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

    Neuronpedia

    INDEX
    Explanations

    Retrospective studies and surveys

    np_max-act · gemini-2.0-flash

    The neuron activates on numerical data tokens—especially decimal numbers and statistical measurements.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    /dialog
    -0.07
    csrf
    -0.07
    List
    -0.07
    INA
    -0.07
    ina
    -0.06
    ..."↵
    -0.06
    VectorXd
    -0.06
    Supplier
    -0.06
    Sex
    -0.06
    unn
    -0.06
    POSITIVE LOGITS
    住宅
    0.06
    removeAttr
    0.06
     travels
    0.06
     COLL
    0.06
    entiful
    0.06
     النه
    0.06
     initState
    0.06
     έχουν
    0.06
     anglais
    0.06
     धर
    0.06
    Act Density 0.067%

    No Known Activations