Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    in mind, allow developers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Crow
    -0.10
     Hiro
    -0.09
     Conway
    -0.09
     unp
    -0.09
     Kah
    -0.09
     hunter
    -0.08
    oader
    -0.08
    âĦ
    -0.08
     Humph
    -0.08
     Tah
    -0.08
    POSITIVE LOGITS
    unic
    0.10
    apg
    0.09
    afb
    0.09
     Hlav
    0.09
     ilma
    0.09
    æŁIJ
    0.09
     æŁ
    0.09
    ¦æĥħ
    0.09
    okt
    0.08
    ¶Į
    0.08
    Act Density 0.264%

    No Known Activations