INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     elective
    -0.08
     reper
    -0.08
     wan
    -0.07
     inclu
    -0.07
     Hubbard
    -0.07
    kal
    -0.07
     anecd
    -0.07
    shopping
    -0.07
    445
    -0.07
    &gt
    -0.07
    POSITIVE LOGITS
    에게
    0.11
    ගේ
    0.09
    들에게
    0.08
     ấy
    0.08
    0.08
     arrested
    0.08
     interviewed
    0.08
     slain
    0.08
     embryos
    0.07
     sœur
    0.07
    Act Density 0.026%

    No Known Activations