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

    Neuronpedia

    INDEX
    Explanations

    file paths/namespaces

    np_max-act · gemini-2.0-flash

    The neuron flags the dot-qualified parts of multi‐segment identifiers (e.g. the “.” and namespace/component tokens in `using` and `namespace` statements).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     commenting
    -0.07
     collider
    -0.07
    เอ
    -0.07
    keeper
    -0.07
    (delegate
    -0.07
     Formula
    -0.07
     keynote
    -0.06
    Template
    -0.06
    Autom
    -0.06
    Hamilton
    -0.06
    POSITIVE LOGITS
     SCM
    0.07
     العن
    0.06
    gment
    0.06
    总
    0.06
     قص
    0.06
    적
    0.06
     erle
    0.06
     Affero
    0.06
     Phillies
    0.06
     overst
    0.06
    Act Density 0.011%

    No Known Activations