INDEX
    Explanations

    instances of the word "discussion" and related forms

    New Auto-Interp
    Negative Logits
    deen
    -0.14
    lemn
    -0.14
    ura
    -0.14
    PIO
    -0.14
     scoop
    -0.14
    igkeit
    -0.14
    tuk
    -0.14
    ray
    -0.14
    mes
    -0.14
    /by
    -0.14
    POSITIVE LOGITS
    conti
    0.18
    åĢ
    0.15
    incare
    0.14
    Decoration
    0.14
    wt
    0.14
    erializer
    0.14
    ative
    0.14
    ariat
    0.14
    nect
    0.14
    alten
    0.14
    Act Density 0.023%

    No Known Activations