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

    Neuronpedia

    INDEX
    Explanations

    a

    np_max-act · gemini-2.0-flash

    This neuron activates on the procedural “how‐to” instruction text in code comments (e.g. “In order to … create or edit a type with the same name …”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    	score
    -0.07
     WEIGHT
    -0.07
    dict
    -0.07
    ::
    -0.06
    (button
    -0.06
    (expr
    -0.06
    ignon
    -0.06
    략
    -0.06
    aser
    -0.06
    tickets
    -0.06
    POSITIVE LOGITS
    名無し
    0.07
     �
    0.06
     yö
    0.06
     chuyên
    0.06
    -inc
    0.06
    	UInt
    0.06
    姓名
    0.06
     معرف
    0.06
     nổi
    0.06
    嘛
    0.06
    Act Density 0.008%

    No Known Activations