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

    Neuronpedia

    INDEX
    Explanations

    :

    np_max-act · gemini-2.0-flash

    This neuron fires on HTTP header field names (e.g. “Content-Type:”, “Content-Disposition:”, “SoapAction:”, etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    dw
    -0.06
    609
    -0.06
    607
    -0.06
     pt
    -0.06
    933
    -0.06
    aims
    -0.06
    _offset
    -0.06
    45
    -0.06
     ambassador
    -0.06
    者
    -0.06
    POSITIVE LOGITS
    _fb
    0.07
     yaklaşık
    0.07
     символ
    0.07
     shack
    0.07
     poets
    0.07
    isky
    0.07
     flakes
    0.07
     thanh
    0.06
    itational
    0.06
    essaging
    0.06
    Act Density 0.003%

    No Known Activations