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

    Neuronpedia

    INDEX
    Explanations

    describing device types/parts

    np_max-act · gemini-2.0-flash

    The neuron is sensitive to discourse markers and enumerators—words like “One,” “Another,” “Finally,” “In,” “Such,” “Type,” etc.—that introduce or signal new sections or examples.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     breaches
    -0.07
     racially
    -0.06
     Husband
    -0.06
     working
    -0.06
     �
    -0.06
    ETERS
    -0.06
     مناطق
    -0.06
    -Semitism
    -0.06
    ,to
    -0.06
    spots
    -0.06
    POSITIVE LOGITS
     آنجا
    0.07
    <ul
    0.06
     ISSN
    0.06
     Tear
    0.06
    /*.
    0.06
     Ebook
    0.06
    τογραφ
    0.06
     مقاله
    0.06
    .panelControl
    0.06
    ony
    0.06
    Act Density 0.033%

    No Known Activations