INDEX
    Explanations

    multiples and divisibility

    New Auto-Interp
    Negative Logits
    …”
    -0.09
     Einige
    -0.09
     vikt
    -0.09
    .coords
    -0.09
     از
    -0.08
    -0.08
    -0.08
    ’를
    -0.08
     competed
    -0.08
     daraus
    -0.08
    POSITIVE LOGITS
     integer
    0.09
    整数
    0.08
     divisible
    0.08
    Inte
    0.08
    -as
    0.08
     fitting
    0.08
     spacing
    0.07
     integers
    0.07
    spacing
    0.07
    Integer
    0.07
    Act Density 0.062%

    No Known Activations