© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 115596
    Prev
    Next
    INDEX
    Explanations

    race and social identity

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

    The neuron fires strongly on occurrences of the word “white” (e.g. “White,” “white privilege,” “white power”), i.e. explicit mentions of whiteness.

    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
     white
    -1.09
     whites
    -1.03
     blacks
    -1.03
     negros
    -0.96
     black
    -0.93
     negro
    -0.90
    黑白
    -0.89
     blackened
    -0.88
    white
    -0.84
    Whites
    -0.84
    POSITIVE LOGITS
     people
    1.73
     folks
    1.42
     skinned
    1.16
     skin
    1.11
     America
    1.09
     folk
    1.06
    washing
    1.06
    人
    0.97
    People
    0.97
     children
    0.93
    Activations Density 0.017%

    No Known Activations