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

    Neuronpedia

    INDEX
    Explanations

    SS

    np_max-act · gemini-2.0-flash

    The neuron fires on title‐ or heading‐style words—i.e. capitalized or all‐caps tokens marking section headings, product names, or other headline‐like labels.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     DD
    -0.07
     starving
    -0.07
     Fe
    -0.07
     immature
    -0.06
     inventor
    -0.06
     launching
    -0.06
     GER
    -0.06
     DEN
    -0.06
    ิกายน
    -0.06
     ku
    -0.06
    POSITIVE LOGITS
     Institution
    0.07
    inspect
    0.07
    Meteor
    0.06
    unar
    0.06
    Monster
    0.06
    STS
    0.06
    HG
    0.06
    Sem
    0.06
     usize
    0.06
    ську
    0.06
    Act Density 0.055%

    No Known Activations