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

    Neuronpedia

    INDEX
    Explanations

    original objects and places

    np_max-act · gemini-2.0-flash

    The neuron fires on descriptive terms that signal historical authenticity or preservation—words like “original,” “preserved,” “intact,” and similar adjectives in building or site descriptions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     bt
    -0.06
     dispers
    -0.06
     підпис
    -0.06
    ReLU
    -0.06
    fstream
    -0.06
    inputs
    -0.06
     Collector
    -0.06
    �
    -0.06
     ◆
    -0.06
    ือด
    -0.06
    POSITIVE LOGITS
    )}}"
    0.08
    ,:)
    0.07
     Pad
    0.07
    APER
    0.07
     Lower
    0.06
     mẹ
    0.06
     Attend
    0.06
    تف
    0.06
     мої
    0.06
     DISPLAY
    0.06
    Act Density 0.121%

    No Known Activations