© 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. 80848
    Prev
    Next
    INDEX
    Explanations

    devotee, Aussie, Seneca, adept, Cuckoo, Confucius, Sufi

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

    The neuron fires on words that label or describe people as members of a group (e.g. demonyms like “Aussie,” place-based group names like “Seneca,” and descriptors of devotees or the devout).

    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
     otras
    -3.17
    蠖
    -2.86
     другие
    -2.67
    л
    -2.48
     renowned
    -2.39
     propios
    -2.33
     remnants
    -2.30
     occasional
    -2.27
     reputable
    -2.25
     asamblea
    -2.23
    POSITIVE LOGITS
    there
    2.94
    </td>
    2.89
     questionSearch
    2.75
    '
    2.44
    骓
    2.38
    
    2.38
     cannes
    2.38
    Despite
    2.33
     rita
    2.31
     toul
    2.31
    Activations Density 0.004%

    No Known Activations