INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     His
    -0.08
    hic
    -0.07
     God
    -0.07
    ouri
    -0.07
    gis
    -0.07
     Ak
    -0.07
     goodness
    -0.07
     Det
    -0.07
     Beer
    -0.07
    ihen
    -0.07
    POSITIVE LOGITS
     아닌
    0.12
     maupun
    0.11
     whereas
    0.11
     tandis
    0.11
     아니라
    0.10
     quella
    0.08
     quello
    0.08
    なく
    0.08
     versus
    0.08
     Whereas
    0.08
    Act Density 0.241%

    No Known Activations