INDEX
    Explanations

    mathematical formulas

    New Auto-Interp
    Negative Logits
     any
    -0.09
     a
    -0.09
     certain
    -0.09
     multiple
    -0.09
     and
    -0.09
    /
    -0.08
     or
    -0.08
     et
    -0.08
     another
    -0.08
     sg
    -0.07
    POSITIVE LOGITS
     Knowing
    0.09
     знать
    0.09
    тах
    0.09
     Accordingly
    0.09
    Knowing
    0.09
    merci
    0.09
     numerator
    0.08
     рәх
    0.08
     ദേശീയ
    0.08
     ,↵↵
    0.08
    Act Density 0.034%

    No Known Activations