INDEX
    Explanations

    variations of the suffix "er" in words

    New Auto-Interp
    Negative Logits
    ing
    -0.41
    ed
    -0.32
    en
    -0.30
    on
    -0.29
    ا
    -0.25
    m
    -0.25
    ic
    -0.23
    d
    -0.23
    al
    -0.23
    icine
    -0.23
    POSITIVE LOGITS
    cury
    0.20
    an
    0.19
    ousel
    0.17
    itage
    0.17
    ilyn
    0.16
    obic
    0.16
    usalem
    0.16
    getic
    0.15
    GES
    0.15
    uida
    0.15
    Act Density 0.035%

    No Known Activations