INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     любые
    0.87
     herhangi
    0.82
     любых
    0.79
     любой
    0.76
     cualquier
    0.75
     qualsiasi
    0.74
     sahiptir
    0.73
     કોઈપણ
    0.72
     mỗi
    0.72
    なる
    0.71
    POSITIVE LOGITS
     whereas
    1.75
    whereas
    1.58
     when
    1.54
     where
    1.53
    when
    1.47
     Whereas
    1.44
     unnecessarily
    1.39
     obwohl
    1.35
     needlessly
    1.32
     যেখানে
    1.31
    Act Density 0.058%

    No Known Activations