INDEX
    Explanations

    Gemma and similar names

    New Auto-Interp
    Negative Logits
    :
    1.34
     I
    1.34
    ,
    1.28
    ?
    1.13
     It
    1.10
    /
    1.04
    -
    1.00
     Eti
    0.96
     for
    0.94
    Mvc
    0.94
    POSITIVE LOGITS
    ن
    1.82
    1.51
    1.48
    که
    1.41
    Т
    1.41
    t
    1.37
    n
    1.36
    Р
    1.33
    1.30
    1.27
    Act Density 0.025%

    No Known Activations