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

    Neuronpedia

    INDEX
    Explanations

    System failures

    np_max-act · gemini-2.0-flash

    The neuron fires on mentions of formal rules, protocols, or safety/security procedures (e.g. “lock and key,” “two-man rule,” “protocol for ensuring crane safety”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     playful
    -0.09
     Kapoor
    -0.08
     faith
    -0.07
    omid
    -0.07
    338
    -0.07
    ooks
    -0.06
    ост
    -0.06
     ire
    -0.06
    Knife
    -0.06
     blanket
    -0.06
    POSITIVE LOGITS
     searchData
    0.07
    Î
    0.07
     Towards
    0.07
    ‌پدیای
    0.06
     finans
    0.06
    _CLR
    0.06
    @return
    0.06
     작성
    0.06
     OnTriggerEnter
    0.06
    (Sprite
    0.06
    Act Density 0.065%

    No Known Activations