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

    Neuronpedia

    INDEX
    Explanations

    Random web content

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on high‐frequency function words (common stopwords like “the,” “and,” “in,” “to,” etc.) rather than content words.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _REGISTRY
    -0.06
     empathy
    -0.06
    аж
    -0.06
    腐
    -0.06
    Cursor
    -0.06
    iterals
    -0.06
    _keeper
    -0.06
     >(
    -0.06
    handleRequest
    -0.06
    IPP
    -0.06
    POSITIVE LOGITS
     phương
    0.07
    кими
    0.07
     Possibly
    0.07
     noticeably
    0.07
     ниж
    0.06
     federation
    0.06
     firstly
    0.06
     absolutely
    0.06
     ما
    0.06
     fell
    0.06
    Act Density 0.011%

    No Known Activations