203,
      "step": 1530
    },
    {
      "epoch": 16.74,
      "learning_rate": 3.293478260869565e-05,
      "loss": 0.3049,
      "step": 1540
    },
    {
      "epoch": 16.85,
      "learning_rate": 3.1847826086956526e-05,
      "loss": 0.1622,
      "step": 1550
    },
    {
      "epoch": 16.96,
      "learning_rate": 3.0760869565217395e-05,
      "loss": 0.2375,
      "step": 1560
    },
    {
      "epoch": 17.07,
      "learning_rate": 2.967391304347826e-05,
      "loss": 0.123,
      "step": 1570
    },
    {
      "epoch": 17.17,
      "learning_rate": 2.8586956521739135e-05,
      "loss": 0.1922,
      "step": 1580
    },
    {
      "epoch": 17.28,
      "learning_rate": 2.7500000000000004e-05,
      "loss": 0.2794,
      "step": 1590
    },
    {
      "epoch": 17.39,
      "learning_rate": 2.6413043478260867e-05,
      "loss": 0.2183,
      "step": 1600
    },
    {
      "epoch": 17.39,
      "eval_accuracy": 0.9615384615384616,
      "eval_loss": 0.13765688240528107,
      "eval_runtime": 2.3734,
      "eval_samples_per_second": 109.546,
      "eval_steps_per_second": 13.904,
      "step": 1600
    },
    {
      "epoch": 17.5,
      "learning_rate": 2.5326086956521737e-05,
      "loss": 0.1748,
      "step": 1610
    },
    {
      "epoch": 17.61,
      "learning_rate": 2.423913043478261e-05,
      "loss": 0.2382,
      "step": 1620
    },
    {
      "epoch": 17.72,
      "learning_rate": 2.315217391304348e-05,
      "loss": 0.2712,
      "step": 1630
    },
    {
      "epoch": 17.83,
      "learning_rate": 2.206521739130435e-05,
      "loss": 0.2208,
      "step": 1640
    },
    {
      "epoch": 17.93,
      "learning_rate": 2.097826086956522e-05,
      "loss": 0.1639,
      "step": 1650
    },
    {
      "epoch": 18.04,
      "learning_rate": 1.9891304347826085e-05,
      "loss": 0.1006,
      "step": 1660
    },
    {
      "epoch": 18.15,
      "learning_rate": 1.8804347826086958e-05,
      "loss": 0.1697,
      "step": 1670
    },
    {
      "epoch": 18.26,
      "learning_rate": 1.7717391304347828e-05,
      "loss": 0.1967,
      "step": 1680
    },
    {
      "epoch": 18.37,
      "learning_rate": 1.6630434782608698e-05,
      "loss": 0.1897,
      "step": 1690
    },
    {
      "epoch": 18.48,
      "learning_rate": 1.5543478260869564e-05,
      "loss": 0.1478,
      "step": 1700
    },
    {
      "epoch": 18.48,
      "eval_accuracy": 0.9769230769230769,
      "eval_loss": 0.12689848244190216,
      "eval_runtime": 1.8016,
      "eval_samples_per_second": 144.314,
      "eval_steps_per_second": 18.317,
      "step": 1700
    },
    {
      "epoch": 18.59,
      "learning_rate": 1.4456521739130435e-05,
      "loss": 0.1515,
      "step": 1710
    },
    {
      "epoch": 18.7,
      "learning_rate": 1.3369565217391305e-05,
      "loss": 0.2222,
      "step": 1720
    },
    {
      "epoch": 18.8,
      "learning_rate": 1.2282608695652175e-05,
      "loss": 0.201,
      "step": 1730
    },
    {
      "epoch": 18.91,
      "learning_rate": 1.1195652173913044e-05,
      "loss": 0.1389,
      "step": 1740
    },
    {
      "epoch": 19.02,
      "learning_rate": 1.0108695652173914e-05,
      "loss": 0.1495,
      "step": 1750
    },
    {
      "epoch": 19.13,
      "learning_rate": 9.021739130434784e-06,
      "loss": 0.2616,
      "step": 1760
    },
    {
      "epoch": 19.24,
      "learning_rate": 7.934782608695653e-06,
      "loss": 0.1864,
      "step": 1770
    },
    {
      "epoch": 19.35,
      "learning_rate": 6.847826086956521e-06,
      "loss": 0.1395,
      "step": 1780
    },
    {
      "epoch": 19.46,
      "learning_rate": 5.760869565217392e-06,
      "loss": 0.1486,
      "step": 1790
    },
    {
      "epoch": 19.57,
      "learning_rate": 4.673913043478261e-06,
      "loss": 0.1944,
      "step": 1800
    },
    {
      "epoch": 19.57,
      "eval_accuracy": 0.9769230769230769,
      "eval_loss": 0.09090477973222733,
      "eval_runtime": 2.3489,
      "eval_samples_per_second": 110.691,
      "eval_steps_per_second": 14.049,
      "step": 1800
    },
    {
      "epoch": 19.67,
      "learning_rate": 3.5869565217391305e-06,
      "loss": 0.1359,
      "step": 1810
    },
    {
      "epoch": 19.78,
      "learning_rate": 2.5e-06,
      "loss": 0.143,
      "step": 1820
    },
    {
      "epoch": 19.89,
      "learning_rate": 1.4130434782608697e-06,
      "loss": 0.1368,
      "step": 1830
    },
    {
      "epoch": 20.0,
      "learning_rate": 3.260869565217391e-07,
      "loss": 0.162,
      "step": 1840
    },
    {
      "epoch": 20.0,
      "step": 1840,
      "total_flos": 7.30261234607063e+17,
      "train_loss": 0.5132673807766126,
      "train_runtime": 395.7262,
      "train_samples_per_second": 74.193,
      "train_steps_per_second": 4.65
    }
  ],
  "logging_steps": 10,
  "max_steps": 1840,
  "num_train_epochs": 20,
  "save_steps": 100,
  "total_flos": 7.30261234607063e+17,
  "trial_name": null,
  "trial_params": null
}