INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     Vidal
    -0.08
     bac
    -0.07
    rawing
    -0.07
    Little
    -0.07
     Blo
    -0.07
     obser
    -0.07
    _From
    -0.07
    Counting
    -0.07
    Captcha
    -0.07
     Gre
    -0.07
    POSITIVE LOGITS
     you'll
    0.09
     تكون
    0.08
     вместо
    0.08
    0.08
     ты
    0.08
    rench
    0.08
     يمكنك
    0.08
     আপনি
    0.08
     encontrarás
    0.07
    :@"%@
    0.07
    Act Density 0.006%

    No Known Activations