INDEX
    Explanations

    descriptive attributes of subjects in the text

    New Auto-Interp
    Negative Logits
     Purple
    -0.15
    és
    -0.15
     OCR
    -0.15
    iday
    -0.14
    chet
    -0.14
    adge
    -0.14
    erland
    -0.14
    elter
    -0.14
    éļł
    -0.14
    chten
    -0.13
    POSITIVE LOGITS
     ла
    0.17
     consist
    0.17
    .Clone
    0.15
    ino
    0.15
    ODULE
    0.14
    nez
    0.14
     모ìĬµ
    0.14
     adultos
    0.14
    åijĪ
    0.14
    ÏģÏĩ
    0.14
    Act Density 0.360%

    No Known Activations