INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     lokal
    -0.08
     Travis
    -0.08
     Crazy
    -0.08
    _cmd
    -0.08
    ieras
    -0.08
     rätt
    -0.07
     handy
    -0.07
    iffen
    -0.07
    -0.07
    stances
    -0.07
    POSITIVE LOGITS
     paro
    0.09
     parto
    0.08
    (crate
    0.08
    0.08
    өө
    0.08
    inzi
    0.08
    0.08
    .pag
    0.07
    -pal
    0.07
    0.07
    Act Density 0.007%

    No Known Activations