INDEX
    Explanations

    words related to violence and criminal behavior

    New Auto-Interp
    Negative Logits
    ciplinary
    -0.80
    iosyncr
    -0.77
     spons
    -0.77
    translation
    -0.76
    taboola
    -0.75
    pora
    -0.70
    pmwiki
    -0.69
    lighting
    -0.69
    aston
    -0.68
    fitted
    -0.66
    POSITIVE LOGITS
     him
    1.32
     somebody
    1.20
     someone
    1.19
     anybody
    1.15
     them
    1.13
     anyone
    1.06
     Him
    1.03
     oneself
    0.96
    someone
    0.96
     whoever
    0.95
    Act Density 0.261%

    No Known Activations