© 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. 34-GEMMASCOPE-RES-131K
    7. 108164
    Prev
    Next
    INDEX
    Explanations

    demographic and identity categories

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

    The neuron fires on enumerations of demographic or protected‐class categories (e.g. “racial/ethnic and socioeconomic status”), detecting those slash‐ or “and”‐separated lists of group attributes.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     Racism
    -1.40
     Racial
    -1.34
     racial
    -1.29
     racially
    -1.27
    种族
    -1.25
     racism
    -1.20
    Racism
    -1.11
     racist
    -1.09
     Ethnicity
    -1.02
     Demographics
    -1.00
    POSITIVE LOGITS
     disability
    1.14
     age
    1.14
     disabled
    1.03
     gender
    1.01
     religion
    1.00
     LGBTQ
    0.97
     income
    0.96
     rural
    0.95
    rural
    0.94
     sexual
    0.93
    Activations Density 0.028%

    No Known Activations