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

    Neuronpedia

    INDEX
    Explanations

    society

    np_max-act · gemini-2.0-flash

    The neuron fires most strongly on nouns naming groups or categories—especially in titles or lists—such as “Partner States,” “basic tastes,” “key resources,” “teams,” “continents,” etc.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     ذكر
    -0.07
    「お
    -0.07
     お
    -0.07
    _tt
    -0.06
     samot
    -0.06
     tren
    -0.06
     잔
    -0.06
    /light
    -0.06
     딸
    -0.06
     아버지
    -0.06
    POSITIVE LOGITS
    igators
    0.06
    _IRQn
    0.06
    WI
    0.06
    errals
    0.06
     اولیه
    0.06
     투
    0.06
     Stap
    0.06
     statements
    0.06
     TCHAR
    0.06
    CoreApplication
    0.06
    Act Density 0.280%

    No Known Activations