INDEX
    Explanations

    words related to artistic and creative professions or attributes

    New Auto-Interp
    Negative Logits
    ï¼Įåħ¶ä¸Ń
    -0.16
     konkrét
    -0.15
    anela
    -0.15
    Including
    -0.15
     коÑĤоÑĢое
    -0.15
     eldre
    -0.15
    plode
    -0.15
    onders
    -0.14
     koje
    -0.14
    ï¼Įå®ĥ
    -0.14
    POSITIVE LOGITS
     who
    0.45
    who
    0.36
     whose
    0.35
     whom
    0.31
     intent
    0.29
    whose
    0.27
     capable
    0.27
     able
    0.26
     Who
    0.26
     with
    0.25
    Act Density 0.478%

    No Known Activations