INDEX
    Explanations

    phrases encouraging engagement with local stories or news

    New Auto-Interp
    Negative Logits
    mist
    -0.17
    ober
    -0.14
    isd
    -0.14
    led
    -0.14
    ÑģÑĭл
    -0.14
    zier
    -0.14
    ÑĢой
    -0.14
    ant
    -0.14
    tie
    -0.13
     Sullivan
    -0.13
    POSITIVE LOGITS
    iani
    0.18
    ixin
    0.16
    atti
    0.15
    ROTO
    0.14
     vidéo
    0.14
    RAD
    0.14
    ienie
    0.14
    atta
    0.14
    otta
    0.14
    GO
    0.13
    Act Density 0.022%

    No Known Activations