© 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. 22-GEMMASCOPE-RES-131K
    7. 98532
    Prev
    Next
    INDEX
    Explanations

    political membership

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

    The neuron is especially sensitive to the word “member,” as in “member of (an assembly/committee/etc.).”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     bilhões
    -1.39
    ׇ
    -1.36
     jurk
    -1.36
    août
    -1.35
     別注
    -1.34
    　“
    -1.33
    """
    -1.30
     ὅ
    -1.30
    prepareStatement
    -1.29
     zarówno
    -1.29
    POSITIVE LOGITS
     on
    1.68
     election
    1.54
     his
    1.52
     elections
    1.38
     voters
    1.32
     electoral
    1.30
     сказки
    1.29
     in
    1.28
     λα
    1.27
     political
    1.26
    Activations Density 0.195%

    No Known Activations