INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     spectro
    -0.08
     tham
    -0.07
    margin
    -0.06
     KB
    -0.06
     sabe
    -0.06
     tempo
    -0.06
    Maker
    -0.06
     roc
    -0.06
     Workbook
    -0.06
    hra
    -0.06
    POSITIVE LOGITS
     kla
    0.06
    ‌هاي
    0.06
     Self
    0.06
    0.06
     şöyle
    0.06
    LOTS
    0.06
     temiz
    0.06
    ่าก
    0.06
     만들
    0.06
    0.06
    Act Density 0.003%

    No Known Activations