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

    Neuronpedia

    INDEX
    Explanations

    quotation mark

    np_max-act · gemini-2.0-flash

    This neuron activates on JSON‐style object keys—the quoted field names (e.g. "type", "properties", "description") that precede colons.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    istory
    -0.07
     अध
    -0.07
    Все
    -0.06
     Elon
    -0.06
     Massage
    -0.06
     Yug
    -0.06
     bạc
    -0.06
     aux
    -0.06
     Deadly
    -0.06
     fred
    -0.06
    POSITIVE LOGITS
    >//
    0.07
     align
    0.07
     Journalism
    0.06
    _Desc
    0.06
    .GetItem
    0.06
     *↵↵
    0.06
     Tower
    0.06
    _cats
    0.06
     Indigenous
    0.06
    :@{
    0.06
    Act Density 0.006%

    No Known Activations