INDEX
    Explanations

    conversational writing

    New Auto-Interp
    Negative Logits
     pole
    -0.08
     ime
    -0.08
    imet
    -0.08
    inent
    -0.08
     ikon
    -0.08
    fake
    -0.08
     hei
    -0.07
    auv
    -0.07
     jak
    -0.07
     cadáver
    -0.07
    POSITIVE LOGITS
     acaso
    0.10
     WF
    0.09
    Bis
    0.08
    wis
    0.08
    0.08
    0.08
     Bass
    0.07
    worthy
    0.07
     COS
    0.07
    Bass
    0.07
    Act Density 0.021%

    No Known Activations