INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     NT
    -0.08
     epit
    -0.07
     branch
    -0.07
    -0.07
     MPI
    -0.07
    krat
    -0.07
     Greek
    -0.07
     Musik
    -0.07
     eru
    -0.07
    quel
    -0.07
    POSITIVE LOGITS
    Shield
    0.09
    0.09
     Shield
    0.09
     diaphr
    0.08
    АР
    0.08
     склада
    0.08
    .sections
    0.08
    етель
    0.08
     retirada
    0.08
    ))+
    0.08
    Act Density 0.006%

    No Known Activations