INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    创造
    -0.08
     pembangunan
    -0.08
    lec
    -0.07
     Obrig
    -0.07
    jén
    -0.07
     misc
    -0.07
    islation
    -0.07
     infraestrutura
    -0.07
     Infrastructure
    -0.07
    gx
    -0.07
    POSITIVE LOGITS
     erläut
    0.09
    /detail
    0.09
     enlarged
    0.09
    _detail
    0.08
    0.08
    caled
    0.08
     detall
    0.08
    0.08
     aclar
    0.08
     unten
    0.08
    Act Density 0.006%

    No Known Activations