INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     faibles
    -0.08
     (+
    -0.08
     strong
    -0.08
     hep
    -0.07
     fortes
    -0.07
     fuertes
    -0.07
     SPL
    -0.07
    _cos
    -0.07
     militar
    -0.07
     weak
    -0.07
    POSITIVE LOGITS
    ppel
    0.08
     transmitir
    0.08
     mystical
    0.08
     вернуть
    0.08
     revisar
    0.08
    orderen
    0.08
     coloration
    0.08
    дігі
    0.08
     aparecer
    0.08
    ndür
    0.08
    Act Density 0.003%

    No Known Activations