{
  "best_metric": 0.027106858789920807,
  "best_model_checkpoint": "./deit-base-distilled-mask-finetuned/checkpoint-600",
  "epoch": 1.9955654101995566,
  "global_step": 900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 0.00019889135254988916,
      "loss": 0.0451,
      "step": 10
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019778270509977829,
      "loss": 0.0465,
      "step": 20
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001966740576496674,
      "loss": 0.2191,
      "step": 30
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019556541019955653,
      "loss": 0.0804,
      "step": 40
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019445676274944569,
      "loss": 0.0719,
      "step": 50
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019334811529933484,
      "loss": 0.0518,
      "step": 60
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019223946784922396,
      "loss": 0.0433,
      "step": 70
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019124168514412417,
      "loss": 0.2892,
      "step": 80
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019013303769401332,
      "loss": 0.0692,
      "step": 90
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00018902439024390244,
      "loss": 0.0765,
      "step": 100
    },
    {
      "epoch": 0.22,
      "eval_accuracy": 0.9888765294771968,
      "eval_loss": 0.05117267370223999,
      "eval_runtime": 195.5169,
      "eval_samples_per_second": 18.392,
      "eval_steps_per_second": 2.302,
      "step": 100
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001879157427937916,
      "loss": 0.1107,
      "step": 110
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00018680709534368072,
      "loss": 0.0967,
      "step": 120
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00018569844789356984,
      "loss": 0.0852,
      "step": 130
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.000184589800443459,
      "loss": 0.082,
      "step": 140
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00018348115299334812,
      "loss": 0.0172,
      "step": 150
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00018237250554323727,
      "loss": 0.0379,
      "step": 160
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0001812638580931264,
      "loss": 0.0773,
      "step": 170
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00018015521064301552,
      "loss": 0.0329,
      "step": 180
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00017904656319290467,
      "loss": 0.0688,
      "step": 190
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001779379157427938,
      "loss": 0.0533,
      "step": 200
    },
    {
      "epoch": 0.44,
      "eval_accuracy": 0.9908231368186874,
      "eval_loss": 0.037399593740701675,
      "eval_runtime": 194.4149,
      "eval_samples_per_second": 18.497,
      "eval_steps_per_second": 2.315,
      "step": 200
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017682926829268295,
      "loss": 0.0536,
      "step": 210
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00017572062084257207,
      "loss": 0.0144,
      "step": 220
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001746119733924612,
      "loss": 0.0535,
      "step": 230
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00017350332594235035,
      "loss": 0.0569,
      "step": 240
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00017239467849223947,
      "loss": 0.0408,
      "step": 250
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00017128603104212862,
      "loss": 0.0529,
      "step": 260
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00017017738359201775,
      "loss": 0.088,
      "step": 270
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00016906873614190687,
      "loss": 0.0147,
      "step": 280
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00016796008869179602,
      "loss": 0.0369,
      "step": 290
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00016685144124168515,
      "loss": 0.0442,
      "step": 300
    },
    {
      "epoch": 0.67,
      "eval_accuracy": 0.9885984427141268,
      "eval_loss": 0.03961510211229324,
      "eval_runtime": 194.5523,
      "eval_samples_per_second": 18.483,
      "eval_steps_per_second": 2.313,
      "step": 300
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0001657427937915743,
      "loss": 0.0191,
      "step": 310
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00016463414634146343,
      "loss": 0.044,
      "step": 320
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00016352549889135255,
      "loss": 0.0274,
      "step": 330
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001624168514412417,
      "loss": 0.064,
      "step": 340
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00016130820399113083,
      "loss": 0.0478,
      "step": 350
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00016019955654101998,
      "loss": 0.0532,
      "step": 360
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001590909090909091,
      "loss": 0.05,
      "step": 370
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00015798226164079823,
      "loss": 0.0197,
      "step": 380
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00015687361419068738,
      "loss": 0.0492,
      "step": 390
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001557649667405765,
      "loss": 0.0359,
      "step": 400
    },
    {
      "epoch": 0.89,
      "eval_accuracy": 0.9885984427141268,
      "eval_loss": 0.034940723329782486,
      "eval_runtime": 193.9815,
      "eval_samples_per_second": 18.538,
      "eval_steps_per_second": 2.32,
      "step": 400
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00015465631929046565,
      "loss": 0.0405,
      "step": 410
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00015354767184035478,
      "loss": 0.03,
      "step": 420
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0001524390243902439,
      "loss": 0.0272,
      "step": 430
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00015133037694013303,
      "loss": 0.0366,
      "step": 440
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00015022172949002218,
      "loss": 0.0676,
      "step": 450
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00014911308203991133,
      "loss": 0.0185,
      "step": 460
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00014800443458980045,
      "loss": 0.0053,
      "step": 470
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00014689578713968958,
      "loss": 0.0309,
      "step": 480
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001457871396895787,
      "loss": 0.0183,
      "step": 490
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00014467849223946785,
      "loss": 0.0777,
      "step": 500
    },
    {
      "epoch": 1.11,
      "eval_accuracy": 0.9869299221357063,
      "eval_loss": 0.042109716683626175,
      "eval_runtime": 193.8529,
      "eval_samples_per_second": 18.55,
      "eval_steps_per_second": 2.321,
      "step": 500
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.000143569844789357,
      "loss": 0.0159,
      "step": 510
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00014246119733924613,
      "loss": 0.0645,
      "step": 520
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00014135254988913525,
      "loss": 0.0632,
      "step": 530
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00014024390243902438,
      "loss": 0.0251,
      "step": 540
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00013913525498891353,
      "loss": 0.036,
      "step": 550
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00013802660753880268,
      "loss": 0.0389,
      "step": 560
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001369179600886918,
      "loss": 0.0353,
      "step": 570
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00013580931263858093,
      "loss": 0.0376,
      "step": 580
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00013470066518847006,
      "loss": 0.0416,
      "step": 590
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0001335920177383592,
      "loss": 0.0051,
      "step": 600
    },
    {
      "epoch": 1.33,
      "eval_accuracy": 0.9922135706340378,
      "eval_loss": 0.027106858789920807,
      "eval_runtime": 193.7443,
      "eval_samples_per_second": 18.561,
      "eval_steps_per_second": 2.323,
      "step": 600
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00013248337028824836,
      "loss": 0.0277,
      "step": 610
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00013137472283813748,
      "loss": 0.0466,
      "step": 620
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001302660753880266,
      "loss": 0.0365,
      "step": 630
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00012915742793791573,
      "loss": 0.0079,
      "step": 640
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00012804878048780488,
      "loss": 0.0291,
      "step": 650
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.000126940133037694,
      "loss": 0.0694,
      "step": 660
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00012583148558758316,
      "loss": 0.0027,
      "step": 670
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0001247228381374723,
      "loss": 0.0179,
      "step": 680
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001236141906873614,
      "loss": 0.0206,
      "step": 690
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00012250554323725056,
      "loss": 0.0112,
      "step": 700
    },
    {
      "epoch": 1.55,
      "eval_accuracy": 0.9849833147942157,
      "eval_loss": 0.051631152629852295,
      "eval_runtime": 194.3203,
      "eval_samples_per_second": 18.506,
      "eval_steps_per_second": 2.316,
      "step": 700
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00012139689578713968,
      "loss": 0.0037,
      "step": 710
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00012028824833702884,
      "loss": 0.0051,
      "step": 720
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00011917960088691797,
      "loss": 0.0014,
      "step": 730
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0001180709534368071,
      "loss": 0.0018,
      "step": 740
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00011696230598669624,
      "loss": 0.0445,
      "step": 750
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00011585365853658536,
      "loss": 0.0294,
      "step": 760
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00011474501108647451,
      "loss": 0.0247,
      "step": 770
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00011363636363636365,
      "loss": 0.0069,
      "step": 780
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011252771618625277,
      "loss": 0.0612,
      "step": 790
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00011141906873614191,
      "loss": 0.0152,
      "step": 800
    },
    {
      "epoch": 1.77,
      "eval_accuracy": 0.9824805339265851,
      "eval_loss": 0.05840621143579483,
      "eval_runtime": 194.8587,
      "eval_samples_per_second": 18.454,
      "eval_steps_per_second": 2.309,
      "step": 800
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00011031042128603104,
      "loss": 0.0137,
      "step": 810
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00010920177383592019,
      "loss": 0.0118,
      "step": 820
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00010809312638580931,
      "loss": 0.0246,
      "step": 830
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00010698447893569845,
      "loss": 0.0246,
      "step": 840
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0001058758314855876,
      "loss": 0.036,
      "step": 850
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00010476718403547671,
      "loss": 0.0248,
      "step": 860
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00010365853658536586,
      "loss": 0.019,
      "step": 870
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00010254988913525499,
      "loss": 0.0387,
      "step": 880
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00010144124168514413,
      "loss": 0.0127,
      "step": 890
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00010033259423503328,
      "loss": 0.0339,
      "step": 900
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9905450500556173,
      "eval_loss": 0.027991166338324547,
      "eval_runtime": 194.6989,
      "eval_samples_per_second": 18.47,
      "eval_steps_per_second": 2.311,
      "step": 900
    },
    {
      "epoch": 2.0,
      "step": 900,
      "total_flos": 2.2314992806549094e+18,
      "train_loss": 0.04378239723129405,
      "train_runtime": 5836.8754,
      "train_samples_per_second": 9.887,
      "train_steps_per_second": 0.309
    }
  ],
  "max_steps": 1804,
  "num_train_epochs": 4,
  "total_flos": 2.2314992806549094e+18,
  "trial_name": null,
  "trial_params": null
}