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

    Neuronpedia

    INDEX
    Explanations

    $(

    np_max-act · gemini-2.0-flash

    The neuron spotlights numeric literals and references (e.g. version numbers, numeric parameters, or numeric constants) in build scripts and makefiles.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ponent
    -0.08
     Likes
    -0.07
    )didReceiveMemoryWarning
    -0.07
     policing
    -0.07
     　 　 　 　
    -0.06
    attice
    -0.06
     Infantry
    -0.06
    .py
    -0.06
    .Com
    -0.06
    (entities
    -0.06
    POSITIVE LOGITS
     "$(
    0.07
    ectar
    0.07
     coeffs
    0.07
    $(
    0.07
     $(
    0.07
     altın
    0.07
    ılır
    0.07
    ($(
    0.07
    ulan
    0.06
    spring
    0.06
    Act Density 0.004%

    No Known Activations