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

    Neuronpedia

    INDEX
    Explanations

    period

    np_max-act · gemini-2.0-flash

    The neuron fires on the single‐word answer tokens provided by the assistant (e.g. “Bulldozer,” “Lemon,” “Egg,” “QTabBar,” “stderr,” “Death”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .findAll
    -0.06
     زیست
    -0.06
    ocoder
    -0.06
    -0.06
    _printf
    -0.06
    Customer
    -0.06
     Ваш
    -0.06
    	include
    -0.06
    knowledge
    -0.06
    зация
    -0.06
    POSITIVE LOGITS
    име
    0.08
    قام
    0.07
     Ödül
    0.07
    Winter
    0.07
     xuống
    0.07
    -Jun
    0.07
     उद
    0.06
     Ins
    0.06
     Civ
    0.06
    选
    0.06
    Act Density 0.071%

    No Known Activations