© 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. 19045
    Prev
    Next
    INDEX
    Explanations

    city names

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

    The neuron detects mentions of geographic place names (cities, regions, or other locations).

    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
     that
    -2.50
     this
    -2.33
     such
    -2.06
     because
    -2.00
     its
    -1.97
     a
    -1.85
     their
    -1.78
     our
    -1.75
     from
    -1.73
     those
    -1.66
    POSITIVE LOGITS
    却不
    1.43
    对她
    1.41
    佩服
    1.38
     EIGHT
    1.38
    绺
    1.37
    气势
    1.36
     TWENTY
    1.35
    诧异
    1.34
     огромное
    1.34
    bedded
    1.34
    Activations Density 0.181%

    No Known Activations