INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     eerie
    -0.08
    ziehen
    -0.08
    apped
    -0.08
     Monument
    -0.07
    Recorder
    -0.07
     monument
    -0.07
     carving
    -0.07
     großer
    -0.07
    ärken
    -0.07
    Gl
    -0.07
    POSITIVE LOGITS
     fashionable
    0.12
     boutiques
    0.11
     trendy
    0.10
     stylish
    0.09
     savvy
    0.09
     outfit
    0.09
     chic
    0.09
    fashion
    0.09
     sofistic
    0.09
     outfits
    0.08
    Act Density 0.029%

    No Known Activations