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

    Neuronpedia

    INDEX
    Explanations

    questions and code snippets

    np_max-act · gemini-2.0-flash

    This neuron fires on the initial imperative or command verb that kicks off a user’s request (e.g. “charger,” “requête,” “refaça,” “finja,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     egy
    -0.07
     sł
    -0.06
     Duc
    -0.06
     سیاست
    -0.06
     INLINE
    -0.06
    536
    -0.06
    耀
    -0.06
     contar
    -0.06
    _phr
    -0.06
    Ionic
    -0.06
    POSITIVE LOGITS
     Residents
    0.06
    Database
    0.06
    asha
    0.06
    iversit
    0.06
     roma
    0.06
    ें↵↵
    0.06
    ema
    0.06
     RelayCommand
    0.06
    USE
    0.06
     chuyên
    0.06
    Act Density 0.088%

    No Known Activations