INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     backpack
    -0.09
    -0.09
     نح
    -0.08
     intime
    -0.08
     conse
    -0.08
    flies
    -0.08
    nums
    -0.08
     barrel
    -0.07
     مرت
    -0.07
    -0.07
    POSITIVE LOGITS
    erezo
    0.09
    指导
    0.08
     gegen
    0.08
     gon
    0.08
     Gis
    0.07
     खत
    0.07
     regulating
    0.07
     Engel
    0.07
     stopping
    0.07
     inhibition
    0.07
    Act Density 0.001%

    No Known Activations