INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     shareholder
    -0.08
     affair
    -0.07
     yam
    -0.07
     shareholders
    -0.07
    Phys
    -0.07
    Generation
    -0.07
     пул
    -0.07
    Carbon
    -0.07
    ंजन
    -0.07
    (shared
    -0.06
    POSITIVE LOGITS
     Klick
    0.09
     dela
    0.08
     Dixie
    0.08
     hindsight
    0.08
     SOM
    0.07
     negatives
    0.07
     нар
    0.07
     Vou
    0.07
    chá
    0.07
     fahr
    0.07
    Act Density 0.001%

    No Known Activations