INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     tagline
    -0.08
     maj
    -0.08
     questionnaire
    -0.07
    	Log
    -0.07
    าน
    -0.07
     therm
    -0.07
     Pandora
    -0.07
     potion
    -0.07
     Lys
    -0.07
     kompani
    -0.07
    POSITIVE LOGITS
     Discipline
    0.08
     дисцип
    0.08
    fuck
    0.08
     тиб
    0.08
     FUCK
    0.08
    cuss
    0.08
     painfully
    0.08
     Bezir
    0.08
     hentai
    0.08
     фай
    0.08
    Act Density 0.010%

    No Known Activations