INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    in
    1.63
    ي
    1.33
    i
    1.13
    נ
    1.13
    re
    1.10
    ності
    1.09
    inase
    1.09
    י
    1.08
    有利于
    1.06
    ۔
    1.04
    POSITIVE LOGITS
     ion
    1.06
    Ion
    1.04
    0
    1.01
    and
    0.98
     maravill
    0.98
     Ion
    0.96
     escuch
    0.96
     as
    0.95
     eléctricas
    0.94
     emocion
    0.93
    Act Density 0.003%

    No Known Activations