© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 72520
    Prev
    Next
    INDEX
    Explanations

    government departments or officials

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

    This neuron responds to formal, bureaucratic or institutional language—especially department/organization names and related terms.

    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
     сахара
    -1.34
    ଼
    -1.23
    ׇ
    -1.19
     десер
    -1.18
     galinha
    -1.15
    有些
    -1.14
    као
    -1.14
     партнер
    -1.13
    apte
    -1.13
     ингреди
    -1.13
    POSITIVE LOGITS
    ַּ
    1.55
    ְּ
    1.38
    ָּ
    1.38
    ֶּ
    1.38
     who
    1.38
     office
    1.31
    ִּ
    1.27
     regarding
    1.23
     then
    1.23
     directly
    1.23
    Activations Density 0.021%

    No Known Activations