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

    Neuronpedia

    INDEX
    Explanations

    headquarters

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of the word "headquarters" (and its variants), effectively detecting mentions of a company’s headquarters location.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Til
    -0.06
    _animation
    -0.06
    Into
    -0.06
    rites
    -0.06
     tan
    -0.06
     Cell
    -0.06
     sorunu
    -0.06
     Mong
    -0.06
    Camb
    -0.06
    (long
    -0.06
    POSITIVE LOGITS
     headquarters
    0.13
     Headquarters
    0.11
     HQ
    0.10
    quarters
    0.09
    quarter
    0.09
    hq
    0.09
    quartered
    0.08
    QS
    0.08
    HQ
    0.08
     مقر
    0.08
    Act Density 0.005%

    No Known Activations