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

    Neuronpedia

    INDEX
    Explanations

    quotation marks

    np_max-act · gemini-2.0-flash

    This neuron activates on spans of dialogue enclosed in quotation marks—i.e. quoted speech in the conversation.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Services
    -0.06
    .CompilerServices
    -0.06
     Kabul
    -0.06
    ики
    -0.06
     furniture
    -0.06
     republican
    -0.06
     Deborah
    -0.06
    	table
    -0.06
     imágenes
    -0.06
     Bucc
    -0.06
    POSITIVE LOGITS
    ující
    0.07
    ويل
    0.07
    ιν
    0.07
     FOOD
    0.06
     вищ
    0.06
    notations
    0.06
    із
    0.06
    по
    0.06
    >>↵↵
    0.06
     syll
    0.06
    Act Density 0.054%

    No Known Activations