INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     primit
    -0.08
    IRONMENT
    -0.08
     Roche
    -0.08
    .at
    -0.08
    eur
    -0.08
     Achter
    -0.08
    다가
    -0.08
    riangle
    -0.08
     Erschein
    -0.07
    dens
    -0.07
    POSITIVE LOGITS
    /master
    0.09
    /detail
    0.08
    fully
    0.08
    0.08
    ত্ব
    0.08
    0.08
    0.07
     engross
    0.07
     electric
    0.07
    0.07
    Act Density 0.018%

    No Known Activations