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

    Neuronpedia

    INDEX
    Explanations

    This neuron seems to activate on multiple different kinds of text including: citations starting with "L.", mathematical/scientific notation with expressions, and place names containing "park." I am unable to create a concise description of the neuron's function

    oai_token-act-pair · gemini-2.0-flash

    Coding/legal/math text

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     [*]
    -0.56
     censiti
    -0.52
    DeleteBehavior
    -0.51
     snippetHide
    -0.50
     копия
    -0.50
     Wikimédia
    -0.48
     Nema
    -0.48
     protoimpl
    -0.48
    propTypes
    -0.48
    errHandler
    -0.48
    POSITIVE LOGITS
    <bos>
    1.97
    section
    0.67
    </table>
    0.54
    //
    0.52
     referenties
    0.50
     مواليد
    0.50
    Today
    0.49
    '
    0.48
     vastaan
    0.47
    Geplaatst
    0.47
    Act Density 0.356%

    No Known Activations