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

    Neuronpedia

    INDEX
    Explanations

    characterized

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Routing
    -0.07
    '><
    -0.07
    ULO
    -0.07
    unj
    -0.07
    ;&#
    -0.07
    /prom
    -0.07
     Inv
    -0.06
    𝐖
    -0.06
     compassionate
    -0.06
    新征程
    -0.06
    POSITIVE LOGITS
    	operator
    0.07
     investigators
    0.07
    㺾
    0.07
    蕺
    0.07
     libert
    0.07
    宰
    0.07
     taxi
    0.06
     left
    0.06
     Ice
    0.06
     cellar
    0.06
    Act Density 0.022%

    No Known Activations