{
  "best_metric": 0.07409419864416122,
  "best_model_checkpoint": "./vit-base-hand-classifier-9\\checkpoint-1200",
  "epoch": 5.0,
  "eval_steps": 100,
  "global_step": 1265,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 4.960474308300396e-05,
      "loss": 0.6889,
      "step": 10
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.9209486166007906e-05,
      "loss": 0.6524,
      "step": 20
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.881422924901186e-05,
      "loss": 0.5933,
      "step": 30
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.841897233201581e-05,
      "loss": 0.5607,
      "step": 40
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.8023715415019764e-05,
      "loss": 0.5662,
      "step": 50
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.762845849802372e-05,
      "loss": 0.5097,
      "step": 60
    },
    {
      "epoch": 0.28,
      "learning_rate": 4.723320158102767e-05,
      "loss": 0.436,
      "step": 70
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.683794466403162e-05,
      "loss": 0.3855,
      "step": 80
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.644268774703557e-05,
      "loss": 0.2617,
      "step": 90
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.6047430830039526e-05,
      "loss": 0.2234,
      "step": 100
    },
    {
      "epoch": 0.4,
      "eval_accuracy": 0.9196428571428571,
      "eval_loss": 0.20466722548007965,
      "eval_runtime": 13.5297,
      "eval_samples_per_second": 16.556,
      "eval_steps_per_second": 2.07,
      "step": 100
    },
    {
      "epoch": 0.43,
      "learning_rate": 4.565217391304348e-05,
      "loss": 0.1498,
      "step": 110
    },
    {
      "epoch": 0.47,
      "learning_rate": 4.525691699604743e-05,
      "loss": 0.1202,
      "step": 120
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.4861660079051384e-05,
      "loss": 0.0587,
      "step": 130
    },
    {
      "epoch": 0.55,
      "learning_rate": 4.446640316205534e-05,
      "loss": 0.1286,
      "step": 140
    },
    {
      "epoch": 0.59,
      "learning_rate": 4.4071146245059295e-05,
      "loss": 0.0628,
      "step": 150
    },
    {
      "epoch": 0.63,
      "learning_rate": 4.367588932806324e-05,
      "loss": 0.0803,
      "step": 160
    },
    {
      "epoch": 0.67,
      "learning_rate": 4.32806324110672e-05,
      "loss": 0.0622,
      "step": 170
    },
    {
      "epoch": 0.71,
      "learning_rate": 4.288537549407115e-05,
      "loss": 0.1115,
      "step": 180
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.24901185770751e-05,
      "loss": 0.1047,
      "step": 190
    },
    {
      "epoch": 0.79,
      "learning_rate": 4.2094861660079056e-05,
      "loss": 0.0905,
      "step": 200
    },
    {
      "epoch": 0.79,
      "eval_accuracy": 0.96875,
      "eval_loss": 0.0953177735209465,
      "eval_runtime": 9.8248,
      "eval_samples_per_second": 22.799,
      "eval_steps_per_second": 2.85,
      "step": 200
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1699604743083005e-05,
      "loss": 0.0506,
      "step": 210
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.130434782608696e-05,
      "loss": 0.0257,
      "step": 220
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.0909090909090915e-05,
      "loss": 0.0165,
      "step": 230
    },
    {
      "epoch": 0.95,
      "learning_rate": 4.051383399209486e-05,
      "loss": 0.0876,
      "step": 240
    },
    {
      "epoch": 0.99,
      "learning_rate": 4.011857707509882e-05,
      "loss": 0.0508,
      "step": 250
    },
    {
      "epoch": 1.03,
      "learning_rate": 3.9723320158102766e-05,
      "loss": 0.0192,
      "step": 260
    },
    {
      "epoch": 1.07,
      "learning_rate": 3.932806324110672e-05,
      "loss": 0.0144,
      "step": 270
    },
    {
      "epoch": 1.11,
      "learning_rate": 3.893280632411067e-05,
      "loss": 0.0206,
      "step": 280
    },
    {
      "epoch": 1.15,
      "learning_rate": 3.8537549407114625e-05,
      "loss": 0.0365,
      "step": 290
    },
    {
      "epoch": 1.19,
      "learning_rate": 3.814229249011858e-05,
      "loss": 0.0614,
      "step": 300
    },
    {
      "epoch": 1.19,
      "eval_accuracy": 0.9732142857142857,
      "eval_loss": 0.08801885694265366,
      "eval_runtime": 9.6908,
      "eval_samples_per_second": 23.115,
      "eval_steps_per_second": 2.889,
      "step": 300
    },
    {
      "epoch": 1.23,
      "learning_rate": 3.774703557312253e-05,
      "loss": 0.0209,
      "step": 310
    },
    {
      "epoch": 1.26,
      "learning_rate": 3.735177865612648e-05,
      "loss": 0.0117,
      "step": 320
    },
    {
      "epoch": 1.3,
      "learning_rate": 3.695652173913043e-05,
      "loss": 0.0096,
      "step": 330
    },
    {
      "epoch": 1.34,
      "learning_rate": 3.656126482213439e-05,
      "loss": 0.0399,
      "step": 340
    },
    {
      "epoch": 1.38,
      "learning_rate": 3.616600790513834e-05,
      "loss": 0.0089,
      "step": 350
    },
    {
      "epoch": 1.42,
      "learning_rate": 3.577075098814229e-05,
      "loss": 0.0657,
      "step": 360
    },
    {
      "epoch": 1.46,
      "learning_rate": 3.5375494071146245e-05,
      "loss": 0.01,
      "step": 370
    },
    {
      "epoch": 1.5,
      "learning_rate": 3.49802371541502e-05,
      "loss": 0.0094,
      "step": 380
    },
    {
      "epoch": 1.54,
      "learning_rate": 3.4584980237154155e-05,
      "loss": 0.0336,
      "step": 390
    },
    {
      "epoch": 1.58,
      "learning_rate": 3.418972332015811e-05,
      "loss": 0.0166,
      "step": 400
    },
    {
      "epoch": 1.58,
      "eval_accuracy": 0.9821428571428571,
      "eval_loss": 0.08657606691122055,
      "eval_runtime": 10.0499,
      "eval_samples_per_second": 22.289,
      "eval_steps_per_second": 2.786,
      "step": 400
    },
    {
      "epoch": 1.62,
      "learning_rate": 3.379446640316206e-05,
      "loss": 0.0103,
      "step": 410
    },
    {
      "epoch": 1.66,
      "learning_rate": 3.3438735177865614e-05,
      "loss": 0.0239,
      "step": 420
    },
    {
      "epoch": 1.7,
      "learning_rate": 3.304347826086956e-05,
      "loss": 0.0075,
      "step": 430
    },
    {
      "epoch": 1.74,
      "learning_rate": 3.264822134387352e-05,
      "loss": 0.0065,
      "step": 440
    },
    {
      "epoch": 1.78,
      "learning_rate": 3.225296442687747e-05,
      "loss": 0.0065,
      "step": 450
    },
    {
      "epoch": 1.82,
      "learning_rate": 3.185770750988142e-05,
      "loss": 0.0061,
      "step": 460
    },
    {
      "epoch": 1.86,
      "learning_rate": 3.1462450592885376e-05,
      "loss": 0.0711,
      "step": 470
    },
    {
      "epoch": 1.9,
      "learning_rate": 3.106719367588933e-05,
      "loss": 0.0062,
      "step": 480
    },
    {
      "epoch": 1.94,
      "learning_rate": 3.0671936758893286e-05,
      "loss": 0.0067,
      "step": 490
    },
    {
      "epoch": 1.98,
      "learning_rate": 3.027667984189723e-05,
      "loss": 0.0381,
      "step": 500
    },
    {
      "epoch": 1.98,
      "eval_accuracy": 0.9821428571428571,
      "eval_loss": 0.08877144753932953,
      "eval_runtime": 9.8395,
      "eval_samples_per_second": 22.765,
      "eval_steps_per_second": 2.846,
      "step": 500
    },
    {
      "epoch": 2.02,
      "learning_rate": 2.9881422924901186e-05,
      "loss": 0.0391,
      "step": 510
    },
    {
      "epoch": 2.06,
      "learning_rate": 2.948616600790514e-05,
      "loss": 0.0075,
      "step": 520
    },
    {
      "epoch": 2.09,
      "learning_rate": 2.909090909090909e-05,
      "loss": 0.0058,
      "step": 530
    },
    {
      "epoch": 2.13,
      "learning_rate": 2.8695652173913044e-05,
      "loss": 0.0053,
      "step": 540
    },
    {
      "epoch": 2.17,
      "learning_rate": 2.8300395256916996e-05,
      "loss": 0.0055,
      "step": 550
    },
    {
      "epoch": 2.21,
      "learning_rate": 2.790513833992095e-05,
      "loss": 0.0549,
      "step": 560
    },
    {
      "epoch": 2.25,
      "learning_rate": 2.7509881422924906e-05,
      "loss": 0.0067,
      "step": 570
    },
    {
      "epoch": 2.29,
      "learning_rate": 2.7114624505928854e-05,
      "loss": 0.0176,
      "step": 580
    },
    {
      "epoch": 2.33,
      "learning_rate": 2.671936758893281e-05,
      "loss": 0.0052,
      "step": 590
    },
    {
      "epoch": 2.37,
      "learning_rate": 2.6324110671936758e-05,
      "loss": 0.0053,
      "step": 600
    },
    {
      "epoch": 2.37,
      "eval_accuracy": 0.9776785714285714,
      "eval_loss": 0.10505396872758865,
      "eval_runtime": 9.7768,
      "eval_samples_per_second": 22.911,
      "eval_steps_per_second": 2.864,
      "step": 600
    },
    {
      "epoch": 2.41,
      "learning_rate": 2.5928853754940713e-05,
      "loss": 0.0049,
      "step": 610
    },
    {
      "epoch": 2.45,
      "learning_rate": 2.5533596837944668e-05,
      "loss": 0.0062,
      "step": 620
    },
    {
      "epoch": 2.49,
      "learning_rate": 2.5138339920948616e-05,
      "loss": 0.0053,
      "step": 630
    },
    {
      "epoch": 2.53,
      "learning_rate": 2.4743083003952568e-05,
      "loss": 0.0051,
      "step": 640
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.4347826086956523e-05,
      "loss": 0.0215,
      "step": 650
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.3952569169960475e-05,
      "loss": 0.0043,
      "step": 660
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3557312252964426e-05,
      "loss": 0.0043,
      "step": 670
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.316205533596838e-05,
      "loss": 0.0042,
      "step": 680
    },
    {
      "epoch": 2.73,
      "learning_rate": 2.2766798418972333e-05,
      "loss": 0.0306,
      "step": 690
    },
    {
      "epoch": 2.77,
      "learning_rate": 2.2371541501976288e-05,
      "loss": 0.0043,
      "step": 700
    },
    {
      "epoch": 2.77,
      "eval_accuracy": 0.9776785714285714,
      "eval_loss": 0.08058255165815353,
      "eval_runtime": 10.2106,
      "eval_samples_per_second": 21.938,
      "eval_steps_per_second": 2.742,
      "step": 700
    },
    {
      "epoch": 2.81,
      "learning_rate": 2.197628458498024e-05,
      "loss": 0.0042,
      "step": 710
    },
    {
      "epoch": 2.85,
      "learning_rate": 2.158102766798419e-05,
      "loss": 0.0041,
      "step": 720
    },
    {
      "epoch": 2.89,
      "learning_rate": 2.1185770750988143e-05,
      "loss": 0.0173,
      "step": 730
    },
    {
      "epoch": 2.92,
      "learning_rate": 2.0790513833992095e-05,
      "loss": 0.0042,
      "step": 740
    },
    {
      "epoch": 2.96,
      "learning_rate": 2.0395256916996047e-05,
      "loss": 0.0039,
      "step": 750
    },
    {
      "epoch": 3.0,
      "learning_rate": 2e-05,
      "loss": 0.0039,
      "step": 760
    },
    {
      "epoch": 3.04,
      "learning_rate": 1.9604743083003953e-05,
      "loss": 0.0036,
      "step": 770
    },
    {
      "epoch": 3.08,
      "learning_rate": 1.9209486166007905e-05,
      "loss": 0.0037,
      "step": 780
    },
    {
      "epoch": 3.12,
      "learning_rate": 1.8814229249011857e-05,
      "loss": 0.0036,
      "step": 790
    },
    {
      "epoch": 3.16,
      "learning_rate": 1.8418972332015812e-05,
      "loss": 0.0035,
      "step": 800
    },
    {
      "epoch": 3.16,
      "eval_accuracy": 0.9732142857142857,
      "eval_loss": 0.08644634485244751,
      "eval_runtime": 9.8015,
      "eval_samples_per_second": 22.854,
      "eval_steps_per_second": 2.857,
      "step": 800
    },
    {
      "epoch": 3.2,
      "learning_rate": 1.8023715415019763e-05,
      "loss": 0.0035,
      "step": 810
    },
    {
      "epoch": 3.24,
      "learning_rate": 1.762845849802372e-05,
      "loss": 0.0034,
      "step": 820
    },
    {
      "epoch": 3.28,
      "learning_rate": 1.723320158102767e-05,
      "loss": 0.0035,
      "step": 830
    },
    {
      "epoch": 3.32,
      "learning_rate": 1.6837944664031622e-05,
      "loss": 0.0223,
      "step": 840
    },
    {
      "epoch": 3.36,
      "learning_rate": 1.6442687747035574e-05,
      "loss": 0.0035,
      "step": 850
    },
    {
      "epoch": 3.4,
      "learning_rate": 1.6047430830039525e-05,
      "loss": 0.0035,
      "step": 860
    },
    {
      "epoch": 3.44,
      "learning_rate": 1.565217391304348e-05,
      "loss": 0.0175,
      "step": 870
    },
    {
      "epoch": 3.48,
      "learning_rate": 1.5256916996047432e-05,
      "loss": 0.0044,
      "step": 880
    },
    {
      "epoch": 3.52,
      "learning_rate": 1.4861660079051384e-05,
      "loss": 0.0032,
      "step": 890
    },
    {
      "epoch": 3.56,
      "learning_rate": 1.4466403162055337e-05,
      "loss": 0.0087,
      "step": 900
    },
    {
      "epoch": 3.56,
      "eval_accuracy": 0.9821428571428571,
      "eval_loss": 0.08384133875370026,
      "eval_runtime": 10.4471,
      "eval_samples_per_second": 21.441,
      "eval_steps_per_second": 2.68,
      "step": 900
    },
    {
      "epoch": 3.6,
      "learning_rate": 1.4071146245059289e-05,
      "loss": 0.0071,
      "step": 910
    },
    {
      "epoch": 3.64,
      "learning_rate": 1.367588932806324e-05,
      "loss": 0.0032,
      "step": 920
    },
    {
      "epoch": 3.68,
      "learning_rate": 1.3280632411067195e-05,
      "loss": 0.0081,
      "step": 930
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.2885375494071147e-05,
      "loss": 0.0031,
      "step": 940
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.2490118577075099e-05,
      "loss": 0.0031,
      "step": 950
    },
    {
      "epoch": 3.79,
      "learning_rate": 1.2094861660079052e-05,
      "loss": 0.0031,
      "step": 960
    },
    {
      "epoch": 3.83,
      "learning_rate": 1.1699604743083006e-05,
      "loss": 0.003,
      "step": 970
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.1304347826086957e-05,
      "loss": 0.003,
      "step": 980
    },
    {
      "epoch": 3.91,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 0.0179,
      "step": 990
    },
    {
      "epoch": 3.95,
      "learning_rate": 1.0513833992094862e-05,
      "loss": 0.003,
      "step": 1000
    },
    {
      "epoch": 3.95,
      "eval_accuracy": 0.9821428571428571,
      "eval_loss": 0.07715751230716705,
      "eval_runtime": 9.6117,
      "eval_samples_per_second": 23.305,
      "eval_steps_per_second": 2.913,
      "step": 1000
    },
    {
      "epoch": 3.99,
      "learning_rate": 1.0118577075098814e-05,
      "loss": 0.003,
      "step": 1010
    },
    {
      "epoch": 4.03,
      "learning_rate": 9.723320158102767e-06,
      "loss": 0.0046,
      "step": 1020
    },
    {
      "epoch": 4.07,
      "learning_rate": 9.32806324110672e-06,
      "loss": 0.0044,
      "step": 1030
    },
    {
      "epoch": 4.11,
      "learning_rate": 8.932806324110672e-06,
      "loss": 0.0029,
      "step": 1040
    },
    {
      "epoch": 4.15,
      "learning_rate": 8.537549407114626e-06,
      "loss": 0.0029,
      "step": 1050
    },
    {
      "epoch": 4.19,
      "learning_rate": 8.142292490118577e-06,
      "loss": 0.0061,
      "step": 1060
    },
    {
      "epoch": 4.23,
      "learning_rate": 7.747035573122529e-06,
      "loss": 0.0029,
      "step": 1070
    },
    {
      "epoch": 4.27,
      "learning_rate": 7.351778656126483e-06,
      "loss": 0.0028,
      "step": 1080
    },
    {
      "epoch": 4.31,
      "learning_rate": 6.956521739130435e-06,
      "loss": 0.0029,
      "step": 1090
    },
    {
      "epoch": 4.35,
      "learning_rate": 6.5612648221343875e-06,
      "loss": 0.0147,
      "step": 1100
    },
    {
      "epoch": 4.35,
      "eval_accuracy": 0.9821428571428571,
      "eval_loss": 0.07519398629665375,
      "eval_runtime": 10.7399,
      "eval_samples_per_second": 20.857,
      "eval_steps_per_second": 2.607,
      "step": 1100
    },
    {
      "epoch": 4.39,
      "learning_rate": 6.16600790513834e-06,
      "loss": 0.0029,
      "step": 1110
    },
    {
      "epoch": 4.43,
      "learning_rate": 5.770750988142293e-06,
      "loss": 0.0029,
      "step": 1120
    },
    {
      "epoch": 4.47,
      "learning_rate": 5.375494071146245e-06,
      "loss": 0.0029,
      "step": 1130
    },
    {
      "epoch": 4.51,
      "learning_rate": 4.980237154150198e-06,
      "loss": 0.0029,
      "step": 1140
    },
    {
      "epoch": 4.55,
      "learning_rate": 4.58498023715415e-06,
      "loss": 0.0029,
      "step": 1150
    },
    {
      "epoch": 4.58,
      "learning_rate": 4.189723320158103e-06,
      "loss": 0.0029,
      "step": 1160
    },
    {
      "epoch": 4.62,
      "learning_rate": 3.7944664031620557e-06,
      "loss": 0.0028,
      "step": 1170
    },
    {
      "epoch": 4.66,
      "learning_rate": 3.3992094861660078e-06,
      "loss": 0.0028,
      "step": 1180
    },
    {
      "epoch": 4.7,
      "learning_rate": 3.0039525691699607e-06,
      "loss": 0.0028,
      "step": 1190
    },
    {
      "epoch": 4.74,
      "learning_rate": 2.608695652173913e-06,
      "loss": 0.0052,
      "step": 1200
    },
    {
      "epoch": 4.74,
      "eval_accuracy": 0.9821428571428571,
      "eval_loss": 0.07409419864416122,
      "eval_runtime": 10.6746,
      "eval_samples_per_second": 20.984,
      "eval_steps_per_second": 2.623,
      "step": 1200
    },
    {
      "epoch": 4.78,
      "learning_rate": 2.2134387351778658e-06,
      "loss": 0.0028,
      "step": 1210
    },
    {
      "epoch": 4.82,
      "learning_rate": 1.818181818181818e-06,
      "loss": 0.0028,
      "step": 1220
    },
    {
      "epoch": 4.86,
      "learning_rate": 1.4229249011857708e-06,
      "loss": 0.0102,
      "step": 1230
    },
    {
      "epoch": 4.9,
      "learning_rate": 1.0276679841897233e-06,
      "loss": 0.0028,
      "step": 1240
    },
    {
      "epoch": 4.94,
      "learning_rate": 6.324110671936759e-07,
      "loss": 0.0122,
      "step": 1250
    },
    {
      "epoch": 4.98,
      "learning_rate": 2.3715415019762848e-07,
      "loss": 0.0098,
      "step": 1260
    },
    {
      "epoch": 5.0,
      "step": 1265,
      "total_flos": 1.568437869799342e+18,
      "train_loss": 0.05678829440015107,
      "train_runtime": 1104.709,
      "train_samples_per_second": 18.322,
      "train_steps_per_second": 1.145
    }
  ],
  "logging_steps": 10,
  "max_steps": 1265,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 100,
  "total_flos": 1.568437869799342e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}