INDEX
    Explanations

    variations of the suffix "-en" in words

    New Auto-Interp
    Negative Logits
     aerial
    -0.16
    adesh
    -0.15
    eldom
    -0.15
    ickers
    -0.15
    resh
    -0.15
    erece
    -0.14
    atti
    -0.14
     Tou
    -0.14
     Y
    -0.14
     brick
    -0.14
    POSITIVE LOGITS
    iscard
    0.18
    onomy
    0.15
    à¥įवव
    0.15
    497
    0.15
    rud
    0.15
    terminal
    0.15
    ixin
    0.14
    _lazy
    0.14
    _gs
    0.14
    stein
    0.14
    Act Density 0.007%

    No Known Activations