© 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. 10-GEMMASCOPE-RES-131K
    7. 111316
    Prev
    Next
    INDEX
    Explanations

    streets, transit, and locations

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

    The main thing this neuron does is find specific geographic place names and address components (e.g., city names, streets, avenues).

    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
    :--
    -1.26
     ajuta
    -1.24
     nicer
    -1.16
     erweitert
    -1.14
     città
    -1.13
     nevera
    -1.12
    .--
    -1.11
    très
    -1.10
     motivasi
    -1.10
     gelöst
    -1.09
    POSITIVE LOGITS
     is
    1.28
    лия
    1.27
     [
    1.24
     include
    1.22
     whereas
    1.20
    kleber
    1.10
    Sons
    1.09
    udahkan
    1.09
     validés
    1.07
    Additionally
    1.06
    Activations Density 0.089%

    No Known Activations