{
  "best_metric": 0.7797476986021139,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-img_orientation/checkpoint-2060",
  "epoch": 4.993939393939394,
  "eval_steps": 500,
  "global_step": 2060,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 2.4271844660194174e-06,
      "loss": 0.5485,
      "step": 10
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.854368932038835e-06,
      "loss": 0.524,
      "step": 20
    },
    {
      "epoch": 0.07,
      "learning_rate": 7.281553398058253e-06,
      "loss": 0.5436,
      "step": 30
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.70873786407767e-06,
      "loss": 0.5309,
      "step": 40
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.2135922330097088e-05,
      "loss": 0.5218,
      "step": 50
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.4563106796116505e-05,
      "loss": 0.4976,
      "step": 60
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.6990291262135926e-05,
      "loss": 0.5168,
      "step": 70
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.941747572815534e-05,
      "loss": 0.5168,
      "step": 80
    },
    {
      "epoch": 0.22,
      "learning_rate": 2.1844660194174756e-05,
      "loss": 0.4882,
      "step": 90
    },
    {
      "epoch": 0.24,
      "learning_rate": 2.4271844660194176e-05,
      "loss": 0.5174,
      "step": 100
    },
    {
      "epoch": 0.27,
      "learning_rate": 2.6699029126213593e-05,
      "loss": 0.5012,
      "step": 110
    },
    {
      "epoch": 0.29,
      "learning_rate": 2.912621359223301e-05,
      "loss": 0.462,
      "step": 120
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.155339805825243e-05,
      "loss": 0.4785,
      "step": 130
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.398058252427185e-05,
      "loss": 0.47,
      "step": 140
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.6407766990291265e-05,
      "loss": 0.4878,
      "step": 150
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.883495145631068e-05,
      "loss": 0.4648,
      "step": 160
    },
    {
      "epoch": 0.41,
      "learning_rate": 4.12621359223301e-05,
      "loss": 0.4636,
      "step": 170
    },
    {
      "epoch": 0.44,
      "learning_rate": 4.368932038834951e-05,
      "loss": 0.4862,
      "step": 180
    },
    {
      "epoch": 0.46,
      "learning_rate": 4.611650485436894e-05,
      "loss": 0.4594,
      "step": 190
    },
    {
      "epoch": 0.48,
      "learning_rate": 4.854368932038835e-05,
      "loss": 0.4882,
      "step": 200
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.9892125134843584e-05,
      "loss": 0.4429,
      "step": 210
    },
    {
      "epoch": 0.53,
      "learning_rate": 4.962243797195254e-05,
      "loss": 0.4683,
      "step": 220
    },
    {
      "epoch": 0.56,
      "learning_rate": 4.935275080906149e-05,
      "loss": 0.4689,
      "step": 230
    },
    {
      "epoch": 0.58,
      "learning_rate": 4.9083063646170444e-05,
      "loss": 0.4786,
      "step": 240
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.88133764832794e-05,
      "loss": 0.5182,
      "step": 250
    },
    {
      "epoch": 0.63,
      "learning_rate": 4.854368932038835e-05,
      "loss": 0.4701,
      "step": 260
    },
    {
      "epoch": 0.65,
      "learning_rate": 4.8274002157497304e-05,
      "loss": 0.4795,
      "step": 270
    },
    {
      "epoch": 0.68,
      "learning_rate": 4.800431499460626e-05,
      "loss": 0.4394,
      "step": 280
    },
    {
      "epoch": 0.7,
      "learning_rate": 4.773462783171521e-05,
      "loss": 0.4252,
      "step": 290
    },
    {
      "epoch": 0.73,
      "learning_rate": 4.7464940668824164e-05,
      "loss": 0.4688,
      "step": 300
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.7195253505933115e-05,
      "loss": 0.4879,
      "step": 310
    },
    {
      "epoch": 0.78,
      "learning_rate": 4.692556634304207e-05,
      "loss": 0.4517,
      "step": 320
    },
    {
      "epoch": 0.8,
      "learning_rate": 4.665587918015103e-05,
      "loss": 0.4492,
      "step": 330
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.638619201725998e-05,
      "loss": 0.4738,
      "step": 340
    },
    {
      "epoch": 0.85,
      "learning_rate": 4.611650485436894e-05,
      "loss": 0.4579,
      "step": 350
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.584681769147789e-05,
      "loss": 0.4413,
      "step": 360
    },
    {
      "epoch": 0.9,
      "learning_rate": 4.557713052858684e-05,
      "loss": 0.508,
      "step": 370
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.530744336569579e-05,
      "loss": 0.4605,
      "step": 380
    },
    {
      "epoch": 0.95,
      "learning_rate": 4.503775620280475e-05,
      "loss": 0.4459,
      "step": 390
    },
    {
      "epoch": 0.97,
      "learning_rate": 4.47680690399137e-05,
      "loss": 0.4568,
      "step": 400
    },
    {
      "epoch": 0.99,
      "learning_rate": 4.449838187702265e-05,
      "loss": 0.4741,
      "step": 410
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7345721104670986,
      "eval_loss": 0.47738537192344666,
      "eval_runtime": 44.109,
      "eval_samples_per_second": 132.989,
      "eval_steps_per_second": 4.171,
      "step": 412
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.422869471413161e-05,
      "loss": 0.4517,
      "step": 420
    },
    {
      "epoch": 1.04,
      "learning_rate": 4.395900755124056e-05,
      "loss": 0.4245,
      "step": 430
    },
    {
      "epoch": 1.07,
      "learning_rate": 4.368932038834951e-05,
      "loss": 0.4318,
      "step": 440
    },
    {
      "epoch": 1.09,
      "learning_rate": 4.341963322545847e-05,
      "loss": 0.4309,
      "step": 450
    },
    {
      "epoch": 1.12,
      "learning_rate": 4.314994606256743e-05,
      "loss": 0.4387,
      "step": 460
    },
    {
      "epoch": 1.14,
      "learning_rate": 4.288025889967638e-05,
      "loss": 0.4605,
      "step": 470
    },
    {
      "epoch": 1.16,
      "learning_rate": 4.261057173678533e-05,
      "loss": 0.4623,
      "step": 480
    },
    {
      "epoch": 1.19,
      "learning_rate": 4.234088457389429e-05,
      "loss": 0.4471,
      "step": 490
    },
    {
      "epoch": 1.21,
      "learning_rate": 4.207119741100324e-05,
      "loss": 0.4685,
      "step": 500
    },
    {
      "epoch": 1.24,
      "learning_rate": 4.180151024811219e-05,
      "loss": 0.4549,
      "step": 510
    },
    {
      "epoch": 1.26,
      "learning_rate": 4.153182308522115e-05,
      "loss": 0.4197,
      "step": 520
    },
    {
      "epoch": 1.28,
      "learning_rate": 4.12621359223301e-05,
      "loss": 0.4243,
      "step": 530
    },
    {
      "epoch": 1.31,
      "learning_rate": 4.099244875943905e-05,
      "loss": 0.4701,
      "step": 540
    },
    {
      "epoch": 1.33,
      "learning_rate": 4.0722761596548e-05,
      "loss": 0.4475,
      "step": 550
    },
    {
      "epoch": 1.36,
      "learning_rate": 4.0453074433656965e-05,
      "loss": 0.444,
      "step": 560
    },
    {
      "epoch": 1.38,
      "learning_rate": 4.0183387270765916e-05,
      "loss": 0.4231,
      "step": 570
    },
    {
      "epoch": 1.41,
      "learning_rate": 3.991370010787487e-05,
      "loss": 0.431,
      "step": 580
    },
    {
      "epoch": 1.43,
      "learning_rate": 3.964401294498382e-05,
      "loss": 0.4441,
      "step": 590
    },
    {
      "epoch": 1.45,
      "learning_rate": 3.9374325782092776e-05,
      "loss": 0.466,
      "step": 600
    },
    {
      "epoch": 1.48,
      "learning_rate": 3.910463861920173e-05,
      "loss": 0.4114,
      "step": 610
    },
    {
      "epoch": 1.5,
      "learning_rate": 3.883495145631068e-05,
      "loss": 0.4523,
      "step": 620
    },
    {
      "epoch": 1.53,
      "learning_rate": 3.8565264293419636e-05,
      "loss": 0.4632,
      "step": 630
    },
    {
      "epoch": 1.55,
      "learning_rate": 3.829557713052859e-05,
      "loss": 0.4373,
      "step": 640
    },
    {
      "epoch": 1.58,
      "learning_rate": 3.802588996763754e-05,
      "loss": 0.4487,
      "step": 650
    },
    {
      "epoch": 1.6,
      "learning_rate": 3.7756202804746496e-05,
      "loss": 0.4818,
      "step": 660
    },
    {
      "epoch": 1.62,
      "learning_rate": 3.7486515641855454e-05,
      "loss": 0.4461,
      "step": 670
    },
    {
      "epoch": 1.65,
      "learning_rate": 3.7216828478964405e-05,
      "loss": 0.4475,
      "step": 680
    },
    {
      "epoch": 1.67,
      "learning_rate": 3.6947141316073356e-05,
      "loss": 0.469,
      "step": 690
    },
    {
      "epoch": 1.7,
      "learning_rate": 3.6677454153182314e-05,
      "loss": 0.448,
      "step": 700
    },
    {
      "epoch": 1.72,
      "learning_rate": 3.6407766990291265e-05,
      "loss": 0.4411,
      "step": 710
    },
    {
      "epoch": 1.75,
      "learning_rate": 3.6138079827400216e-05,
      "loss": 0.4529,
      "step": 720
    },
    {
      "epoch": 1.77,
      "learning_rate": 3.5868392664509174e-05,
      "loss": 0.4553,
      "step": 730
    },
    {
      "epoch": 1.79,
      "learning_rate": 3.5598705501618125e-05,
      "loss": 0.4558,
      "step": 740
    },
    {
      "epoch": 1.82,
      "learning_rate": 3.5329018338727076e-05,
      "loss": 0.4976,
      "step": 750
    },
    {
      "epoch": 1.84,
      "learning_rate": 3.505933117583603e-05,
      "loss": 0.4474,
      "step": 760
    },
    {
      "epoch": 1.87,
      "learning_rate": 3.4789644012944984e-05,
      "loss": 0.4646,
      "step": 770
    },
    {
      "epoch": 1.89,
      "learning_rate": 3.451995685005394e-05,
      "loss": 0.4537,
      "step": 780
    },
    {
      "epoch": 1.92,
      "learning_rate": 3.425026968716289e-05,
      "loss": 0.4649,
      "step": 790
    },
    {
      "epoch": 1.94,
      "learning_rate": 3.398058252427185e-05,
      "loss": 0.4751,
      "step": 800
    },
    {
      "epoch": 1.96,
      "learning_rate": 3.37108953613808e-05,
      "loss": 0.4954,
      "step": 810
    },
    {
      "epoch": 1.99,
      "learning_rate": 3.344120819848975e-05,
      "loss": 0.4958,
      "step": 820
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7569041936583703,
      "eval_loss": 0.440324068069458,
      "eval_runtime": 44.2872,
      "eval_samples_per_second": 132.454,
      "eval_steps_per_second": 4.155,
      "step": 825
    },
    {
      "epoch": 2.01,
      "learning_rate": 3.3171521035598704e-05,
      "loss": 0.4511,
      "step": 830
    },
    {
      "epoch": 2.04,
      "learning_rate": 3.290183387270766e-05,
      "loss": 0.4175,
      "step": 840
    },
    {
      "epoch": 2.06,
      "learning_rate": 3.263214670981661e-05,
      "loss": 0.4511,
      "step": 850
    },
    {
      "epoch": 2.08,
      "learning_rate": 3.2362459546925564e-05,
      "loss": 0.4188,
      "step": 860
    },
    {
      "epoch": 2.11,
      "learning_rate": 3.209277238403452e-05,
      "loss": 0.453,
      "step": 870
    },
    {
      "epoch": 2.13,
      "learning_rate": 3.182308522114347e-05,
      "loss": 0.4392,
      "step": 880
    },
    {
      "epoch": 2.16,
      "learning_rate": 3.155339805825243e-05,
      "loss": 0.4585,
      "step": 890
    },
    {
      "epoch": 2.18,
      "learning_rate": 3.128371089536138e-05,
      "loss": 0.434,
      "step": 900
    },
    {
      "epoch": 2.21,
      "learning_rate": 3.101402373247034e-05,
      "loss": 0.4336,
      "step": 910
    },
    {
      "epoch": 2.23,
      "learning_rate": 3.074433656957929e-05,
      "loss": 0.446,
      "step": 920
    },
    {
      "epoch": 2.25,
      "learning_rate": 3.0474649406688245e-05,
      "loss": 0.4559,
      "step": 930
    },
    {
      "epoch": 2.28,
      "learning_rate": 3.0204962243797196e-05,
      "loss": 0.4529,
      "step": 940
    },
    {
      "epoch": 2.3,
      "learning_rate": 2.993527508090615e-05,
      "loss": 0.4681,
      "step": 950
    },
    {
      "epoch": 2.33,
      "learning_rate": 2.96655879180151e-05,
      "loss": 0.4512,
      "step": 960
    },
    {
      "epoch": 2.35,
      "learning_rate": 2.9395900755124056e-05,
      "loss": 0.4373,
      "step": 970
    },
    {
      "epoch": 2.38,
      "learning_rate": 2.912621359223301e-05,
      "loss": 0.4706,
      "step": 980
    },
    {
      "epoch": 2.4,
      "learning_rate": 2.885652642934196e-05,
      "loss": 0.4582,
      "step": 990
    },
    {
      "epoch": 2.42,
      "learning_rate": 2.8586839266450923e-05,
      "loss": 0.4442,
      "step": 1000
    },
    {
      "epoch": 2.45,
      "learning_rate": 2.8317152103559874e-05,
      "loss": 0.4593,
      "step": 1010
    },
    {
      "epoch": 2.47,
      "learning_rate": 2.8047464940668828e-05,
      "loss": 0.4525,
      "step": 1020
    },
    {
      "epoch": 2.5,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.4512,
      "step": 1030
    },
    {
      "epoch": 2.52,
      "learning_rate": 2.7508090614886734e-05,
      "loss": 0.4668,
      "step": 1040
    },
    {
      "epoch": 2.55,
      "learning_rate": 2.7238403451995688e-05,
      "loss": 0.4669,
      "step": 1050
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.696871628910464e-05,
      "loss": 0.4513,
      "step": 1060
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.6699029126213593e-05,
      "loss": 0.4727,
      "step": 1070
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.6429341963322544e-05,
      "loss": 0.4638,
      "step": 1080
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.61596548004315e-05,
      "loss": 0.473,
      "step": 1090
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.5889967637540453e-05,
      "loss": 0.4535,
      "step": 1100
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.5620280474649404e-05,
      "loss": 0.4874,
      "step": 1110
    },
    {
      "epoch": 2.72,
      "learning_rate": 2.5350593311758362e-05,
      "loss": 0.4746,
      "step": 1120
    },
    {
      "epoch": 2.74,
      "learning_rate": 2.5080906148867317e-05,
      "loss": 0.4474,
      "step": 1130
    },
    {
      "epoch": 2.76,
      "learning_rate": 2.481121898597627e-05,
      "loss": 0.4717,
      "step": 1140
    },
    {
      "epoch": 2.79,
      "learning_rate": 2.4541531823085222e-05,
      "loss": 0.4634,
      "step": 1150
    },
    {
      "epoch": 2.81,
      "learning_rate": 2.4271844660194176e-05,
      "loss": 0.478,
      "step": 1160
    },
    {
      "epoch": 2.84,
      "learning_rate": 2.400215749730313e-05,
      "loss": 0.5012,
      "step": 1170
    },
    {
      "epoch": 2.86,
      "learning_rate": 2.3732470334412082e-05,
      "loss": 0.4592,
      "step": 1180
    },
    {
      "epoch": 2.88,
      "learning_rate": 2.3462783171521036e-05,
      "loss": 0.519,
      "step": 1190
    },
    {
      "epoch": 2.91,
      "learning_rate": 2.319309600862999e-05,
      "loss": 0.5202,
      "step": 1200
    },
    {
      "epoch": 2.93,
      "learning_rate": 2.2923408845738945e-05,
      "loss": 0.4819,
      "step": 1210
    },
    {
      "epoch": 2.96,
      "learning_rate": 2.2653721682847896e-05,
      "loss": 0.4969,
      "step": 1220
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.238403451995685e-05,
      "loss": 0.473,
      "step": 1230
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7659393112853733,
      "eval_loss": 0.40928295254707336,
      "eval_runtime": 44.1748,
      "eval_samples_per_second": 132.79,
      "eval_steps_per_second": 4.165,
      "step": 1237
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.2114347357065805e-05,
      "loss": 0.4556,
      "step": 1240
    },
    {
      "epoch": 3.03,
      "learning_rate": 2.1844660194174756e-05,
      "loss": 0.4626,
      "step": 1250
    },
    {
      "epoch": 3.05,
      "learning_rate": 2.1574973031283714e-05,
      "loss": 0.4897,
      "step": 1260
    },
    {
      "epoch": 3.08,
      "learning_rate": 2.1305285868392665e-05,
      "loss": 0.4705,
      "step": 1270
    },
    {
      "epoch": 3.1,
      "learning_rate": 2.103559870550162e-05,
      "loss": 0.4876,
      "step": 1280
    },
    {
      "epoch": 3.13,
      "learning_rate": 2.0765911542610574e-05,
      "loss": 0.4844,
      "step": 1290
    },
    {
      "epoch": 3.15,
      "learning_rate": 2.0496224379719525e-05,
      "loss": 0.4737,
      "step": 1300
    },
    {
      "epoch": 3.18,
      "learning_rate": 2.0226537216828483e-05,
      "loss": 0.4978,
      "step": 1310
    },
    {
      "epoch": 3.2,
      "learning_rate": 1.9956850053937434e-05,
      "loss": 0.4744,
      "step": 1320
    },
    {
      "epoch": 3.22,
      "learning_rate": 1.9687162891046388e-05,
      "loss": 0.4588,
      "step": 1330
    },
    {
      "epoch": 3.25,
      "learning_rate": 1.941747572815534e-05,
      "loss": 0.4344,
      "step": 1340
    },
    {
      "epoch": 3.27,
      "learning_rate": 1.9147788565264294e-05,
      "loss": 0.4651,
      "step": 1350
    },
    {
      "epoch": 3.3,
      "learning_rate": 1.8878101402373248e-05,
      "loss": 0.4813,
      "step": 1360
    },
    {
      "epoch": 3.32,
      "learning_rate": 1.8608414239482202e-05,
      "loss": 0.4614,
      "step": 1370
    },
    {
      "epoch": 3.35,
      "learning_rate": 1.8338727076591157e-05,
      "loss": 0.4687,
      "step": 1380
    },
    {
      "epoch": 3.37,
      "learning_rate": 1.8069039913700108e-05,
      "loss": 0.4735,
      "step": 1390
    },
    {
      "epoch": 3.39,
      "learning_rate": 1.7799352750809062e-05,
      "loss": 0.4798,
      "step": 1400
    },
    {
      "epoch": 3.42,
      "learning_rate": 1.7529665587918013e-05,
      "loss": 0.4288,
      "step": 1410
    },
    {
      "epoch": 3.44,
      "learning_rate": 1.725997842502697e-05,
      "loss": 0.4592,
      "step": 1420
    },
    {
      "epoch": 3.47,
      "learning_rate": 1.6990291262135926e-05,
      "loss": 0.4694,
      "step": 1430
    },
    {
      "epoch": 3.49,
      "learning_rate": 1.6720604099244877e-05,
      "loss": 0.4409,
      "step": 1440
    },
    {
      "epoch": 3.52,
      "learning_rate": 1.645091693635383e-05,
      "loss": 0.4536,
      "step": 1450
    },
    {
      "epoch": 3.54,
      "learning_rate": 1.6181229773462782e-05,
      "loss": 0.4635,
      "step": 1460
    },
    {
      "epoch": 3.56,
      "learning_rate": 1.5911542610571736e-05,
      "loss": 0.4554,
      "step": 1470
    },
    {
      "epoch": 3.59,
      "learning_rate": 1.564185544768069e-05,
      "loss": 0.4504,
      "step": 1480
    },
    {
      "epoch": 3.61,
      "learning_rate": 1.5372168284789645e-05,
      "loss": 0.445,
      "step": 1490
    },
    {
      "epoch": 3.64,
      "learning_rate": 1.5102481121898598e-05,
      "loss": 0.4476,
      "step": 1500
    },
    {
      "epoch": 3.66,
      "learning_rate": 1.483279395900755e-05,
      "loss": 0.4828,
      "step": 1510
    },
    {
      "epoch": 3.68,
      "learning_rate": 1.4563106796116505e-05,
      "loss": 0.4698,
      "step": 1520
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.4293419633225461e-05,
      "loss": 0.4603,
      "step": 1530
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.4023732470334414e-05,
      "loss": 0.4558,
      "step": 1540
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.3754045307443367e-05,
      "loss": 0.4632,
      "step": 1550
    },
    {
      "epoch": 3.78,
      "learning_rate": 1.348435814455232e-05,
      "loss": 0.4785,
      "step": 1560
    },
    {
      "epoch": 3.81,
      "learning_rate": 1.3214670981661272e-05,
      "loss": 0.4593,
      "step": 1570
    },
    {
      "epoch": 3.83,
      "learning_rate": 1.2944983818770227e-05,
      "loss": 0.439,
      "step": 1580
    },
    {
      "epoch": 3.85,
      "learning_rate": 1.2675296655879181e-05,
      "loss": 0.437,
      "step": 1590
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.2405609492988135e-05,
      "loss": 0.4649,
      "step": 1600
    },
    {
      "epoch": 3.9,
      "learning_rate": 1.2135922330097088e-05,
      "loss": 0.4976,
      "step": 1610
    },
    {
      "epoch": 3.93,
      "learning_rate": 1.1866235167206041e-05,
      "loss": 0.4667,
      "step": 1620
    },
    {
      "epoch": 3.95,
      "learning_rate": 1.1596548004314995e-05,
      "loss": 0.4591,
      "step": 1630
    },
    {
      "epoch": 3.98,
      "learning_rate": 1.1326860841423948e-05,
      "loss": 0.4113,
      "step": 1640
    },
    {
      "epoch": 4.0,
      "learning_rate": 1.1057173678532903e-05,
      "loss": 0.4476,
      "step": 1650
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7734401636549608,
      "eval_loss": 0.4010549485683441,
      "eval_runtime": 43.1679,
      "eval_samples_per_second": 135.888,
      "eval_steps_per_second": 4.262,
      "step": 1650
    },
    {
      "epoch": 4.02,
      "learning_rate": 1.0787486515641857e-05,
      "loss": 0.4337,
      "step": 1660
    },
    {
      "epoch": 4.05,
      "learning_rate": 1.051779935275081e-05,
      "loss": 0.429,
      "step": 1670
    },
    {
      "epoch": 4.07,
      "learning_rate": 1.0248112189859762e-05,
      "loss": 0.4329,
      "step": 1680
    },
    {
      "epoch": 4.1,
      "learning_rate": 9.978425026968717e-06,
      "loss": 0.451,
      "step": 1690
    },
    {
      "epoch": 4.12,
      "learning_rate": 9.70873786407767e-06,
      "loss": 0.4556,
      "step": 1700
    },
    {
      "epoch": 4.15,
      "learning_rate": 9.439050701186624e-06,
      "loss": 0.466,
      "step": 1710
    },
    {
      "epoch": 4.17,
      "learning_rate": 9.169363538295578e-06,
      "loss": 0.4192,
      "step": 1720
    },
    {
      "epoch": 4.19,
      "learning_rate": 8.899676375404531e-06,
      "loss": 0.4304,
      "step": 1730
    },
    {
      "epoch": 4.22,
      "learning_rate": 8.629989212513486e-06,
      "loss": 0.4474,
      "step": 1740
    },
    {
      "epoch": 4.24,
      "learning_rate": 8.360302049622438e-06,
      "loss": 0.4011,
      "step": 1750
    },
    {
      "epoch": 4.27,
      "learning_rate": 8.090614886731391e-06,
      "loss": 0.4566,
      "step": 1760
    },
    {
      "epoch": 4.29,
      "learning_rate": 7.820927723840345e-06,
      "loss": 0.417,
      "step": 1770
    },
    {
      "epoch": 4.32,
      "learning_rate": 7.551240560949299e-06,
      "loss": 0.4359,
      "step": 1780
    },
    {
      "epoch": 4.34,
      "learning_rate": 7.281553398058253e-06,
      "loss": 0.4322,
      "step": 1790
    },
    {
      "epoch": 4.36,
      "learning_rate": 7.011866235167207e-06,
      "loss": 0.4457,
      "step": 1800
    },
    {
      "epoch": 4.39,
      "learning_rate": 6.74217907227616e-06,
      "loss": 0.4896,
      "step": 1810
    },
    {
      "epoch": 4.41,
      "learning_rate": 6.472491909385113e-06,
      "loss": 0.4384,
      "step": 1820
    },
    {
      "epoch": 4.44,
      "learning_rate": 6.202804746494068e-06,
      "loss": 0.4464,
      "step": 1830
    },
    {
      "epoch": 4.46,
      "learning_rate": 5.9331175836030205e-06,
      "loss": 0.4671,
      "step": 1840
    },
    {
      "epoch": 4.48,
      "learning_rate": 5.663430420711974e-06,
      "loss": 0.4436,
      "step": 1850
    },
    {
      "epoch": 4.51,
      "learning_rate": 5.3937432578209285e-06,
      "loss": 0.4403,
      "step": 1860
    },
    {
      "epoch": 4.53,
      "learning_rate": 5.124056094929881e-06,
      "loss": 0.4329,
      "step": 1870
    },
    {
      "epoch": 4.56,
      "learning_rate": 4.854368932038835e-06,
      "loss": 0.4651,
      "step": 1880
    },
    {
      "epoch": 4.58,
      "learning_rate": 4.584681769147789e-06,
      "loss": 0.424,
      "step": 1890
    },
    {
      "epoch": 4.61,
      "learning_rate": 4.314994606256743e-06,
      "loss": 0.4539,
      "step": 1900
    },
    {
      "epoch": 4.63,
      "learning_rate": 4.0453074433656955e-06,
      "loss": 0.4237,
      "step": 1910
    },
    {
      "epoch": 4.65,
      "learning_rate": 3.7756202804746495e-06,
      "loss": 0.4337,
      "step": 1920
    },
    {
      "epoch": 4.68,
      "learning_rate": 3.5059331175836035e-06,
      "loss": 0.4322,
      "step": 1930
    },
    {
      "epoch": 4.7,
      "learning_rate": 3.2362459546925567e-06,
      "loss": 0.441,
      "step": 1940
    },
    {
      "epoch": 4.73,
      "learning_rate": 2.9665587918015102e-06,
      "loss": 0.4151,
      "step": 1950
    },
    {
      "epoch": 4.75,
      "learning_rate": 2.6968716289104642e-06,
      "loss": 0.4437,
      "step": 1960
    },
    {
      "epoch": 4.78,
      "learning_rate": 2.4271844660194174e-06,
      "loss": 0.4187,
      "step": 1970
    },
    {
      "epoch": 4.8,
      "learning_rate": 2.1574973031283714e-06,
      "loss": 0.4216,
      "step": 1980
    },
    {
      "epoch": 4.82,
      "learning_rate": 1.8878101402373248e-06,
      "loss": 0.4153,
      "step": 1990
    },
    {
      "epoch": 4.85,
      "learning_rate": 1.6181229773462783e-06,
      "loss": 0.4495,
      "step": 2000
    },
    {
      "epoch": 4.87,
      "learning_rate": 1.3484358144552321e-06,
      "loss": 0.4395,
      "step": 2010
    },
    {
      "epoch": 4.9,
      "learning_rate": 1.0787486515641857e-06,
      "loss": 0.4226,
      "step": 2020
    },
    {
      "epoch": 4.92,
      "learning_rate": 8.090614886731392e-07,
      "loss": 0.4334,
      "step": 2030
    },
    {
      "epoch": 4.95,
      "learning_rate": 5.393743257820928e-07,
      "loss": 0.4169,
      "step": 2040
    },
    {
      "epoch": 4.97,
      "learning_rate": 2.696871628910464e-07,
      "loss": 0.4196,
      "step": 2050
    },
    {
      "epoch": 4.99,
      "learning_rate": 0.0,
      "loss": 0.4093,
      "step": 2060
    },
    {
      "epoch": 4.99,
      "eval_accuracy": 0.7797476986021139,
      "eval_loss": 0.3965856432914734,
      "eval_runtime": 45.954,
      "eval_samples_per_second": 127.649,
      "eval_steps_per_second": 4.004,
      "step": 2060
    },
    {
      "epoch": 4.99,
      "step": 2060,
      "total_flos": 6.553200292249928e+18,
      "train_loss": 0.4588474294514332,
      "train_runtime": 4338.0094,
      "train_samples_per_second": 60.844,
      "train_steps_per_second": 0.475
    }
  ],
  "logging_steps": 10,
  "max_steps": 2060,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 6.553200292249928e+18,
  "trial_name": null,
  "trial_params": null
}