INDEX
    Explanations

    references to white supremacist groups and ideologies

    New Auto-Interp
    Negative Logits
    abic
    -0.15
    عاÙĦ
    -0.15
    INTR
    -0.15
    bjerg
    -0.15
    _beam
    -0.14
    ä»Ļ
    -0.14
    765
    -0.14
    ãĥĢãĥ¼
    -0.14
    rokes
    -0.14
    rij
    -0.14
    POSITIVE LOGITS
     offline
    0.17
     Explicit
    0.16
     overlap
    0.15
     recruitment
    0.15
     explicit
    0.15
     incel
    0.15
     Recruitment
    0.14
     gent
    0.14
     overlaps
    0.14
     active
    0.14
    Act Density 0.043%

    No Known Activations