INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    Restore
    -0.08
     barber
    -0.08
    Fax
    -0.08
    Italia
    -0.08
    จริง
    -0.08
    Flutter
    -0.07
     gesch
    -0.07
    ascus
    -0.07
    -0.07
    African
    -0.07
    POSITIVE LOGITS
     interm
    0.08
    portal
    0.07
     bolig
    0.07
     overshadow
    0.07
     sati
    0.07
     Molly
    0.07
     endocrine
    0.07
     Ull
    0.07
     jose
    0.07
     moda
    0.07
    Act Density 0.004%

    No Known Activations