INDEX
    Explanations

    references to extremist groups and their affiliations

    New Auto-Interp
    Negative Logits
    apan
    -0.17
    /INFO
    -0.15
    ffe
    -0.15
    ickle
    -0.15
    isu
    -0.15
    angan
    -0.14
    -flash
    -0.14
    uzu
    -0.14
    обов
    -0.14
    abin
    -0.14
    POSITIVE LOGITS
     membership
    0.17
    membership
    0.17
     Membership
    0.16
    ãģ°
    0.15
     memberships
    0.15
     member
    0.15
     members
    0.15
     Exist
    0.14
    rale
    0.14
     formed
    0.14
    Act Density 0.435%

    No Known Activations