INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     blonde
    -0.08
    olu
    -0.07
     widespread
    -0.07
     COVER
    -0.07
    -0.06
    _bool
    -0.06
    -band
    -0.06
    -0.06
    _square
    -0.06
     Borders
    -0.06
    POSITIVE LOGITS
    //[
    0.06
    INATION
    0.06
    (article
    0.06
    hle
    0.06
    0.06
     portrayal
    0.06
     niche
    0.06
    gün
    0.06
     modific
    0.06
    vector
    0.06
    Act Density 0.002%

    No Known Activations