INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    Widths
    -0.08
     alas
    -0.08
    dur
    -0.08
     Morph
    -0.07
    -0.07
     widths
    -0.07
     downs
    -0.07
     grounded
    -0.07
     fabr
    -0.07
     Grip
    -0.07
    POSITIVE LOGITS
    عي
    0.09
     Trudeau
    0.09
     обучение
    0.08
     Beckham
    0.08
     supplementation
    0.08
     увеличение
    0.08
    0.08
     damaging
    0.08
     Herce
    0.07
    0.07
    Act Density 0.002%

    No Known Activations