INDEX
    Explanations

    references to tobacco and its products

    New Auto-Interp
    Negative Logits
    ures
    -0.16
    uml
    -0.15
    inkel
    -0.14
    uth
    -0.14
    ãĤ½ãĥ³
    -0.14
    orama
    -0.14
    Ïĥο
    -0.14
    ecute
    -0.14
    vy
    -0.13
    ners
    -0.13
    POSITIVE LOGITS
    ancock
    0.15
    bidden
    0.15
    Ú¯ÛĮ
    0.14
    oÅĻ
    0.14
    strand
    0.14
    ayette
    0.14
    .skills
    0.14
    curacy
    0.14
    erve
    0.14
    auge
    0.13
    Act Density 0.003%

    No Known Activations