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

    Neuronpedia

    INDEX
    Explanations

    Code and math functions

    np_max-act · gemini-2.0-flash

    The neuron is detecting lines that include or import external modules or headers (e.g. “#include”, “import”, “include”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    duğu
    -0.07
    lobals
    -0.07
     deadlock
    -0.06
    aras
    -0.06
     Ş
    -0.06
    ть
    -0.06
    Logger
    -0.06
     RETURNS
    -0.06
    두
    -0.06
    *L
    -0.06
    POSITIVE LOGITS
    О
    0.07
     derec
    0.06
    وا
    0.06
     sph
    0.06
     staging
    0.06
    ersed
    0.06
    际
    0.06
     unearth
    0.06
     вспом
    0.06
    θήκη
    0.05
    Act Density 0.029%

    No Known Activations