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

    Neuronpedia

    INDEX
    Explanations

    tokenized in

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Pron
    -0.09
     pron
    -0.09
     weap
    -0.08
    ï¼ģ”;\n\n
    -0.08
    ï¼Ł”;\n\n
    -0.08
    rape
    -0.08
    amas
    -0.08
     either
    -0.08
     â
    -0.07
     (?)
    -0.07
    POSITIVE LOGITS
    ÂĢÂĢ
    0.10
    įng
    0.09
    ¦æĥħ
    0.09
    ¶Į
    0.09
    łéĻ¤
    0.09
    ekim
    0.09
    .Formatter
    0.09
    ³ç´°
    0.08
    ÑŁ
    0.08
    ¿ÃĤ
    0.08
    Act Density 0.362%

    No Known Activations