INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     amad
    -0.08
     washes
    -0.08
    izando
    -0.08
    -0.08
     dut
    -0.07
    lande
    -0.07
    PRISE
    -0.07
     fixing
    -0.07
     onboarding
    -0.07
     washing
    -0.07
    POSITIVE LOGITS
    利益
    0.10
    竞争
    0.10
    0.09
     perju
    0.08
     antagon
    0.08
    0.08
    vie
    0.08
    -share
    0.08
    .extract
    0.08
    ellipsis
    0.08
    Act Density 0.006%

    No Known Activations