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

    Neuronpedia

    INDEX
    Explanations

    Additions and continuations

    np_max-act · gemini-2.0-flash

    The neuron is selectively activating on longer, multi-syllable, Latinate (formal/technical) words (e.g. “conducted,” “incorporated,” “exposure”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     đọ
    -0.07
     keypoints
    -0.07
    .interval
    -0.06
    .ArrayAdapter
    -0.06
     items
    -0.06
     SOME
    -0.06
    _Element
    -0.06
     dope
    -0.06
    .setHeader
    -0.06
     network
    -0.06
    POSITIVE LOGITS
    ividad
    0.07
    zipcode
    0.06
    ometrics
    0.06
    اید
    0.06
    Independ
    0.06
     customizable
    0.06
    >↵↵↵↵↵
    0.06
    Reviews
    0.06
    _Metadata
    0.06
     Marketplace
    0.06
    Act Density 0.131%

    No Known Activations