INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    +i
    -0.07
    Trial
    -0.07
     nucle
    -0.07
     north
    -0.07
     mountains
    -0.07
     ostensibly
    -0.06
    ynam
    -0.06
     wor
    -0.06
    utschein
    -0.06
     attractive
    -0.06
    POSITIVE LOGITS
     resize
    0.09
     resizing
    0.09
    jpeg
    0.08
    resize
    0.08
    ogui
    0.07
     Resize
    0.07
    Resize
    0.07
    кость
    0.07
    .resize
    0.07
     resized
    0.07
    Act Density 0.004%

    No Known Activations