INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     breathing
    -0.10
    Br
    -0.08
     a
    -0.08
    _bal
    -0.08
    -0.08
    -0.08
    маз
    -0.08
     passie
    -0.08
     bruto
    -0.08
    br
    -0.07
    POSITIVE LOGITS
     предотвращ
    0.13
     prevents
    0.12
    0.12
     prohibits
    0.12
     Already
    0.11
     forb
    0.11
     Prevent
    0.11
     Tracking
    0.11
     forbid
    0.11
     Duplicate
    0.11
    Act Density 0.011%

    No Known Activations