INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     rak
    -0.08
     Sex
    -0.08
    -0.07
     Quentin
    -0.07
    -0.07
     sex
    -0.07
     מד
    -0.07
     sexes
    -0.07
     salut
    -0.07
     Spray
    -0.07
    POSITIVE LOGITS
     wrongdoing
    0.10
     capitalism
    0.09
     whistle
    0.09
    čí
    0.08
    保証
    0.08
     workplace
    0.08
     safegu
    0.08
    blower
    0.08
     upheld
    0.08
     ultime
    0.08
    Act Density 0.005%

    No Known Activations