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

    Neuronpedia

    INDEX
    Explanations

    RUN

    np_max-act · gemini-2.0-flash

    The neuron fires on automated test‐driver directives embedded in comments (e.g. lines beginning with “// RUN:”, “// CHECK:”, etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Nab
    -0.07
     Fit
    -0.07
     Naz
    -0.06
    coop
    -0.06
    ělí
    -0.06
    Dream
    -0.06
    ira
    -0.06
    aber
    -0.06
    部分
    -0.06
     decipher
    -0.06
    POSITIVE LOGITS
     berk
    0.07
    义
    0.06
    criteria
    0.06
     padr
    0.06
    QUERY
    0.06
    асти
    0.06
    (sym
    0.06
    %">↵
    0.06
    ерти
    0.06
     ansible
    0.06
    Act Density 0.020%

    No Known Activations