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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 129916
    Prev
    Next
    INDEX
    Explanations

    first names followed by last names

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

    The neuron activates on tokens that are parts of people’s proper names (first or last names) in news text.

    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
     what
    -1.66
     these
    -1.66
    尤其
    -1.57
     kapcsol
    -1.48
     ۱۱
    -1.46
    ﹠
    -1.46
    Pengertian
    -1.41
     not
    -1.41
    WWW
    -1.34
     ۱۵
    -1.31
    POSITIVE LOGITS
     who
    1.88
    ଶ
    1.64
     use
    1.64
     previously
    1.51
     السياس
    1.49
    </tfoot>
    1.47
    OrWhiteSpace
    1.46
    ಇ
    1.46
    使用
    1.42
     attempt
    1.42
    Activations Density 0.259%

    No Known Activations