© 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. 75106
    Prev
    Next
    INDEX
    Explanations

    design patterns

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

    The neuron fires on the standalone, capitalized token “Design,” as used (typically as a heading or part of a title) across various contexts.

    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
     figuring
    -2.31
     manten
    -2.16
    犼
    -2.09
    嘬
    -2.09
    鮓
    -2.02
     it
    -2.02
    琑
    -2.02
     inscrição
    -1.98
    壆
    -1.98
    蕻
    -1.96
    POSITIVE LOGITS
    --
    2.69
    {
    2.42
    2
    2.38
    </strong>
    2.28
    ة
    2.23
    了两
    2.17
    4
    2.17
    -
    2.13
    You
    2.08
     vorsichtig
    2.08
    Activations Density 0.009%

    No Known Activations