INDEX
    Explanations

    references to races, racism, and related societal issues

    New Auto-Interp
    Negative Logits
    ican
    -0.17
    lingen
    -0.15
    ¯¯
    -0.14
     ApiController
    -0.14
    enaire
    -0.14
     sparing
    -0.14
    less
    -0.14
     Lama
    -0.14
     Crafts
    -0.13
    linger
    -0.13
    POSITIVE LOGITS
    assic
    0.15
    acho
    0.15
    adamente
    0.14
    ewis
    0.14
    NetMessage
    0.14
    tuk
    0.14
    /filepath
    0.13
    avou
    0.13
     Ders
    0.13
    είο
    0.13
    Act Density 0.195%

    No Known Activations