INDEX
    Explanations

    and before specific nouns

    New Auto-Interp
    Negative Logits
    ZIONE
    -0.88
    ZIONI
    -0.87
    ($(
    -0.87
     pObj
    -0.86
     culturel
    -0.83
    CIL
    -0.82
    -0.81
     fVar
    -0.81
     Sharon
    -0.77
    }});
    -0.77
    POSITIVE LOGITS
    related
    0.98
    PENUTUP
    0.89
    olden
    0.87
    licative
    0.84
    HEMAT
    0.84
     success
    0.84
     扇
    0.82
    ?!”
    0.81
     halter
    0.79
     výkon
    0.79
    Act Density 0.077%

    No Known Activations