INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    glise
    -0.08
     happen
    -0.07
     прием
    -0.07
     invaluable
    -0.07
    _hook
    -0.07
     дад
    -0.07
    भूम
    -0.07
    328
    -0.07
     пропис
    -0.07
    -0.07
    POSITIVE LOGITS
     bearings
    0.08
     ducha
    0.08
     ado
    0.08
     Agile
    0.08
     Bearings
    0.08
    pager
    0.08
    keys
    0.08
    .keys
    0.08
     wallpaper
    0.07
    hag
    0.07
    Act Density 0.001%

    No Known Activations