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

    Neuronpedia

    INDEX
    Explanations

    sink

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    (pDX
    -0.06
     TestUtils
    -0.06
    ﭪ
    -0.06
    胸怀
    -0.06
    Indices
    -0.06
     Ann
    -0.06
    🔲
    -0.06
    Peace
    -0.06
    當您
    -0.06
     anonymity
    -0.06
    POSITIVE LOGITS
    echa
    0.07
    ख
    0.07
    兢
    0.07
    缩小
    0.07
    皱
    0.07
    DOCTYPE
    0.07
    орт
    0.07
    лон
    0.07
     diminishing
    0.07
    креп
    0.07
    Act Density 0.004%

    No Known Activations