INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    ार्थ
    -0.09
     Plaint
    -0.08
     imposs
    -0.08
    라는
    -0.07
    overs
    -0.07
     confession
    -0.07
     inoc
    -0.07
     truthful
    -0.07
    -0.07
     Craig
    -0.07
    POSITIVE LOGITS
    是多少
    0.09
    -dependent
    0.09
     lama
    0.09
     tinggi
    0.08
     dependant
    0.08
     زیادی
    0.08
    (Window
    0.07
    ierung
    0.07
     Cosm
    0.07
     Vogue
    0.07
    Act Density 0.009%

    No Known Activations