INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     cheque
    -0.08
    .flatten
    -0.08
     ограничения
    -0.07
     Prism
    -0.07
    Wal
    -0.07
    -We
    -0.07
     Scouts
    -0.07
     requisito
    -0.07
     принимать
    -0.07
     unig
    -0.07
    POSITIVE LOGITS
                      
    0.08
     efficacy
    0.08
    plant
    0.08
    ocation
    0.08
    fitness
    0.07
     Haush
    0.07
     household
    0.07
                     
    0.07
     eficácia
    0.07
    0.07
    Act Density 0.008%

    No Known Activations