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

    Neuronpedia

    INDEX
    Explanations

    "is" and "has" (verbs)

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of “have not been verified” (i.e. negative perfect‐passive phrases indicating unverified information).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Produces
    -0.06
     Process
    -0.06
    odon
    -0.06
    .effect
    -0.06
    estion
    -0.06
     Voter
    -0.06
     Expansion
    -0.06
     铁
    -0.06
     Frankfurt
    -0.05
    ektor
    -0.05
    POSITIVE LOGITS
     Dave
    0.07
     priceless
    0.07
    声明
    0.07
    mute
    0.07
     кг
    0.07
    inheritDoc
    0.07
    arem
    0.07
    ><?
    0.06
    overview
    0.06
     [_
    0.06
    Act Density 0.001%

    No Known Activations