{
  "best_metric": 0.9161991584852734,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-classification/checkpoint-802",
  "epoch": 9.975062344139651,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "grad_norm": 4.872883319854736,
      "learning_rate": 1e-05,
      "loss": 0.2252,
      "step": 10
    },
    {
      "epoch": 0.2,
      "grad_norm": 4.769281387329102,
      "learning_rate": 2e-05,
      "loss": 0.2281,
      "step": 20
    },
    {
      "epoch": 0.3,
      "grad_norm": 4.990223407745361,
      "learning_rate": 3e-05,
      "loss": 0.2281,
      "step": 30
    },
    {
      "epoch": 0.4,
      "grad_norm": 4.2269463539123535,
      "learning_rate": 4e-05,
      "loss": 0.2359,
      "step": 40
    },
    {
      "epoch": 0.5,
      "grad_norm": 5.273340702056885,
      "learning_rate": 5e-05,
      "loss": 0.2345,
      "step": 50
    },
    {
      "epoch": 0.6,
      "grad_norm": 6.166986465454102,
      "learning_rate": 6e-05,
      "loss": 0.2625,
      "step": 60
    },
    {
      "epoch": 0.7,
      "grad_norm": 5.066626071929932,
      "learning_rate": 7e-05,
      "loss": 0.2738,
      "step": 70
    },
    {
      "epoch": 0.8,
      "grad_norm": 4.902966499328613,
      "learning_rate": 8e-05,
      "loss": 0.2793,
      "step": 80
    },
    {
      "epoch": 0.9,
      "grad_norm": 7.076936721801758,
      "learning_rate": 9e-05,
      "loss": 0.3018,
      "step": 90
    },
    {
      "epoch": 1.0,
      "grad_norm": 7.968544960021973,
      "learning_rate": 0.0001,
      "loss": 0.3193,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8716690042075736,
      "eval_loss": 0.3754436671733856,
      "eval_runtime": 39.3279,
      "eval_samples_per_second": 72.518,
      "eval_steps_per_second": 1.144,
      "step": 100
    },
    {
      "epoch": 1.1,
      "grad_norm": 6.099853992462158,
      "learning_rate": 9.888888888888889e-05,
      "loss": 0.2997,
      "step": 110
    },
    {
      "epoch": 1.2,
      "grad_norm": 8.707666397094727,
      "learning_rate": 9.777777777777778e-05,
      "loss": 0.3066,
      "step": 120
    },
    {
      "epoch": 1.3,
      "grad_norm": 6.50078010559082,
      "learning_rate": 9.666666666666667e-05,
      "loss": 0.2916,
      "step": 130
    },
    {
      "epoch": 1.4,
      "grad_norm": 6.648171901702881,
      "learning_rate": 9.555555555555557e-05,
      "loss": 0.318,
      "step": 140
    },
    {
      "epoch": 1.5,
      "grad_norm": 5.3696746826171875,
      "learning_rate": 9.444444444444444e-05,
      "loss": 0.2995,
      "step": 150
    },
    {
      "epoch": 1.6,
      "grad_norm": 4.172982692718506,
      "learning_rate": 9.333333333333334e-05,
      "loss": 0.2971,
      "step": 160
    },
    {
      "epoch": 1.7,
      "grad_norm": 5.951164245605469,
      "learning_rate": 9.222222222222223e-05,
      "loss": 0.3165,
      "step": 170
    },
    {
      "epoch": 1.8,
      "grad_norm": 5.611052513122559,
      "learning_rate": 9.111111111111112e-05,
      "loss": 0.2885,
      "step": 180
    },
    {
      "epoch": 1.9,
      "grad_norm": 6.186773777008057,
      "learning_rate": 9e-05,
      "loss": 0.2863,
      "step": 190
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.462760925292969,
      "learning_rate": 8.888888888888889e-05,
      "loss": 0.292,
      "step": 200
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.864656381486676,
      "eval_loss": 0.4085911512374878,
      "eval_runtime": 38.7867,
      "eval_samples_per_second": 73.53,
      "eval_steps_per_second": 1.16,
      "step": 200
    },
    {
      "epoch": 2.09,
      "grad_norm": 6.041274547576904,
      "learning_rate": 8.777777777777778e-05,
      "loss": 0.2882,
      "step": 210
    },
    {
      "epoch": 2.19,
      "grad_norm": 5.336978435516357,
      "learning_rate": 8.666666666666667e-05,
      "loss": 0.269,
      "step": 220
    },
    {
      "epoch": 2.29,
      "grad_norm": 5.773736953735352,
      "learning_rate": 8.555555555555556e-05,
      "loss": 0.2675,
      "step": 230
    },
    {
      "epoch": 2.39,
      "grad_norm": 5.7196784019470215,
      "learning_rate": 8.444444444444444e-05,
      "loss": 0.256,
      "step": 240
    },
    {
      "epoch": 2.49,
      "grad_norm": 4.928184986114502,
      "learning_rate": 8.333333333333334e-05,
      "loss": 0.2463,
      "step": 250
    },
    {
      "epoch": 2.59,
      "grad_norm": 7.451691627502441,
      "learning_rate": 8.222222222222222e-05,
      "loss": 0.2677,
      "step": 260
    },
    {
      "epoch": 2.69,
      "grad_norm": 5.755824089050293,
      "learning_rate": 8.111111111111112e-05,
      "loss": 0.2507,
      "step": 270
    },
    {
      "epoch": 2.79,
      "grad_norm": 5.09450101852417,
      "learning_rate": 8e-05,
      "loss": 0.2766,
      "step": 280
    },
    {
      "epoch": 2.89,
      "grad_norm": 5.240322589874268,
      "learning_rate": 7.88888888888889e-05,
      "loss": 0.2674,
      "step": 290
    },
    {
      "epoch": 2.99,
      "grad_norm": 4.865029811859131,
      "learning_rate": 7.777777777777778e-05,
      "loss": 0.2531,
      "step": 300
    },
    {
      "epoch": 2.99,
      "eval_accuracy": 0.894109396914446,
      "eval_loss": 0.34070128202438354,
      "eval_runtime": 39.4778,
      "eval_samples_per_second": 72.243,
      "eval_steps_per_second": 1.14,
      "step": 300
    },
    {
      "epoch": 3.09,
      "grad_norm": 4.403306007385254,
      "learning_rate": 7.666666666666667e-05,
      "loss": 0.2141,
      "step": 310
    },
    {
      "epoch": 3.19,
      "grad_norm": 4.91873836517334,
      "learning_rate": 7.555555555555556e-05,
      "loss": 0.2257,
      "step": 320
    },
    {
      "epoch": 3.29,
      "grad_norm": 5.522466659545898,
      "learning_rate": 7.444444444444444e-05,
      "loss": 0.2419,
      "step": 330
    },
    {
      "epoch": 3.39,
      "grad_norm": 4.829797267913818,
      "learning_rate": 7.333333333333333e-05,
      "loss": 0.2522,
      "step": 340
    },
    {
      "epoch": 3.49,
      "grad_norm": 6.162938117980957,
      "learning_rate": 7.222222222222222e-05,
      "loss": 0.2533,
      "step": 350
    },
    {
      "epoch": 3.59,
      "grad_norm": 5.2231340408325195,
      "learning_rate": 7.111111111111112e-05,
      "loss": 0.2284,
      "step": 360
    },
    {
      "epoch": 3.69,
      "grad_norm": 4.9288716316223145,
      "learning_rate": 7e-05,
      "loss": 0.244,
      "step": 370
    },
    {
      "epoch": 3.79,
      "grad_norm": 4.194519996643066,
      "learning_rate": 6.88888888888889e-05,
      "loss": 0.2433,
      "step": 380
    },
    {
      "epoch": 3.89,
      "grad_norm": 5.996556282043457,
      "learning_rate": 6.777777777777778e-05,
      "loss": 0.2428,
      "step": 390
    },
    {
      "epoch": 3.99,
      "grad_norm": 5.954503536224365,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.2242,
      "step": 400
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8927068723702665,
      "eval_loss": 0.33171677589416504,
      "eval_runtime": 39.2032,
      "eval_samples_per_second": 72.749,
      "eval_steps_per_second": 1.148,
      "step": 401
    },
    {
      "epoch": 4.09,
      "grad_norm": 5.060131072998047,
      "learning_rate": 6.555555555555556e-05,
      "loss": 0.209,
      "step": 410
    },
    {
      "epoch": 4.19,
      "grad_norm": 5.29101037979126,
      "learning_rate": 6.444444444444446e-05,
      "loss": 0.2045,
      "step": 420
    },
    {
      "epoch": 4.29,
      "grad_norm": 7.005785942077637,
      "learning_rate": 6.333333333333333e-05,
      "loss": 0.2063,
      "step": 430
    },
    {
      "epoch": 4.39,
      "grad_norm": 3.808072328567505,
      "learning_rate": 6.222222222222222e-05,
      "loss": 0.1935,
      "step": 440
    },
    {
      "epoch": 4.49,
      "grad_norm": 4.201041221618652,
      "learning_rate": 6.111111111111112e-05,
      "loss": 0.2062,
      "step": 450
    },
    {
      "epoch": 4.59,
      "grad_norm": 5.4039835929870605,
      "learning_rate": 6e-05,
      "loss": 0.1949,
      "step": 460
    },
    {
      "epoch": 4.69,
      "grad_norm": 6.116047382354736,
      "learning_rate": 5.8888888888888896e-05,
      "loss": 0.2042,
      "step": 470
    },
    {
      "epoch": 4.79,
      "grad_norm": 4.799409866333008,
      "learning_rate": 5.7777777777777776e-05,
      "loss": 0.2135,
      "step": 480
    },
    {
      "epoch": 4.89,
      "grad_norm": 5.615736484527588,
      "learning_rate": 5.666666666666667e-05,
      "loss": 0.2098,
      "step": 490
    },
    {
      "epoch": 4.99,
      "grad_norm": 4.442112922668457,
      "learning_rate": 5.555555555555556e-05,
      "loss": 0.2181,
      "step": 500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8983169705469846,
      "eval_loss": 0.3073332607746124,
      "eval_runtime": 39.2249,
      "eval_samples_per_second": 72.709,
      "eval_steps_per_second": 1.147,
      "step": 501
    },
    {
      "epoch": 5.09,
      "grad_norm": 3.2966055870056152,
      "learning_rate": 5.4444444444444446e-05,
      "loss": 0.1809,
      "step": 510
    },
    {
      "epoch": 5.19,
      "grad_norm": 3.9496381282806396,
      "learning_rate": 5.333333333333333e-05,
      "loss": 0.1904,
      "step": 520
    },
    {
      "epoch": 5.29,
      "grad_norm": 4.369997978210449,
      "learning_rate": 5.222222222222223e-05,
      "loss": 0.1817,
      "step": 530
    },
    {
      "epoch": 5.39,
      "grad_norm": 4.793045520782471,
      "learning_rate": 5.111111111111111e-05,
      "loss": 0.1889,
      "step": 540
    },
    {
      "epoch": 5.49,
      "grad_norm": 3.8780226707458496,
      "learning_rate": 5e-05,
      "loss": 0.1713,
      "step": 550
    },
    {
      "epoch": 5.59,
      "grad_norm": 5.775850296020508,
      "learning_rate": 4.888888888888889e-05,
      "loss": 0.1916,
      "step": 560
    },
    {
      "epoch": 5.69,
      "grad_norm": 4.524603843688965,
      "learning_rate": 4.7777777777777784e-05,
      "loss": 0.1771,
      "step": 570
    },
    {
      "epoch": 5.79,
      "grad_norm": 4.741029262542725,
      "learning_rate": 4.666666666666667e-05,
      "loss": 0.1722,
      "step": 580
    },
    {
      "epoch": 5.89,
      "grad_norm": 4.951327323913574,
      "learning_rate": 4.555555555555556e-05,
      "loss": 0.156,
      "step": 590
    },
    {
      "epoch": 5.99,
      "grad_norm": 3.7907869815826416,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.1829,
      "step": 600
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9032258064516129,
      "eval_loss": 0.31035441160202026,
      "eval_runtime": 38.6346,
      "eval_samples_per_second": 73.82,
      "eval_steps_per_second": 1.165,
      "step": 601
    },
    {
      "epoch": 6.08,
      "grad_norm": 4.384369850158691,
      "learning_rate": 4.3333333333333334e-05,
      "loss": 0.1656,
      "step": 610
    },
    {
      "epoch": 6.18,
      "grad_norm": 5.536717891693115,
      "learning_rate": 4.222222222222222e-05,
      "loss": 0.1573,
      "step": 620
    },
    {
      "epoch": 6.28,
      "grad_norm": 5.531068801879883,
      "learning_rate": 4.111111111111111e-05,
      "loss": 0.1585,
      "step": 630
    },
    {
      "epoch": 6.38,
      "grad_norm": 4.487759590148926,
      "learning_rate": 4e-05,
      "loss": 0.1618,
      "step": 640
    },
    {
      "epoch": 6.48,
      "grad_norm": 4.500083923339844,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.1567,
      "step": 650
    },
    {
      "epoch": 6.58,
      "grad_norm": 4.1614298820495605,
      "learning_rate": 3.777777777777778e-05,
      "loss": 0.1548,
      "step": 660
    },
    {
      "epoch": 6.68,
      "grad_norm": 4.733586311340332,
      "learning_rate": 3.6666666666666666e-05,
      "loss": 0.1607,
      "step": 670
    },
    {
      "epoch": 6.78,
      "grad_norm": 5.126717567443848,
      "learning_rate": 3.555555555555556e-05,
      "loss": 0.1542,
      "step": 680
    },
    {
      "epoch": 6.88,
      "grad_norm": 3.771245002746582,
      "learning_rate": 3.444444444444445e-05,
      "loss": 0.1455,
      "step": 690
    },
    {
      "epoch": 6.98,
      "grad_norm": 5.217066287994385,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.1426,
      "step": 700
    },
    {
      "epoch": 6.99,
      "eval_accuracy": 0.9049789621318373,
      "eval_loss": 0.29764774441719055,
      "eval_runtime": 39.4594,
      "eval_samples_per_second": 72.277,
      "eval_steps_per_second": 1.14,
      "step": 701
    },
    {
      "epoch": 7.08,
      "grad_norm": 6.066614627838135,
      "learning_rate": 3.222222222222223e-05,
      "loss": 0.1429,
      "step": 710
    },
    {
      "epoch": 7.18,
      "grad_norm": 3.363190174102783,
      "learning_rate": 3.111111111111111e-05,
      "loss": 0.1385,
      "step": 720
    },
    {
      "epoch": 7.28,
      "grad_norm": 5.055250644683838,
      "learning_rate": 3e-05,
      "loss": 0.1373,
      "step": 730
    },
    {
      "epoch": 7.38,
      "grad_norm": 5.0901384353637695,
      "learning_rate": 2.8888888888888888e-05,
      "loss": 0.1392,
      "step": 740
    },
    {
      "epoch": 7.48,
      "grad_norm": 3.7394046783447266,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1235,
      "step": 750
    },
    {
      "epoch": 7.58,
      "grad_norm": 4.67645263671875,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 0.1466,
      "step": 760
    },
    {
      "epoch": 7.68,
      "grad_norm": 4.7584381103515625,
      "learning_rate": 2.5555555555555554e-05,
      "loss": 0.1228,
      "step": 770
    },
    {
      "epoch": 7.78,
      "grad_norm": 5.486098289489746,
      "learning_rate": 2.4444444444444445e-05,
      "loss": 0.1355,
      "step": 780
    },
    {
      "epoch": 7.88,
      "grad_norm": 5.71726131439209,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 0.1299,
      "step": 790
    },
    {
      "epoch": 7.98,
      "grad_norm": 4.222195625305176,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1472,
      "step": 800
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9161991584852734,
      "eval_loss": 0.2736373543739319,
      "eval_runtime": 39.0364,
      "eval_samples_per_second": 73.06,
      "eval_steps_per_second": 1.153,
      "step": 802
    },
    {
      "epoch": 8.08,
      "grad_norm": 3.898097038269043,
      "learning_rate": 2.111111111111111e-05,
      "loss": 0.1214,
      "step": 810
    },
    {
      "epoch": 8.18,
      "grad_norm": 4.837826728820801,
      "learning_rate": 2e-05,
      "loss": 0.1176,
      "step": 820
    },
    {
      "epoch": 8.28,
      "grad_norm": 4.239289283752441,
      "learning_rate": 1.888888888888889e-05,
      "loss": 0.1191,
      "step": 830
    },
    {
      "epoch": 8.38,
      "grad_norm": 4.8459320068359375,
      "learning_rate": 1.777777777777778e-05,
      "loss": 0.1233,
      "step": 840
    },
    {
      "epoch": 8.48,
      "grad_norm": 4.943118095397949,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.1274,
      "step": 850
    },
    {
      "epoch": 8.58,
      "grad_norm": 3.3787436485290527,
      "learning_rate": 1.5555555555555555e-05,
      "loss": 0.1277,
      "step": 860
    },
    {
      "epoch": 8.68,
      "grad_norm": 4.009765148162842,
      "learning_rate": 1.4444444444444444e-05,
      "loss": 0.1044,
      "step": 870
    },
    {
      "epoch": 8.78,
      "grad_norm": 3.8422799110412598,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.1195,
      "step": 880
    },
    {
      "epoch": 8.88,
      "grad_norm": 2.927946090698242,
      "learning_rate": 1.2222222222222222e-05,
      "loss": 0.1077,
      "step": 890
    },
    {
      "epoch": 8.98,
      "grad_norm": 5.300693988800049,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.1314,
      "step": 900
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9133941093969145,
      "eval_loss": 0.27998465299606323,
      "eval_runtime": 39.0226,
      "eval_samples_per_second": 73.086,
      "eval_steps_per_second": 1.153,
      "step": 902
    },
    {
      "epoch": 9.08,
      "grad_norm": 4.2179107666015625,
      "learning_rate": 1e-05,
      "loss": 0.1172,
      "step": 910
    },
    {
      "epoch": 9.18,
      "grad_norm": 4.46065092086792,
      "learning_rate": 8.88888888888889e-06,
      "loss": 0.1076,
      "step": 920
    },
    {
      "epoch": 9.28,
      "grad_norm": 3.4499449729919434,
      "learning_rate": 7.777777777777777e-06,
      "loss": 0.1103,
      "step": 930
    },
    {
      "epoch": 9.38,
      "grad_norm": 4.1871418952941895,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.1041,
      "step": 940
    },
    {
      "epoch": 9.48,
      "grad_norm": 4.666812896728516,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.1175,
      "step": 950
    },
    {
      "epoch": 9.58,
      "grad_norm": 3.4452712535858154,
      "learning_rate": 4.444444444444445e-06,
      "loss": 0.1143,
      "step": 960
    },
    {
      "epoch": 9.68,
      "grad_norm": 4.585890293121338,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.1156,
      "step": 970
    },
    {
      "epoch": 9.78,
      "grad_norm": 3.672333240509033,
      "learning_rate": 2.2222222222222225e-06,
      "loss": 0.1157,
      "step": 980
    },
    {
      "epoch": 9.88,
      "grad_norm": 3.946244478225708,
      "learning_rate": 1.1111111111111112e-06,
      "loss": 0.1019,
      "step": 990
    },
    {
      "epoch": 9.98,
      "grad_norm": 2.6940207481384277,
      "learning_rate": 0.0,
      "loss": 0.1139,
      "step": 1000
    },
    {
      "epoch": 9.98,
      "eval_accuracy": 0.914796633941094,
      "eval_loss": 0.2748475670814514,
      "eval_runtime": 39.0773,
      "eval_samples_per_second": 72.984,
      "eval_steps_per_second": 1.152,
      "step": 1000
    },
    {
      "epoch": 9.98,
      "step": 1000,
      "total_flos": 6.364199987970048e+18,
      "train_loss": 0.19688368451595306,
      "train_runtime": 4893.4366,
      "train_samples_per_second": 52.446,
      "train_steps_per_second": 0.204
    }
  ],
  "logging_steps": 10,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 6.364199987970048e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}