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

    Neuronpedia

    INDEX
    Explanations

    )

    np_max-act · gemini-2.0-flash

    The neuron flags uses of recursive directory traversal (e.g. calls to `os.walk` and related path‐walking constructs).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Аф
    -0.07
     analytic
    -0.07
     Spain
    -0.06
    _priority
    -0.06
    JsonValue
    -0.06
    -vector
    -0.06
     маши
    -0.06
    confirm
    -0.06
    (csv
    -0.06
     glyphs
    -0.06
    POSITIVE LOGITS
    duplicate
    0.08
    ingle
    0.07
     elles
    0.06
    terminated
    0.06
    _SelectedIndexChanged
    0.06
    jev
    0.06
    _Checked
    0.06
     jewellery
    0.06
     (£
    0.06
    antages
    0.06
    Act Density 0.002%

    No Known Activations