INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     Reform
    -0.06
    гот
    -0.06
     Pett
    -0.06
    door
    -0.06
     gangs
    -0.06
    ét
    -0.06
    (!_
    -0.06
    business
    -0.06
     phát
    -0.06
    jíž
    -0.06
    POSITIVE LOGITS
     ultrasound
    0.13
    trasound
    0.10
     inscription
    0.07
    0.07
    영상
    0.07
     Cry
    0.06
    0.06
    ून
    0.06
     فوق
    0.06
    Instr
    0.06
    Act Density 0.003%

    No Known Activations