INDEX
    Explanations

    Sobel edge detection

    New Auto-Interp
    Negative Logits
    .ui
    -0.08
     nug
    -0.08
     brag
    -0.07
    oleon
    -0.07
     toughness
    -0.07
     ware
    -0.07
     smart
    -0.07
     barang
    -0.07
    .file
    -0.07
     broker
    -0.07
    POSITIVE LOGITS
     gris
    0.08
     Sob
    0.08
     disparity
    0.08
     sob
    0.08
     máscara
    0.08
     diferente
    0.08
    waarden
    0.08
    ")))
    0.07
    нете
    0.07
     fabrikant
    0.07
    Act Density 0.001%

    No Known Activations