INDEX
    Explanations

    say the ending -ated

    New Auto-Interp
    Negative Logits
    ued
    -0.08
    rex
    -0.08
    cret
    -0.07
    uée
    -0.07
    ues
    -0.07
    Arabic
    -0.07
     voeg
    -0.07
     chemin
    -0.07
     agricole
    -0.07
    =str
    -0.07
    POSITIVE LOGITS
    ated
    0.21
    ating
    0.21
    atings
    0.18
    ATED
    0.18
    ATING
    0.16
    ats
    0.16
    ater
    0.14
    atable
    0.14
    AT
    0.12
    aters
    0.12
    Act Density 0.003%

    No Known Activations