INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     ReLU
    0.49
     নো
    0.46
     socialista
    0.46
     sali
    0.46
     Operand
    0.46
     CQL
    0.45
     SimSun
    0.44
    論文
    0.44
     dystopian
    0.44
     anarchist
    0.44
    POSITIVE LOGITS
     golf
    2.34
     golfers
    2.22
    Golf
    2.14
     golfing
    2.08
     Golf
    2.06
     golfer
    2.03
    golf
    1.99
    ゴルフ
    1.82
     ゴルフ
    1.76
    🏌
    1.70
    Act Density 0.013%

    No Known Activations