INDEX
    Explanations

    harmful, unethical, racist, sexist, toxic, dangerous, or illegal

    New Auto-Interp
    Negative Logits
     entrambe
    0.46
     entrambi
    0.46
     oba
    0.44
     beide
    0.42
    xticks
    0.41
    لعاب
    0.40
     ambos
    0.40
    দীর
    0.40
     ridges
    0.39
     obu
    0.38
    POSITIVE LOGITS
     All
    0.40
     அனைத்தும்
    0.39
    All
    0.38
     എന്നിവ
    0.38
     Все
    0.37
    Sense
    0.37
    0.36
     ALL
    0.36
    0.36
     НЕ
    0.36
    Act Density 0.039%

    No Known Activations