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

    Neuronpedia

    INDEX
    Explanations

    perform

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on words referring to performance or how well something “performs,” especially forms of the verb “perform.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     llam
    -0.07
     txn
    -0.06
    UCH
    -0.06
    450
    -0.06
     مخروط
    -0.06
     оптим
    -0.06
     INA
    -0.06
     deneyim
    -0.06
     ÜNİ
    -0.06
    arResult
    -0.06
    POSITIVE LOGITS
     perform
    0.08
     performing
    0.07
    wrapped
    0.07
    /Product
    0.07
    _flags
    0.06
     behaved
    0.06
    .constants
    0.06
     Rates
    0.06
    itag
    0.06
    ographically
    0.06
    Act Density 0.073%

    No Known Activations