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

    Neuronpedia

    INDEX
    Explanations

    magnitude

    np_max-act · gemini-2.0-flash

    The neuron activates on words that describe or qualify motion speed (e.g. “fast,” “slowly,” “quickly,” “speed,” “higher,” “lower”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     documentaries
    -0.07
    /provider
    -0.07
    Opcode
    -0.06
     Rad
    -0.06
    Silver
    -0.06
     Beck
    -0.06
     reprodu
    -0.06
     Lazar
    -0.06
     Barbara
    -0.06
     Stefan
    -0.06
    POSITIVE LOGITS
    _sort
    0.07
     mRNA
    0.07
     influences
    0.06
     Cory
    0.06
     upkeep
    0.06
     toch
    0.06
    Türkiye
    0.06
    .strings
    0.06
    .FILE
    0.06
     useEffect
    0.06
    Act Density 0.081%

    No Known Activations