INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     parti
    -0.59
     sing
    -0.56
    to
    -0.52
     case
    -0.50
     specific
    -0.50
    te
    -0.49
    ten
    -0.48
     he
    -0.47
    the
    -0.47
    ing
    -0.47
    POSITIVE LOGITS
     صوتيه
    0.92
     विश्वसनीयता
    0.90
    expandindo
    0.90
     незавершена
    0.87
     ویکی‌پدی
    0.82
    Aktualisiert
    0.82
    fören
    0.81
     للمعارف
    0.81
    usahaan
    0.77
     متعلقه
    0.77
    Act Density 0.778%

    No Known Activations