INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     določ
    -0.08
     Rund
    -0.08
    -0.08
    lod
    -0.07
     Brexit
    -0.07
    verbs
    -0.07
     Exodus
    -0.07
    ived
    -0.07
    aye
    -0.07
     lived
    -0.07
    POSITIVE LOGITS
     construir
    0.09
    .construct
    0.08
     обработки
    0.07
     bouwen
    0.07
     chained
    0.07
    0.07
    -sites
    0.07
     Literary
    0.07
     driving
    0.07
     iterations
    0.07
    Act Density 0.001%

    No Known Activations