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

    Neuronpedia

    INDEX
    Explanations

    Buzz

    np_max-act · gemini-2.0-flash

    The neuron activates strongly on occurrences of the tokens “fizz” and “buzz,” i.e. it detects mentions of the “fizzbuzz” keyword.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Cf
    -0.07
    andan
    -0.07
     mobility
    -0.07
     seed
    -0.06
    ως
    -0.06
     har
    -0.06
    окси
    -0.06
    ходим
    -0.06
     sphere
    -0.06
     catalyst
    -0.06
    POSITIVE LOGITS
    ifar
    0.07
     commerc
    0.06
    _tooltip
    0.06
    _PUR
    0.06
    _shuffle
    0.06
    isVisible
    0.06
     fencing
    0.06
    ivement
    0.06
    .delete
    0.06
    992
    0.06
    Act Density 0.002%

    No Known Activations