INDEX
    Explanations

    limits as x approaches 0

    New Auto-Interp
    Negative Logits
     face
    -0.10
     Face
    -0.09
    \Bundle
    -0.09
    (face
    -0.09
    Face
    -0.09
    _face
    -0.08
    .Face
    -0.08
    .face
    -0.08
     Daha
    -0.08
     Kaffee
    -0.08
    POSITIVE LOGITS
     negligible
    0.09
     vanish
    0.08
     محور
    0.08
     creeping
    0.08
     visibilité
    0.08
    sweet
    0.08
    iesa
    0.07
     dispara
    0.07
    gone
    0.07
     ചേ
    0.07
    Act Density 0.037%

    No Known Activations