INDEX
    Explanations

    questions or competitions

    New Auto-Interp
    Negative Logits
     ilç
    -0.79
     ĝ
    -0.75
    genres
    -0.73
     menampilkan
    -0.73
    frutar
    -0.72
     kennis
    -0.71
    helf
    -0.71
     combines
    -0.71
    doir
    -0.71
    Odon
    -0.71
    POSITIVE LOGITS
    tizens
    0.81
    ثل
    0.79
     sonder
    0.79
     ป
    0.73
    ijan
    0.73
    0.73
    onday
    0.72
     сайта
    0.71
     вечер
    0.71
     limestone
    0.71
    Act Density 0.011%

    No Known Activations