INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    er
    1.84
    ΙΑ
    1.23
    in
    1.19
    en
    1.14
    ر
    1.13
     xấu
    1.10
    ddar
    1.10
    island
    1.10
    kid
    1.10
    lll
    1.10
    POSITIVE LOGITS
     ayudarte
    1.32
    clesiastical
    1.18
     چاندی
    1.15
    1.14
     fün
    1.12
     spoonful
    1.11
     средство
    1.10
    یه
    1.09
     muze
    1.09
     Panasonic
    1.06
    Act Density 0.001%

    No Known Activations