INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     ligação
    -0.08
     containment
    -0.08
    sover
    -0.08
     taken
    -0.07
    -0.07
     organizada
    -0.07
     lifecycle
    -0.07
    solutely
    -0.07
     a
    -0.07
     тэр
    -0.07
    POSITIVE LOGITS
     redesigned
    0.09
    ufig
    0.08
    ителям
    0.08
     Reduced
    0.08
     개선
    0.08
     amélior
    0.08
    Stub
    0.08
     Schre
    0.08
    Reduced
    0.08
    تمبر
    0.08
    Act Density 0.003%

    No Known Activations