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

    Neuronpedia

    INDEX
    Explanations

    Scientific publications citations

    np_max-act · gemini-2.0-flash

    This neuron is tuned to detect in-text citation markers or reference key tokens (e.g. the bracketed author–year or DOI reference fragments).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Ag
    -0.06
     xmlns
    -0.06
     overwhelmingly
    -0.06
    .getWorld
    -0.06
    InputGroup
    -0.06
    Creat
    -0.06
    fed
    -0.06
     twig
    -0.06
     newPath
    -0.06
    .home
    -0.06
    POSITIVE LOGITS
    -,
    0.06
    (acc
    0.06
     construction
    0.06
     capacidad
    0.06
    ublic
    0.06
     Phar
    0.06
    IOD
    0.06
    бы
    0.06
    驗
    0.06
     translates
    0.06
    Act Density 0.004%

    No Known Activations