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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on object-oriented class declarations and their accompanying structural tokens (e.g. “class,” “extends,” and “{”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     carbohydrate
    -0.07
     på
    -0.07
    import
    -0.07
    wendung
    -0.06
     zač
    -0.06
    Steel
    -0.06
     MPH
    -0.06
    .stderr
    -0.06
     Wag
    -0.06
     disruptions
    -0.06
    POSITIVE LOGITS
    领域
    0.07
     grosse
    0.06
    enaire
    0.06
     setType
    0.06
     seating
    0.06
     관리자
    0.06
    ScrollBar
    0.06
     odstran
    0.06
     dlou
    0.06
     desarrollo
    0.06
    Act Density 0.073%

    No Known Activations