INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    وتر
    -0.91
     it
    -0.87
     both
    -0.86
     countless
    -0.83
     several
    -0.83
    ارف
    -0.79
     so
    -0.79
     this
    -0.79
     cumbersome
    -0.79
     limitations
    -0.78
    POSITIVE LOGITS
     pixels
    1.45
     Pixel
    1.16
    pixel
    1.10
    Pixel
    1.09
     pixel
    1.06
     INDEPENDENT
    1.04
     Pixels
    1.04
    pixels
    0.97
     Px
    0.97
     yp
    0.96
    Act Density 0.001%

    No Known Activations