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

    Neuronpedia

    INDEX
    Explanations

    Scientific publications

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on domain-specific technical terms and jargon—e.g. specialized scientific names, acronyms, and proper nouns (like gene or protein names, distribution names, species names)—in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    姓
    -0.08
     включ
    -0.07
    _make
    -0.07
     remar
    -0.07
    vero
    -0.06
    ocial
    -0.06
    leans
    -0.06
    Attribute
    -0.06
    寝
    -0.06
    -posts
    -0.06
    POSITIVE LOGITS
     янва
    0.06
     necesita
    0.06
    _WITH
    0.06
    .favorite
    0.06
     عاشق
    0.06
    ...");↵↵
    0.06
    ​​
    0.06
    ικές
    0.06
     NavLink
    0.06
     důležité
    0.06
    Act Density 0.104%

    No Known Activations