INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     offenses
    -0.08
     quantit
    -0.08
     depan
    -0.08
     kickoff
    -0.07
     üks
    -0.07
     halve
    -0.07
     abuses
    -0.07
    -0.07
     узнать
    -0.07
    teach
    -0.07
    POSITIVE LOGITS
    普通
    0.10
     acrylic
    0.09
    crylic
    0.08
     ceramics
    0.08
    0.08
     натураль
    0.08
     soms
    0.08
     parfois
    0.08
     precious
    0.08
     ballast
    0.08
    Act Density 0.029%

    No Known Activations