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

    Neuronpedia

    INDEX
    Explanations

    !

    np_max-act · gemini-2.0-flash

    The neuron activates on template code that inspects a document’s metadata—particularly on expressions accessing or comparing properties like doc.kind, doc.signature, and similar doc.* fields.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     jd
    -0.07
     деятель
    -0.07
    nym
    -0.06
    Rate
    -0.06
     journée
    -0.06
    conexion
    -0.06
    .cl
    -0.06
    ности
    -0.06
     Cultural
    -0.06
    icana
    -0.06
    POSITIVE LOGITS
     Msg
    0.07
     Massive
    0.07
    日
    0.07
    TypeInfo
    0.07
    C
    0.06
    .toolStripSeparator
    0.06
     Un
    0.06
     akin
    0.06
     Aust
    0.06
     بنا
    0.06
    Act Density 0.003%

    No Known Activations