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

    Neuronpedia

    INDEX
    Explanations

    AUTHOR

    np_max-act · gemini-2.0-flash

    The neuron activates on the uppercase token “AUTHOR” (as used in software license header comments).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    GraphNode
    -0.07
     egregious
    -0.06
     Essays
    -0.06
     مذه
    -0.06
    idenav
    -0.06
    .refs
    -0.06
     sharedInstance
    -0.06
    iangle
    -0.06
    Banner
    -0.06
    声明
    -0.06
    POSITIVE LOGITS
     Bankası
    0.08
    [port
    0.07
    "url
    0.07
    袖
    0.07
    มา
    0.07
     aer
    0.07
    《
    0.06
    ;
↵
    0.06
    959
    0.06
     živ
    0.06
    Act Density 0.002%

    No Known Activations