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

    Neuronpedia

    INDEX
    Explanations

    safeguard

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    飞
    -0.08
     trả
    -0.07
    ╔
    -0.07
    纻
    -0.07
     descended
    -0.07
     ö
    -0.07
     translateY
    -0.07
    忘记了
    -0.07
     מאות
    -0.07
    远处
    -0.07
    POSITIVE LOGITS
    7
    0.07
     caveat
    0.07
    八卦
    0.07
    HELP
    0.07
     Whole
    0.06
     lethal
    0.06
    إصلاح
    0.06
    ﮨ
    0.06
     Balanced
    0.06
     Palace
    0.06
    Act Density 0.004%

    No Known Activations