INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     waffle
    -0.08
     carpenter
    -0.08
    -0.08
     Lotus
    -0.08
     buoy
    -0.08
     pillars
    -0.08
     gazebo
    -0.08
     umbrellas
    -0.08
     crm
    -0.08
     Pill
    -0.08
    POSITIVE LOGITS
     grayscale
    0.13
     pixels
    0.11
    _pixels
    0.10
     pixel
    0.10
    pixels
    0.10
    _gray
    0.10
    .Pixel
    0.10
    Pixel
    0.10
    Pixels
    0.10
     Pixel
    0.09
    Act Density 0.007%

    No Known Activations