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

    Neuronpedia

    INDEX
    Explanations

    Unreadable or strange text

    np_max-act · gemini-2.0-flash

    This neuron detects placeholder tokens for named entities (e.g., “NAME_1,” “NAME_2,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    	Simple
    -0.07
    "];↵↵
    -0.07
    [prop
    -0.07
    riv
    -0.06
     muslim
    -0.06
    .Ph
    -0.06
     쉬
    -0.06
     başlayan
    -0.06
    .syn
    -0.06
     PPP
    -0.06
    POSITIVE LOGITS
    ække
    0.07
    Async
    0.06
    (Collection
    0.06
    ości
    0.06
    =headers
    0.06
    document
    0.06
    .GetType
    0.06
    _RENDER
    0.06
     ng
    0.06
    ,Integer
    0.06
    Act Density 0.041%

    No Known Activations