INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     samtidigt
    -1.48
    τερα
    -1.18
    костюм
    -1.15
    handedly
    -1.08
     précé
    -1.05
     Secondly
    -1.03
     berbeda
    -1.02
    nhi
    -1.01
    спомина
    -1.01
    ompi
    -1.01
    POSITIVE LOGITS
     enfin
    1.70
    Lastly
    1.61
    Finally
    1.59
     finally
    1.58
     lastly
    1.49
     third
    1.48
     Lastly
    1.37
     last
    1.34
     schließlich
    1.33
    Finalmente
    1.30
    Act Density 0.019%

    No Known Activations