{
  "best_metric": 0.92511454202441,
  "best_model_checkpoint": "vit-msn-small-wbc-blur-detector/checkpoint-1040",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1040,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04807692307692308,
      "grad_norm": 8.808553695678711,
      "learning_rate": 4.807692307692308e-06,
      "loss": 0.5649,
      "step": 10
    },
    {
      "epoch": 0.09615384615384616,
      "grad_norm": 10.409209251403809,
      "learning_rate": 9.615384615384616e-06,
      "loss": 0.5411,
      "step": 20
    },
    {
      "epoch": 0.14423076923076922,
      "grad_norm": 11.638080596923828,
      "learning_rate": 1.4423076923076923e-05,
      "loss": 0.4686,
      "step": 30
    },
    {
      "epoch": 0.19230769230769232,
      "grad_norm": 8.448880195617676,
      "learning_rate": 1.923076923076923e-05,
      "loss": 0.467,
      "step": 40
    },
    {
      "epoch": 0.2403846153846154,
      "grad_norm": 10.3389253616333,
      "learning_rate": 2.4038461538461542e-05,
      "loss": 0.4375,
      "step": 50
    },
    {
      "epoch": 0.28846153846153844,
      "grad_norm": 14.958161354064941,
      "learning_rate": 2.8846153846153845e-05,
      "loss": 0.4374,
      "step": 60
    },
    {
      "epoch": 0.33653846153846156,
      "grad_norm": 5.999127388000488,
      "learning_rate": 3.365384615384616e-05,
      "loss": 0.4244,
      "step": 70
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 15.743879318237305,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.4003,
      "step": 80
    },
    {
      "epoch": 0.4326923076923077,
      "grad_norm": 11.51312255859375,
      "learning_rate": 4.326923076923077e-05,
      "loss": 0.4035,
      "step": 90
    },
    {
      "epoch": 0.4807692307692308,
      "grad_norm": 11.792448043823242,
      "learning_rate": 4.8076923076923084e-05,
      "loss": 0.3758,
      "step": 100
    },
    {
      "epoch": 0.5288461538461539,
      "grad_norm": 16.453128814697266,
      "learning_rate": 4.9679487179487185e-05,
      "loss": 0.4207,
      "step": 110
    },
    {
      "epoch": 0.5769230769230769,
      "grad_norm": 12.526341438293457,
      "learning_rate": 4.9145299145299147e-05,
      "loss": 0.4373,
      "step": 120
    },
    {
      "epoch": 0.625,
      "grad_norm": 14.725676536560059,
      "learning_rate": 4.8611111111111115e-05,
      "loss": 0.3985,
      "step": 130
    },
    {
      "epoch": 0.6730769230769231,
      "grad_norm": 8.387413024902344,
      "learning_rate": 4.8076923076923084e-05,
      "loss": 0.3891,
      "step": 140
    },
    {
      "epoch": 0.7211538461538461,
      "grad_norm": 9.76562213897705,
      "learning_rate": 4.7542735042735045e-05,
      "loss": 0.3595,
      "step": 150
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 6.494562149047852,
      "learning_rate": 4.700854700854701e-05,
      "loss": 0.3769,
      "step": 160
    },
    {
      "epoch": 0.8173076923076923,
      "grad_norm": 4.907495021820068,
      "learning_rate": 4.6474358974358976e-05,
      "loss": 0.3606,
      "step": 170
    },
    {
      "epoch": 0.8653846153846154,
      "grad_norm": 10.716683387756348,
      "learning_rate": 4.594017094017094e-05,
      "loss": 0.4022,
      "step": 180
    },
    {
      "epoch": 0.9134615384615384,
      "grad_norm": 4.651486873626709,
      "learning_rate": 4.5405982905982906e-05,
      "loss": 0.3586,
      "step": 190
    },
    {
      "epoch": 0.9615384615384616,
      "grad_norm": 9.122421264648438,
      "learning_rate": 4.4871794871794874e-05,
      "loss": 0.3471,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8939668116890617,
      "eval_loss": 0.29600855708122253,
      "eval_runtime": 79.6532,
      "eval_samples_per_second": 326.076,
      "eval_steps_per_second": 5.097,
      "step": 208
    },
    {
      "epoch": 1.0096153846153846,
      "grad_norm": 4.985302448272705,
      "learning_rate": 4.4337606837606836e-05,
      "loss": 0.3561,
      "step": 210
    },
    {
      "epoch": 1.0576923076923077,
      "grad_norm": 8.867368698120117,
      "learning_rate": 4.3803418803418805e-05,
      "loss": 0.3629,
      "step": 220
    },
    {
      "epoch": 1.1057692307692308,
      "grad_norm": 5.275181770324707,
      "learning_rate": 4.326923076923077e-05,
      "loss": 0.3352,
      "step": 230
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 7.328935146331787,
      "learning_rate": 4.2735042735042735e-05,
      "loss": 0.3593,
      "step": 240
    },
    {
      "epoch": 1.2019230769230769,
      "grad_norm": 7.934610366821289,
      "learning_rate": 4.2200854700854704e-05,
      "loss": 0.3651,
      "step": 250
    },
    {
      "epoch": 1.25,
      "grad_norm": 4.802842140197754,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.3517,
      "step": 260
    },
    {
      "epoch": 1.2980769230769231,
      "grad_norm": 5.414445400238037,
      "learning_rate": 4.1132478632478634e-05,
      "loss": 0.3527,
      "step": 270
    },
    {
      "epoch": 1.3461538461538463,
      "grad_norm": 6.663522243499756,
      "learning_rate": 4.05982905982906e-05,
      "loss": 0.3397,
      "step": 280
    },
    {
      "epoch": 1.3942307692307692,
      "grad_norm": 4.669881820678711,
      "learning_rate": 4.006410256410257e-05,
      "loss": 0.3378,
      "step": 290
    },
    {
      "epoch": 1.4423076923076923,
      "grad_norm": 5.708807945251465,
      "learning_rate": 3.952991452991453e-05,
      "loss": 0.3119,
      "step": 300
    },
    {
      "epoch": 1.4903846153846154,
      "grad_norm": 5.911713600158691,
      "learning_rate": 3.89957264957265e-05,
      "loss": 0.329,
      "step": 310
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 4.569310188293457,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.3348,
      "step": 320
    },
    {
      "epoch": 1.5865384615384617,
      "grad_norm": 6.366770267486572,
      "learning_rate": 3.7927350427350425e-05,
      "loss": 0.3268,
      "step": 330
    },
    {
      "epoch": 1.6346153846153846,
      "grad_norm": 4.477969646453857,
      "learning_rate": 3.739316239316239e-05,
      "loss": 0.3327,
      "step": 340
    },
    {
      "epoch": 1.6826923076923077,
      "grad_norm": 5.906105995178223,
      "learning_rate": 3.685897435897436e-05,
      "loss": 0.316,
      "step": 350
    },
    {
      "epoch": 1.7307692307692308,
      "grad_norm": 4.78982400894165,
      "learning_rate": 3.6324786324786323e-05,
      "loss": 0.339,
      "step": 360
    },
    {
      "epoch": 1.7788461538461537,
      "grad_norm": 4.8963117599487305,
      "learning_rate": 3.579059829059829e-05,
      "loss": 0.3247,
      "step": 370
    },
    {
      "epoch": 1.8269230769230769,
      "grad_norm": 4.7771501541137695,
      "learning_rate": 3.525641025641026e-05,
      "loss": 0.3168,
      "step": 380
    },
    {
      "epoch": 1.875,
      "grad_norm": 5.485278606414795,
      "learning_rate": 3.472222222222222e-05,
      "loss": 0.3107,
      "step": 390
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 5.954066753387451,
      "learning_rate": 3.418803418803419e-05,
      "loss": 0.3196,
      "step": 400
    },
    {
      "epoch": 1.9711538461538463,
      "grad_norm": 4.826352119445801,
      "learning_rate": 3.365384615384616e-05,
      "loss": 0.3113,
      "step": 410
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9087898972009394,
      "eval_loss": 0.2550640106201172,
      "eval_runtime": 77.7682,
      "eval_samples_per_second": 333.98,
      "eval_steps_per_second": 5.221,
      "step": 416
    },
    {
      "epoch": 2.019230769230769,
      "grad_norm": 4.100191116333008,
      "learning_rate": 3.311965811965812e-05,
      "loss": 0.3039,
      "step": 420
    },
    {
      "epoch": 2.0673076923076925,
      "grad_norm": 5.670529842376709,
      "learning_rate": 3.258547008547009e-05,
      "loss": 0.3228,
      "step": 430
    },
    {
      "epoch": 2.1153846153846154,
      "grad_norm": 4.010565757751465,
      "learning_rate": 3.205128205128206e-05,
      "loss": 0.3096,
      "step": 440
    },
    {
      "epoch": 2.1634615384615383,
      "grad_norm": 5.74422550201416,
      "learning_rate": 3.151709401709402e-05,
      "loss": 0.3119,
      "step": 450
    },
    {
      "epoch": 2.2115384615384617,
      "grad_norm": 3.280287742614746,
      "learning_rate": 3.098290598290599e-05,
      "loss": 0.3305,
      "step": 460
    },
    {
      "epoch": 2.2596153846153846,
      "grad_norm": 4.927590847015381,
      "learning_rate": 3.0448717948717947e-05,
      "loss": 0.3034,
      "step": 470
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 3.297640562057495,
      "learning_rate": 2.9914529914529915e-05,
      "loss": 0.3081,
      "step": 480
    },
    {
      "epoch": 2.355769230769231,
      "grad_norm": 4.635345458984375,
      "learning_rate": 2.9380341880341884e-05,
      "loss": 0.3096,
      "step": 490
    },
    {
      "epoch": 2.4038461538461537,
      "grad_norm": 4.222609519958496,
      "learning_rate": 2.8846153846153845e-05,
      "loss": 0.3242,
      "step": 500
    },
    {
      "epoch": 2.451923076923077,
      "grad_norm": 5.433651924133301,
      "learning_rate": 2.8311965811965814e-05,
      "loss": 0.2856,
      "step": 510
    },
    {
      "epoch": 2.5,
      "grad_norm": 3.512269973754883,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.3165,
      "step": 520
    },
    {
      "epoch": 2.5480769230769234,
      "grad_norm": 4.383204936981201,
      "learning_rate": 2.724358974358974e-05,
      "loss": 0.3309,
      "step": 530
    },
    {
      "epoch": 2.5961538461538463,
      "grad_norm": 3.9306111335754395,
      "learning_rate": 2.670940170940171e-05,
      "loss": 0.3012,
      "step": 540
    },
    {
      "epoch": 2.644230769230769,
      "grad_norm": 4.230567455291748,
      "learning_rate": 2.6175213675213678e-05,
      "loss": 0.2931,
      "step": 550
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 4.904375076293945,
      "learning_rate": 2.564102564102564e-05,
      "loss": 0.3042,
      "step": 560
    },
    {
      "epoch": 2.7403846153846154,
      "grad_norm": 4.499465465545654,
      "learning_rate": 2.5106837606837608e-05,
      "loss": 0.2904,
      "step": 570
    },
    {
      "epoch": 2.7884615384615383,
      "grad_norm": 4.0428242683410645,
      "learning_rate": 2.4572649572649573e-05,
      "loss": 0.3019,
      "step": 580
    },
    {
      "epoch": 2.8365384615384617,
      "grad_norm": 5.506063938140869,
      "learning_rate": 2.4038461538461542e-05,
      "loss": 0.2976,
      "step": 590
    },
    {
      "epoch": 2.8846153846153846,
      "grad_norm": 3.252035617828369,
      "learning_rate": 2.3504273504273504e-05,
      "loss": 0.2874,
      "step": 600
    },
    {
      "epoch": 2.9326923076923075,
      "grad_norm": 3.8732314109802246,
      "learning_rate": 2.297008547008547e-05,
      "loss": 0.3022,
      "step": 610
    },
    {
      "epoch": 2.980769230769231,
      "grad_norm": 4.38186502456665,
      "learning_rate": 2.2435897435897437e-05,
      "loss": 0.3104,
      "step": 620
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9212258884225927,
      "eval_loss": 0.21063366532325745,
      "eval_runtime": 77.9718,
      "eval_samples_per_second": 333.107,
      "eval_steps_per_second": 5.207,
      "step": 624
    },
    {
      "epoch": 3.0288461538461537,
      "grad_norm": 3.352865695953369,
      "learning_rate": 2.1901709401709402e-05,
      "loss": 0.2898,
      "step": 630
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 4.9008402824401855,
      "learning_rate": 2.1367521367521368e-05,
      "loss": 0.326,
      "step": 640
    },
    {
      "epoch": 3.125,
      "grad_norm": 5.130583763122559,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 0.293,
      "step": 650
    },
    {
      "epoch": 3.173076923076923,
      "grad_norm": 4.387807369232178,
      "learning_rate": 2.02991452991453e-05,
      "loss": 0.2777,
      "step": 660
    },
    {
      "epoch": 3.2211538461538463,
      "grad_norm": 5.279751300811768,
      "learning_rate": 1.9764957264957266e-05,
      "loss": 0.3026,
      "step": 670
    },
    {
      "epoch": 3.269230769230769,
      "grad_norm": 3.5947065353393555,
      "learning_rate": 1.923076923076923e-05,
      "loss": 0.2881,
      "step": 680
    },
    {
      "epoch": 3.3173076923076925,
      "grad_norm": 3.5021965503692627,
      "learning_rate": 1.8696581196581197e-05,
      "loss": 0.277,
      "step": 690
    },
    {
      "epoch": 3.3653846153846154,
      "grad_norm": 5.785918712615967,
      "learning_rate": 1.8162393162393162e-05,
      "loss": 0.3025,
      "step": 700
    },
    {
      "epoch": 3.4134615384615383,
      "grad_norm": 3.405780553817749,
      "learning_rate": 1.762820512820513e-05,
      "loss": 0.2825,
      "step": 710
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 3.8835082054138184,
      "learning_rate": 1.7094017094017095e-05,
      "loss": 0.2754,
      "step": 720
    },
    {
      "epoch": 3.5096153846153846,
      "grad_norm": 3.6545193195343018,
      "learning_rate": 1.655982905982906e-05,
      "loss": 0.2736,
      "step": 730
    },
    {
      "epoch": 3.5576923076923075,
      "grad_norm": 3.788748025894165,
      "learning_rate": 1.602564102564103e-05,
      "loss": 0.3084,
      "step": 740
    },
    {
      "epoch": 3.605769230769231,
      "grad_norm": 3.0023396015167236,
      "learning_rate": 1.5491452991452994e-05,
      "loss": 0.2797,
      "step": 750
    },
    {
      "epoch": 3.6538461538461537,
      "grad_norm": 3.6981606483459473,
      "learning_rate": 1.4957264957264958e-05,
      "loss": 0.278,
      "step": 760
    },
    {
      "epoch": 3.7019230769230766,
      "grad_norm": 4.383445739746094,
      "learning_rate": 1.4423076923076923e-05,
      "loss": 0.2602,
      "step": 770
    },
    {
      "epoch": 3.75,
      "grad_norm": 4.581905364990234,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.2877,
      "step": 780
    },
    {
      "epoch": 3.7980769230769234,
      "grad_norm": 4.475015640258789,
      "learning_rate": 1.3354700854700855e-05,
      "loss": 0.2733,
      "step": 790
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 4.552759170532227,
      "learning_rate": 1.282051282051282e-05,
      "loss": 0.2927,
      "step": 800
    },
    {
      "epoch": 3.894230769230769,
      "grad_norm": 4.798213958740234,
      "learning_rate": 1.2286324786324787e-05,
      "loss": 0.2756,
      "step": 810
    },
    {
      "epoch": 3.9423076923076925,
      "grad_norm": 3.405367136001587,
      "learning_rate": 1.1752136752136752e-05,
      "loss": 0.2893,
      "step": 820
    },
    {
      "epoch": 3.9903846153846154,
      "grad_norm": 5.285249710083008,
      "learning_rate": 1.1217948717948719e-05,
      "loss": 0.2855,
      "step": 830
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9221114234012243,
      "eval_loss": 0.21014131605625153,
      "eval_runtime": 78.3645,
      "eval_samples_per_second": 331.438,
      "eval_steps_per_second": 5.181,
      "step": 832
    },
    {
      "epoch": 4.038461538461538,
      "grad_norm": 4.0211920738220215,
      "learning_rate": 1.0683760683760684e-05,
      "loss": 0.2639,
      "step": 840
    },
    {
      "epoch": 4.086538461538462,
      "grad_norm": 3.0746471881866455,
      "learning_rate": 1.014957264957265e-05,
      "loss": 0.2866,
      "step": 850
    },
    {
      "epoch": 4.134615384615385,
      "grad_norm": 3.945941686630249,
      "learning_rate": 9.615384615384616e-06,
      "loss": 0.2624,
      "step": 860
    },
    {
      "epoch": 4.1826923076923075,
      "grad_norm": 3.996899366378784,
      "learning_rate": 9.081196581196581e-06,
      "loss": 0.2795,
      "step": 870
    },
    {
      "epoch": 4.230769230769231,
      "grad_norm": 3.078308582305908,
      "learning_rate": 8.547008547008548e-06,
      "loss": 0.2662,
      "step": 880
    },
    {
      "epoch": 4.278846153846154,
      "grad_norm": 4.386013507843018,
      "learning_rate": 8.012820512820515e-06,
      "loss": 0.2607,
      "step": 890
    },
    {
      "epoch": 4.326923076923077,
      "grad_norm": 4.05147647857666,
      "learning_rate": 7.478632478632479e-06,
      "loss": 0.2837,
      "step": 900
    },
    {
      "epoch": 4.375,
      "grad_norm": 3.7638943195343018,
      "learning_rate": 6.944444444444445e-06,
      "loss": 0.2588,
      "step": 910
    },
    {
      "epoch": 4.423076923076923,
      "grad_norm": 3.4808619022369385,
      "learning_rate": 6.41025641025641e-06,
      "loss": 0.2709,
      "step": 920
    },
    {
      "epoch": 4.471153846153846,
      "grad_norm": 4.455379486083984,
      "learning_rate": 5.876068376068376e-06,
      "loss": 0.2743,
      "step": 930
    },
    {
      "epoch": 4.519230769230769,
      "grad_norm": 3.4500174522399902,
      "learning_rate": 5.341880341880342e-06,
      "loss": 0.2501,
      "step": 940
    },
    {
      "epoch": 4.5673076923076925,
      "grad_norm": 3.780879020690918,
      "learning_rate": 4.807692307692308e-06,
      "loss": 0.2666,
      "step": 950
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 3.8015003204345703,
      "learning_rate": 4.273504273504274e-06,
      "loss": 0.2516,
      "step": 960
    },
    {
      "epoch": 4.663461538461538,
      "grad_norm": 3.866867780685425,
      "learning_rate": 3.7393162393162394e-06,
      "loss": 0.2673,
      "step": 970
    },
    {
      "epoch": 4.711538461538462,
      "grad_norm": 3.153127431869507,
      "learning_rate": 3.205128205128205e-06,
      "loss": 0.2472,
      "step": 980
    },
    {
      "epoch": 4.759615384615385,
      "grad_norm": 3.0481035709381104,
      "learning_rate": 2.670940170940171e-06,
      "loss": 0.2669,
      "step": 990
    },
    {
      "epoch": 4.8076923076923075,
      "grad_norm": 3.5989151000976562,
      "learning_rate": 2.136752136752137e-06,
      "loss": 0.2748,
      "step": 1000
    },
    {
      "epoch": 4.855769230769231,
      "grad_norm": 4.3564839363098145,
      "learning_rate": 1.6025641025641025e-06,
      "loss": 0.2651,
      "step": 1010
    },
    {
      "epoch": 4.903846153846154,
      "grad_norm": 3.6109776496887207,
      "learning_rate": 1.0683760683760685e-06,
      "loss": 0.2631,
      "step": 1020
    },
    {
      "epoch": 4.951923076923077,
      "grad_norm": 3.1704578399658203,
      "learning_rate": 5.341880341880342e-07,
      "loss": 0.2665,
      "step": 1030
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.165510654449463,
      "learning_rate": 0.0,
      "loss": 0.2497,
      "step": 1040
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.92511454202441,
      "eval_loss": 0.20453155040740967,
      "eval_runtime": 78.1441,
      "eval_samples_per_second": 332.373,
      "eval_steps_per_second": 5.196,
      "step": 1040
    },
    {
      "epoch": 5.0,
      "step": 1040,
      "total_flos": 5.20812700794667e+18,
      "train_loss": 0.32158288405491753,
      "train_runtime": 2381.0335,
      "train_samples_per_second": 111.775,
      "train_steps_per_second": 0.437
    }
  ],
  "logging_steps": 10,
  "max_steps": 1040,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.20812700794667e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}