INDEX
    Explanations

    punctuation marks, such as semicolons, parentheses, and commas, suggesting it is looking for structural elements typical in programming or logical expressions

    New Auto-Interp
    Negative Logits
     NDEBUG
    -0.53
    cap
    -0.46
    BuildContext
    -0.46
    anka
    -0.46
     nowrap
    -0.45
     cápsulas
    -0.45
    ADES
    -0.45
    maan
    -0.45
    roon
    -0.44
    des
    -0.43
    POSITIVE LOGITS
     himo
    0.97
    )))))
    0.94
    )))));
    0.93
    enderror
    0.89
     חיצוניים
    0.89
     kaarangay
    0.87
     Италијани
    0.84
     ostavi
    0.82
    ))));
    0.82
     مشين
    0.81
    Act Density 0.671%

    No Known Activations