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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron is activated by the punctuation tokens used in code comments and documentation (e.g. the leading “*” and “@” markers in Javadoc‐style comments).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ubernetes
    -0.07
     tkinter
    -0.07
     coastal
    -0.07
    _runner
    -0.07
    formik
    -0.07
     spectra
    -0.06
     Erica
    -0.06
    jspx
    -0.06
    postalcode
    -0.06
    bedo
    -0.06
    POSITIVE LOGITS
     ок
    0.07
     вов
    0.06
     homeowners
    0.06
    oom
    0.06
    ńst
    0.06
     Komm
    0.06
    мати
    0.06
    лад
    0.06
    чили
    0.06
     stagnant
    0.06
    Act Density 0.082%

    No Known Activations