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

    Neuronpedia

    INDEX
    Explanations

    t

    np_max-act · gemini-2.0-flash

    The neuron fires on domain-specific technical jargon—multi-syllable or specialized terms (e.g. drug names, sports positions, camera/lens acronyms, telecom terminology) rather than common words.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    她们
    -0.06
    ของค
    -0.06
    لق
    -0.06
    .setSelection
    -0.06
     spreads
    -0.06
    just
    -0.06
    _wo
    -0.06
    neutral
    -0.06
    ้าน
    -0.06
     fading
    -0.06
    POSITIVE LOGITS
    (cancel
    0.08
    ��
    0.07
     str
    0.07
    .prev
    0.07
     antibiot
    0.06
    otten
    0.06
    宋
    0.06
     Curso
    0.06
    ิย
    0.06
    (Thread
    0.06
    Act Density 0.110%

    No Known Activations