{
  "best_metric": 0.9808994233422476,
  "best_model_checkpoint": "./outputs/dinov2-base-imagenet1k-1-layer-boulderspot-vN/checkpoint-1015",
  "epoch": 4.993849938499385,
  "eval_steps": 500,
  "global_step": 1015,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 21.444149017333984,
      "learning_rate": 3.92156862745098e-06,
      "loss": 0.3077,
      "step": 10
    },
    {
      "epoch": 0.1,
      "grad_norm": 20.83109474182129,
      "learning_rate": 7.84313725490196e-06,
      "loss": 0.0953,
      "step": 20
    },
    {
      "epoch": 0.15,
      "grad_norm": 8.993614196777344,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 0.1968,
      "step": 30
    },
    {
      "epoch": 0.2,
      "grad_norm": 32.97685241699219,
      "learning_rate": 1.568627450980392e-05,
      "loss": 0.1265,
      "step": 40
    },
    {
      "epoch": 0.25,
      "grad_norm": 10.921149253845215,
      "learning_rate": 1.9607843137254903e-05,
      "loss": 0.166,
      "step": 50
    },
    {
      "epoch": 0.3,
      "grad_norm": 30.291000366210938,
      "learning_rate": 1.9995698998770955e-05,
      "loss": 0.1484,
      "step": 60
    },
    {
      "epoch": 0.34,
      "grad_norm": 7.144667625427246,
      "learning_rate": 1.998083609002402e-05,
      "loss": 0.1529,
      "step": 70
    },
    {
      "epoch": 0.39,
      "grad_norm": 72.38031005859375,
      "learning_rate": 1.995537395500004e-05,
      "loss": 0.1588,
      "step": 80
    },
    {
      "epoch": 0.44,
      "grad_norm": 9.280795097351074,
      "learning_rate": 1.9919339633410737e-05,
      "loss": 0.1575,
      "step": 90
    },
    {
      "epoch": 0.49,
      "grad_norm": 8.867383003234863,
      "learning_rate": 1.9872771392183334e-05,
      "loss": 0.1345,
      "step": 100
    },
    {
      "epoch": 0.54,
      "grad_norm": 11.048301696777344,
      "learning_rate": 1.981571868482269e-05,
      "loss": 0.1832,
      "step": 110
    },
    {
      "epoch": 0.59,
      "grad_norm": 18.908288955688477,
      "learning_rate": 1.974824209889377e-05,
      "loss": 0.1392,
      "step": 120
    },
    {
      "epoch": 0.64,
      "grad_norm": 6.0579047203063965,
      "learning_rate": 1.9670413291680223e-05,
      "loss": 0.1179,
      "step": 130
    },
    {
      "epoch": 0.69,
      "grad_norm": 6.6037774085998535,
      "learning_rate": 1.9582314914087344e-05,
      "loss": 0.1258,
      "step": 140
    },
    {
      "epoch": 0.74,
      "grad_norm": 17.16813087463379,
      "learning_rate": 1.9484040522870333e-05,
      "loss": 0.1914,
      "step": 150
    },
    {
      "epoch": 0.79,
      "grad_norm": 15.085983276367188,
      "learning_rate": 1.9375694481280965e-05,
      "loss": 0.1747,
      "step": 160
    },
    {
      "epoch": 0.84,
      "grad_norm": 6.54490852355957,
      "learning_rate": 1.9257391848238212e-05,
      "loss": 0.0749,
      "step": 170
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.9332025051116943,
      "learning_rate": 1.9129258256140556e-05,
      "loss": 0.0706,
      "step": 180
    },
    {
      "epoch": 0.93,
      "grad_norm": 26.89777946472168,
      "learning_rate": 1.8991429777449674e-05,
      "loss": 0.0852,
      "step": 190
    },
    {
      "epoch": 0.98,
      "grad_norm": 4.798658847808838,
      "learning_rate": 1.884405278018722e-05,
      "loss": 0.1596,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9766081871345029,
      "eval_f1": 0.9758655635300373,
      "eval_loss": 0.0732586681842804,
      "eval_matthews_correlation": 0.8078903073020254,
      "eval_precision": 0.9756885037132272,
      "eval_recall": 0.9766081871345029,
      "eval_runtime": 3.9698,
      "eval_samples_per_second": 172.303,
      "eval_steps_per_second": 10.832,
      "step": 203
    },
    {
      "epoch": 1.03,
      "grad_norm": 13.787192344665527,
      "learning_rate": 1.8687283772498205e-05,
      "loss": 0.1367,
      "step": 210
    },
    {
      "epoch": 1.08,
      "grad_norm": 5.066114902496338,
      "learning_rate": 1.852128923644593e-05,
      "loss": 0.1308,
      "step": 220
    },
    {
      "epoch": 1.13,
      "grad_norm": 2.300729990005493,
      "learning_rate": 1.8346245451215068e-05,
      "loss": 0.1011,
      "step": 230
    },
    {
      "epoch": 1.18,
      "grad_norm": 6.617341995239258,
      "learning_rate": 1.8162338305910636e-05,
      "loss": 0.0879,
      "step": 240
    },
    {
      "epoch": 1.23,
      "grad_norm": 5.020946025848389,
      "learning_rate": 1.79697631021516e-05,
      "loss": 0.0969,
      "step": 250
    },
    {
      "epoch": 1.28,
      "grad_norm": 7.580730438232422,
      "learning_rate": 1.776872434666882e-05,
      "loss": 0.0942,
      "step": 260
    },
    {
      "epoch": 1.33,
      "grad_norm": 1.9427905082702637,
      "learning_rate": 1.7559435534127534e-05,
      "loss": 0.0745,
      "step": 270
    },
    {
      "epoch": 1.38,
      "grad_norm": 10.397661209106445,
      "learning_rate": 1.7342118920405035e-05,
      "loss": 0.1028,
      "step": 280
    },
    {
      "epoch": 1.43,
      "grad_norm": 8.107540130615234,
      "learning_rate": 1.7117005286564344e-05,
      "loss": 0.0941,
      "step": 290
    },
    {
      "epoch": 1.48,
      "grad_norm": 13.503528594970703,
      "learning_rate": 1.688433369377444e-05,
      "loss": 0.1162,
      "step": 300
    },
    {
      "epoch": 1.53,
      "grad_norm": 15.07189655303955,
      "learning_rate": 1.6644351229437416e-05,
      "loss": 0.1301,
      "step": 310
    },
    {
      "epoch": 1.57,
      "grad_norm": 3.9105443954467773,
      "learning_rate": 1.63973127447921e-05,
      "loss": 0.1441,
      "step": 320
    },
    {
      "epoch": 1.62,
      "grad_norm": 7.494752883911133,
      "learning_rate": 1.6143480584272794e-05,
      "loss": 0.1002,
      "step": 330
    },
    {
      "epoch": 1.67,
      "grad_norm": 2.2176589965820312,
      "learning_rate": 1.5883124306910563e-05,
      "loss": 0.0731,
      "step": 340
    },
    {
      "epoch": 1.72,
      "grad_norm": 5.566643714904785,
      "learning_rate": 1.5616520400072963e-05,
      "loss": 0.093,
      "step": 350
    },
    {
      "epoch": 1.77,
      "grad_norm": 4.072403907775879,
      "learning_rate": 1.5343951985846096e-05,
      "loss": 0.0899,
      "step": 360
    },
    {
      "epoch": 1.82,
      "grad_norm": 4.870898246765137,
      "learning_rate": 1.5065708520370943e-05,
      "loss": 0.0781,
      "step": 370
    },
    {
      "epoch": 1.87,
      "grad_norm": 21.179044723510742,
      "learning_rate": 1.4782085486453155e-05,
      "loss": 0.0807,
      "step": 380
    },
    {
      "epoch": 1.92,
      "grad_norm": 4.826471328735352,
      "learning_rate": 1.4493384079772815e-05,
      "loss": 0.0852,
      "step": 390
    },
    {
      "epoch": 1.97,
      "grad_norm": 3.290867328643799,
      "learning_rate": 1.4199910889027335e-05,
      "loss": 0.0635,
      "step": 400
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9473684210526315,
      "eval_f1": 0.9522155218554862,
      "eval_loss": 0.12761278450489044,
      "eval_matthews_correlation": 0.6844503635019795,
      "eval_precision": 0.9618507818612543,
      "eval_recall": 0.9473684210526315,
      "eval_runtime": 4.0112,
      "eval_samples_per_second": 170.521,
      "eval_steps_per_second": 10.72,
      "step": 406
    },
    {
      "epoch": 2.02,
      "grad_norm": 16.522815704345703,
      "learning_rate": 1.390197757034721e-05,
      "loss": 0.0853,
      "step": 410
    },
    {
      "epoch": 2.07,
      "grad_norm": 5.749844074249268,
      "learning_rate": 1.3599900516330382e-05,
      "loss": 0.0685,
      "step": 420
    },
    {
      "epoch": 2.12,
      "grad_norm": 4.761476993560791,
      "learning_rate": 1.3294000520046666e-05,
      "loss": 0.086,
      "step": 430
    },
    {
      "epoch": 2.16,
      "grad_norm": 5.712699890136719,
      "learning_rate": 1.2984602434369058e-05,
      "loss": 0.0927,
      "step": 440
    },
    {
      "epoch": 2.21,
      "grad_norm": 7.019559860229492,
      "learning_rate": 1.2672034826993716e-05,
      "loss": 0.0771,
      "step": 450
    },
    {
      "epoch": 2.26,
      "grad_norm": 5.033693313598633,
      "learning_rate": 1.235662963151493e-05,
      "loss": 0.065,
      "step": 460
    },
    {
      "epoch": 2.31,
      "grad_norm": 7.356439590454102,
      "learning_rate": 1.2038721794925689e-05,
      "loss": 0.0782,
      "step": 470
    },
    {
      "epoch": 2.36,
      "grad_norm": 13.046978950500488,
      "learning_rate": 1.1718648921918112e-05,
      "loss": 0.1074,
      "step": 480
    },
    {
      "epoch": 2.41,
      "grad_norm": 2.489901542663574,
      "learning_rate": 1.1396750916361526e-05,
      "loss": 0.0891,
      "step": 490
    },
    {
      "epoch": 2.46,
      "grad_norm": 8.636452674865723,
      "learning_rate": 1.1073369620338928e-05,
      "loss": 0.0922,
      "step": 500
    },
    {
      "epoch": 2.51,
      "grad_norm": 3.626593589782715,
      "learning_rate": 1.074884845112512e-05,
      "loss": 0.0832,
      "step": 510
    },
    {
      "epoch": 2.56,
      "grad_norm": 2.680762529373169,
      "learning_rate": 1.0423532036492077e-05,
      "loss": 0.0659,
      "step": 520
    },
    {
      "epoch": 2.61,
      "grad_norm": 9.335197448730469,
      "learning_rate": 1.0097765848728825e-05,
      "loss": 0.0718,
      "step": 530
    },
    {
      "epoch": 2.66,
      "grad_norm": 2.929506301879883,
      "learning_rate": 9.771895837764438e-06,
      "loss": 0.0975,
      "step": 540
    },
    {
      "epoch": 2.71,
      "grad_norm": 1.251734733581543,
      "learning_rate": 9.446268063783853e-06,
      "loss": 0.0343,
      "step": 550
    },
    {
      "epoch": 2.76,
      "grad_norm": 11.404159545898438,
      "learning_rate": 9.121228329726563e-06,
      "loss": 0.0488,
      "step": 560
    },
    {
      "epoch": 2.8,
      "grad_norm": 3.3042876720428467,
      "learning_rate": 8.797121814058502e-06,
      "loss": 0.0641,
      "step": 570
    },
    {
      "epoch": 2.85,
      "grad_norm": 10.385183334350586,
      "learning_rate": 8.474292704207095e-06,
      "loss": 0.0951,
      "step": 580
    },
    {
      "epoch": 2.9,
      "grad_norm": 7.184572219848633,
      "learning_rate": 8.153083831048772e-06,
      "loss": 0.0591,
      "step": 590
    },
    {
      "epoch": 2.95,
      "grad_norm": 4.226497650146484,
      "learning_rate": 7.833836304837022e-06,
      "loss": 0.1031,
      "step": 600
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9751461988304093,
      "eval_f1": 0.9755012041745224,
      "eval_loss": 0.06017656996846199,
      "eval_matthews_correlation": 0.8118305972172924,
      "eval_precision": 0.9759749663327615,
      "eval_recall": 0.9751461988304093,
      "eval_runtime": 4.1933,
      "eval_samples_per_second": 163.119,
      "eval_steps_per_second": 10.255,
      "step": 609
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.557692527770996,
      "learning_rate": 7.516889152957744e-06,
      "loss": 0.0533,
      "step": 610
    },
    {
      "epoch": 3.05,
      "grad_norm": 3.751300096511841,
      "learning_rate": 7.202578959896491e-06,
      "loss": 0.0782,
      "step": 620
    },
    {
      "epoch": 3.1,
      "grad_norm": 4.593023777008057,
      "learning_rate": 6.891239509799932e-06,
      "loss": 0.0627,
      "step": 630
    },
    {
      "epoch": 3.15,
      "grad_norm": 4.492953777313232,
      "learning_rate": 6.583201432011217e-06,
      "loss": 0.0564,
      "step": 640
    },
    {
      "epoch": 3.2,
      "grad_norm": 3.022425889968872,
      "learning_rate": 6.278791849955583e-06,
      "loss": 0.0719,
      "step": 650
    },
    {
      "epoch": 3.25,
      "grad_norm": 4.909082889556885,
      "learning_rate": 5.978334033749076e-06,
      "loss": 0.0531,
      "step": 660
    },
    {
      "epoch": 3.3,
      "grad_norm": 2.2064766883850098,
      "learning_rate": 5.682147056899361e-06,
      "loss": 0.0628,
      "step": 670
    },
    {
      "epoch": 3.35,
      "grad_norm": 5.478167533874512,
      "learning_rate": 5.390545457463134e-06,
      "loss": 0.0705,
      "step": 680
    },
    {
      "epoch": 3.39,
      "grad_norm": 5.376189708709717,
      "learning_rate": 5.103838904019993e-06,
      "loss": 0.0888,
      "step": 690
    },
    {
      "epoch": 3.44,
      "grad_norm": 3.5109596252441406,
      "learning_rate": 4.822331866817478e-06,
      "loss": 0.0577,
      "step": 700
    },
    {
      "epoch": 3.49,
      "grad_norm": 0.5529822707176208,
      "learning_rate": 4.546323294436556e-06,
      "loss": 0.0421,
      "step": 710
    },
    {
      "epoch": 3.54,
      "grad_norm": 10.782859802246094,
      "learning_rate": 4.276106296320828e-06,
      "loss": 0.0579,
      "step": 720
    },
    {
      "epoch": 3.59,
      "grad_norm": 0.36797773838043213,
      "learning_rate": 4.0119678315067025e-06,
      "loss": 0.0381,
      "step": 730
    },
    {
      "epoch": 3.64,
      "grad_norm": 3.724026679992676,
      "learning_rate": 3.754188403885013e-06,
      "loss": 0.057,
      "step": 740
    },
    {
      "epoch": 3.69,
      "grad_norm": 6.234977722167969,
      "learning_rate": 3.5030417643177416e-06,
      "loss": 0.0556,
      "step": 750
    },
    {
      "epoch": 3.74,
      "grad_norm": 6.538670063018799,
      "learning_rate": 3.258794619926159e-06,
      "loss": 0.0624,
      "step": 760
    },
    {
      "epoch": 3.79,
      "grad_norm": 1.8592917919158936,
      "learning_rate": 3.021706350859147e-06,
      "loss": 0.0544,
      "step": 770
    },
    {
      "epoch": 3.84,
      "grad_norm": 1.5543180704116821,
      "learning_rate": 2.792028734842418e-06,
      "loss": 0.0321,
      "step": 780
    },
    {
      "epoch": 3.89,
      "grad_norm": 6.113970756530762,
      "learning_rate": 2.5700056798012164e-06,
      "loss": 0.052,
      "step": 790
    },
    {
      "epoch": 3.94,
      "grad_norm": 6.493689060211182,
      "learning_rate": 2.3558729648404065e-06,
      "loss": 0.046,
      "step": 800
    },
    {
      "epoch": 3.99,
      "grad_norm": 3.7167036533355713,
      "learning_rate": 2.1498579898570228e-06,
      "loss": 0.0587,
      "step": 810
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9736842105263158,
      "eval_f1": 0.9734187929958467,
      "eval_loss": 0.05119941756129265,
      "eval_matthews_correlation": 0.7904630921052955,
      "eval_precision": 0.9732101510950039,
      "eval_recall": 0.9736842105263158,
      "eval_runtime": 4.2261,
      "eval_samples_per_second": 161.853,
      "eval_steps_per_second": 10.175,
      "step": 813
    },
    {
      "epoch": 4.03,
      "grad_norm": 6.378363132476807,
      "learning_rate": 1.952179534051183e-06,
      "loss": 0.0389,
      "step": 820
    },
    {
      "epoch": 4.08,
      "grad_norm": 5.403367042541504,
      "learning_rate": 1.763047523591831e-06,
      "loss": 0.0727,
      "step": 830
    },
    {
      "epoch": 4.13,
      "grad_norm": 5.556785583496094,
      "learning_rate": 1.5826628086839968e-06,
      "loss": 0.0365,
      "step": 840
    },
    {
      "epoch": 4.18,
      "grad_norm": 10.821802139282227,
      "learning_rate": 1.41121695027438e-06,
      "loss": 0.0419,
      "step": 850
    },
    {
      "epoch": 4.23,
      "grad_norm": 2.722320795059204,
      "learning_rate": 1.2488920166217034e-06,
      "loss": 0.0404,
      "step": 860
    },
    {
      "epoch": 4.28,
      "grad_norm": 11.169462203979492,
      "learning_rate": 1.095860389947928e-06,
      "loss": 0.0522,
      "step": 870
    },
    {
      "epoch": 4.33,
      "grad_norm": 4.065892696380615,
      "learning_rate": 9.522845833756001e-07,
      "loss": 0.0497,
      "step": 880
    },
    {
      "epoch": 4.38,
      "grad_norm": 11.717745780944824,
      "learning_rate": 8.183170683457986e-07,
      "loss": 0.0543,
      "step": 890
    },
    {
      "epoch": 4.43,
      "grad_norm": 3.715181589126587,
      "learning_rate": 6.941001126998892e-07,
      "loss": 0.065,
      "step": 900
    },
    {
      "epoch": 4.48,
      "grad_norm": 8.879691123962402,
      "learning_rate": 5.797656295970955e-07,
      "loss": 0.0546,
      "step": 910
    },
    {
      "epoch": 4.53,
      "grad_norm": 6.542267799377441,
      "learning_rate": 4.754350374283001e-07,
      "loss": 0.0491,
      "step": 920
    },
    {
      "epoch": 4.58,
      "grad_norm": 5.4677557945251465,
      "learning_rate": 3.8121913087483033e-07,
      "loss": 0.0434,
      "step": 930
    },
    {
      "epoch": 4.62,
      "grad_norm": 5.965972900390625,
      "learning_rate": 2.972179632491989e-07,
      "loss": 0.0594,
      "step": 940
    },
    {
      "epoch": 4.67,
      "grad_norm": 5.097690105438232,
      "learning_rate": 2.23520740242712e-07,
      "loss": 0.0632,
      "step": 950
    },
    {
      "epoch": 4.72,
      "grad_norm": 5.078763961791992,
      "learning_rate": 1.602057251927891e-07,
      "loss": 0.0448,
      "step": 960
    },
    {
      "epoch": 4.77,
      "grad_norm": 4.476650714874268,
      "learning_rate": 1.0734015597060222e-07,
      "loss": 0.0481,
      "step": 970
    },
    {
      "epoch": 4.82,
      "grad_norm": 3.210946798324585,
      "learning_rate": 6.498017357731035e-08,
      "loss": 0.0604,
      "step": 980
    },
    {
      "epoch": 4.87,
      "grad_norm": 6.374156951904297,
      "learning_rate": 3.317076252467133e-08,
      "loss": 0.0406,
      "step": 990
    },
    {
      "epoch": 4.92,
      "grad_norm": 2.580352306365967,
      "learning_rate": 1.1945703063402925e-08,
      "loss": 0.0343,
      "step": 1000
    },
    {
      "epoch": 4.97,
      "grad_norm": 1.094759464263916,
      "learning_rate": 1.327535309979533e-09,
      "loss": 0.038,
      "step": 1010
    },
    {
      "epoch": 4.99,
      "eval_accuracy": 0.9809941520467836,
      "eval_f1": 0.9808994233422476,
      "eval_loss": 0.051854074001312256,
      "eval_matthews_correlation": 0.8500768147494288,
      "eval_precision": 0.9808194985441887,
      "eval_recall": 0.9809941520467836,
      "eval_runtime": 4.1997,
      "eval_samples_per_second": 162.87,
      "eval_steps_per_second": 10.239,
      "step": 1015
    },
    {
      "epoch": 4.99,
      "step": 1015,
      "total_flos": 6.629580853384053e+18,
      "train_loss": 0.08605182834446724,
      "train_runtime": 583.274,
      "train_samples_per_second": 111.397,
      "train_steps_per_second": 1.74
    }
  ],
  "logging_steps": 10,
  "max_steps": 1015,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 6.629580853384053e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}