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

    Neuronpedia

    INDEX
    Explanations

    ="

    np_max-act · gemini-2.0-flash

    elements related to government publications and metadata.

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

    This neuron activates on XML/HTML attribute names and their associated punctuation (e.g. the type, authority, media_type, xmlns, “=” and opening quote tokens), effectively detecting attribute definitions in markup.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Arabian
    -0.06
     usa
    -0.06
     Sanctuary
    -0.06
     bluff
    -0.06
    Clinton
    -0.06
     EST
    -0.06
     Laden
    -0.06
    iling
    -0.06
     Minuten
    -0.05
     Alaska
    -0.05
    POSITIVE LOGITS
     personalized
    0.07
     lượng
    0.07
     viewType
    0.07
    _STANDARD
    0.07
     FStar
    0.06
     thăm
    0.06
    _density
    0.06
     Reign
    0.06
    خی
    0.06
     '}↵
    0.06
    Act Density 0.002%

    No Known Activations