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

    Neuronpedia

    INDEX
    Explanations

    Judge

    np_max-act · gemini-2.0-flash

    The neuron detects occurrences of the word “Judge” (i.e. judicial titles in court‐opinion headings).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Reid
    -0.07
    preced
    -0.07
    PROPERTY
    -0.06
    aptic
    -0.06
     standards
    -0.06
    ))))
    -0.06
     harmless
    -0.06
    conversion
    -0.06
     conviction
    -0.06
    REL
    -0.06
    POSITIVE LOGITS
     Mega
    0.07
    /Register
    0.07
    ิสต
    0.06
     delta
    0.06
     helm
    0.06
     bere
    0.06
     combo
    0.06
    _NPC
    0.06
     ine
    0.06
    umin
    0.06
    Act Density 0.002%

    No Known Activations