INDEX
    Explanations

    would affect or suggest variations

    New Auto-Interp
    Negative Logits
     bourgeois
    0.44
    ósfera
    0.44
     skoro
    0.44
    ত্রী
    0.40
    𝜁
    0.40
    ieß
    0.40
     ইস্যুতে
    0.40
    是因为
    0.40
    ћин
    0.39
     года
    0.39
    POSITIVE LOGITS
     correspondingly
    0.48
     involve
    0.46
     involves
    0.45
    やや
    0.44
    それぞれ
    0.44
     requires
    0.44
     slightly
    0.42
     mathematically
    0.41
     specjal
    0.41
     especiales
    0.41
    Act Density 0.002%

    No Known Activations