© 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. 34-GEMMASCOPE-RES-131K
    7. 33137
    Prev
    Next
    INDEX
    Explanations

    historical town buildings

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

    This neuron fires on structured metadata entries—like dates, addresses, architect names, and other infobox‐style fields—in a building’s info box.

    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
    typically
    -0.96
     gese
    -0.91
     verspre
    -0.88
     vivent
    -0.87
     afirmou
    -0.85
    Faça
    -0.85
     gewo
    -0.85
     poff
    -0.85
     münchen
    -0.84
     lachen
    -0.83
    POSITIVE LOGITS
     livery
    1.14
     hotel
    1.14
     county
    1.12
     courthouse
    1.07
     saloon
    1.05
     township
    1.03
     buggy
    1.02
     Hotel
    0.99
     prairie
    0.97
     blacksmith
    0.95
    Activations Density 0.047%

    No Known Activations