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

    Neuronpedia

    INDEX
    Explanations

    Code/error messages

    np_max-act · gemini-2.0-flash

    The neuron fires on uppercase programming identifiers—especially error‐code or constant names (all‐caps words with underscores).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Ver
    -0.07
    clared
    -0.06
    /pr
    -0.06
     ked
    -0.06
     cit
    -0.06
     DLL
    -0.06
     závis
    -0.06
     volont
    -0.06
     videa
    -0.06
    usal
    -0.06
    POSITIVE LOGITS
     Wholesale
    0.07
     فيه
    0.07
    Animate
    0.06
    оюз
    0.06
     Grimm
    0.06
     Benson
    0.06
    imleri
    0.06
     Levine
    0.06
     Stocks
    0.06
    属于
    0.06
    Act Density 0.017%

    No Known Activations