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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    This neuron detects programming API identifiers—i.e. tokens that are parts of fully qualified class, interface, function, or module names.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     GTX
    -0.06
    Frequency
    -0.06
     ра
    -0.06
     baiser
    -0.06
     Huawei
    -0.06
    (){
↵
    -0.06
    svc
    -0.06
    _numero
    -0.06
     Burg
    -0.06
    _COUNTRY
    -0.06
    POSITIVE LOGITS
     proceeds
    0.07
     elemental
    0.06
    弗
    0.06
     militia
    0.06
     opportunity
    0.06
    ,list
    0.06
     militias
    0.06
     πολ
    0.06
    (dest
    0.06
     discussion
    0.06
    Act Density 0.024%

    No Known Activations