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

    Neuronpedia

    INDEX
    Explanations

    Quotation marks

    np_max-act · gemini-2.0-flash

    The neuron fires on short alphanumeric code tokens—e.g. MIME extensions, ISO language tags, currency codes—so it’s looking for standardized identifier abbreviations.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Than
    -0.07
    URRENT
    -0.07
    PMENT
    -0.07
    пор
    -0.06
     Policies
    -0.06
     payment
    -0.06
    _gap
    -0.06
     laz
    -0.06
    و
    -0.06
     colum
    -0.06
    POSITIVE LOGITS
    -->
    0.07
     *>
    0.06
    )!↵
    0.06
     laboratory
    0.06
    /mod
    0.06
     joueur
    0.06
    -.
    0.06
    _Execute
    0.06
    ्रद
    0.06
    运行
    0.06
    Act Density 0.003%

    No Known Activations