INDEX
    Explanations

    endings like ia, ion, ian

    New Auto-Interp
    Negative Logits
    د
    1.48
    ד
    1.28
    al
    1.18
    ed
    1.17
    1.15
    et
    1.13
    de
    1.09
    1.09
    ור
    1.05
    e
    1.04
    POSITIVE LOGITS
    во
    1.29
    یر
    1.19
    _
    1.19
    ا
    1.19
    1.16
    не
    1.15
    ı
    1.15
    ang
    1.13
    е
    1.13
    و
    1.09
    Act Density 0.291%

    No Known Activations