INDEX
    Explanations

    in followed by origin or manner

    New Auto-Interp
    Negative Logits
    ambilan
    -0.46
    setVerticalGroup
    -0.41
    mtliche
    -0.40
     relevancia
    -0.37
     Viertel
    -0.37
     abusos
    -0.36
     hangat
    -0.36
     Soares
    -0.36
     violência
    -0.36
     violencia
    -0.35
    POSITIVE LOGITS
    LookAnd
    0.63
    цездатний
    0.58
    ThroughAttribute
    0.51
     Produced
    0.51
    хьтан
    0.50
     Crust
    0.50
    Климат
    0.49
     PyLong
    0.49
    InputBorder
    0.48
     Made
    0.48
    Act Density 0.037%

    No Known Activations