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

    Neuronpedia

    INDEX
    Explanations

    Quotations

    np_max-act · gemini-2.0-flash

    The neuron fires on the single‐word punchline token that ends a two-line joke (e.g. “Dam!” or “Supplies!”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _grp
    -0.06
     Mime
    -0.06
     škol
    -0.06
     Bran
    -0.06
    .);↵
    -0.06
     AU
    -0.06
     Nintendo
    -0.06
     raspberry
    -0.06
     Mater
    -0.06
     Wei
    -0.06
    POSITIVE LOGITS
    Daily
    0.07
    /www
    0.07
    .poster
    0.07
     здійснення
    0.07
     оттен
    0.06
    ाइट
    0.06
    -dismiss
    0.06
    Permission
    0.06
     ایشان
    0.06
     legalize
    0.06
    Act Density 0.006%

    No Known Activations