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

    Neuronpedia

    INDEX
    Explanations

    Proper nouns

    np_max-act · gemini-2.0-flash

    The neuron lights up on proper-name tokens—that is, words or word fragments belonging to named entities (people, organizations, places, etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    “(
    -0.06
    ''
    -0.06
    Graphic
    -0.06
    ивши
    -0.06
    .mi
    -0.06
    instances
    -0.06
    “What
    -0.06
    _att
    -0.05
    ुण
    -0.05
     succeeds
    -0.05
    POSITIVE LOGITS
     seinen
    0.07
     bufio
    0.07
    .strokeStyle
    0.06
    .metroLabel
    0.06
     أبريل
    0.06
    _Register
    0.06
    _subplot
    0.06
     eaten
    0.06
     совсем
    0.06
     pleasantly
    0.06
    Act Density 0.156%

    No Known Activations