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

    Neuronpedia

    INDEX
    Explanations

    code punctuation

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on numeric literals (especially the “0” tokens) in the code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Away
    -0.07
    >w
    -0.07
    〇
    -0.06
    .RightToLeft
    -0.06
     اختص
    -0.06
     Einsatz
    -0.06
    ALLED
    -0.06
     ліка
    -0.06
    /"↵↵
    -0.06
    ENCY
    -0.06
    POSITIVE LOGITS
     zvyš
    0.07
    Fl
    0.07
    Graph
    0.06
     Swagger
    0.06
    {{$
    0.06
     decreasing
    0.06
     Truth
    0.06
    는데
    0.06
    Argument
    0.06
     ncols
    0.06
    Act Density 0.006%

    No Known Activations