© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 65239
    Prev
    Next
    INDEX
    Explanations

    proper nouns and names

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is sensitive to rare or unusual tokens—things like uncommon proper names, foreign‐language words (often with diacritics), and outlier terms (e.g. “webp”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Ꮔ
    -3.41
    Ꮈ
    -3.36
    幛
    -3.31
    鬈
    -3.23
    acterísticas
    -3.14
    棪
    -2.78
    釙
    -2.70
    蟶
    -2.69
    撘
    -2.69
     técn
    -2.67
    POSITIVE LOGITS
    a
    3.28
    但
    3.20
    缡
    3.09
    the
    2.92
    2
    2.88
    ).
    2.66
    )$
    2.63
    鮨
    2.58
     --
    2.55
    !!!!!!
    2.52
    Activations Density 0.168%

    No Known Activations