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

    Neuronpedia

    INDEX
    Explanations

    addresses

    np_max-act · gemini-2.0-flash

    specific components of addresses, including street names and numbers, to aid in parsing and structuring address information.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron specifically fires on occurrences of the word “address” (and closely related location-descriptor tokens), marking when an address is being discussed or parsed.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _SENS
    -0.08
    esch
    -0.07
    学院
    -0.06
     Path
    -0.06
     Iz
    -0.06
    č
    -0.06
     اظ
    -0.06
     pace
    -0.06
     Intl
    -0.06
     Odin
    -0.06
    POSITIVE LOGITS
    ');");↵
    0.07
     darüber
    0.06
     پشت
    0.06
     عنوان
    0.06
     nec
    0.06
    ;$
    0.06
    /<
    0.06
    (element
    0.06
    인이
    0.06
    陷
    0.06
    Act Density 0.016%

    No Known Activations