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

    Neuronpedia

    INDEX
    Explanations

    unread

    np_max-act · gemini-2.0-flash

    distinctive proper nouns and named entities (uncommon names of people, places, organizations, brands, or acronyms).

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
     dung
    -0.06
     tiên
    -0.06
    .http
    -0.05
     defensively
    -0.05
    Regions
    -0.05
     ł
    -0.05
    �
    -0.05
    fidf
    -0.05
    Mil
    -0.05
     Juni
    -0.05
    POSITIVE LOGITS
    According
    0.07
     paso
    0.07
    .extract
    0.07
     yapım
    0.07
     architect
    0.07
    .loaded
    0.07
    การแข
    0.06
    Batman
    0.06
     Royale
    0.06
    ؤول
    0.06
    Act Density 4.873%

    No Known Activations