INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    (),
    -0.08
     beard
    -0.08
     steadfast
    -0.08
     astonishing
    -0.08
     Brow
    -0.07
    الت
    -0.07
     nonetheless
    -0.07
     Menge
    -0.07
     boh
    -0.07
    -0.07
    POSITIVE LOGITS
     Memo
    0.08
     Props
    0.08
     propensity
    0.08
     RJ
    0.08
     Raised
    0.08
    FW
    0.07
    0.07
    적으로
    0.07
     SR
    0.07
    ports
    0.07
    Act Density 0.002%

    No Known Activations