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

    Neuronpedia

    INDEX
    Explanations

    recounting events with people

    np_max-act · gemini-2.0-flash

    The neuron fires on mentions of people (real or generic) and their roles or actions in a news‐reporting context—for example tokens like “individual,” “agent,” “plot,” or “assertions” that call out actors and what they did.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ank
    -0.08
    _o
    -0.07
    ANK
    -0.07
    .SpringBootTest
    -0.07
    .coeff
    -0.06
     ссыл
    -0.06
     Lancaster
    -0.06
     INNER
    -0.06
     extraordinary
    -0.06
     GDP
    -0.06
    POSITIVE LOGITS
     @"
    0.07
    검
    0.06
    key
    0.06
     đứng
    0.06
    [dim
    0.06
    odynamics
    0.06
    handles
    0.06
     accidental
    0.06
    (cd
    0.06
     affection
    0.06
    Act Density 0.017%

    No Known Activations