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

    Neuronpedia

    INDEX
    Explanations

    sectional

    np_max-act · gemini-2.0-flash

    the neuron detects instances of figure-reference phrases describing views (e.g. “cross-sectional view,” “plan view,” “FIG.”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     homework
    -0.07
     /*<<<
    -0.07
    wc
    -0.07
    _tpl
    -0.07
     все
    -0.07
    _Start
    -0.07
    文献
    -0.07
     Quarter
    -0.06
     pcap
    -0.06
    ським
    -0.06
    POSITIVE LOGITS
    regation
    0.06
    lásil
    0.06
     capitalism
    0.06
     comparatively
    0.06
     keynote
    0.06
    amm
    0.06
     价格
    0.06
     admitted
    0.06
     meaning
    0.06
     accompanying
    0.06
    Act Density 0.003%

    No Known Activations