© 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. 64764
    Prev
    Next
    INDEX
    Explanations

    place names AND regions

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

    The neuron detects proper place names (geographic locations such as towns, stations, regions, 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
     sesu
    -1.81
     jepang
    -1.64
     kijang
    -1.59
     cewek
    -1.56
     boneka
    -1.55
    
    -1.52
    ваме
    -1.52
     uhr
    -1.50
     sebag
    -1.48
    神
    -1.48
    POSITIVE LOGITS
     >
    1.54
    </h2>
    1.52
    咐
    1.49
    OpenGL
    1.44
     Any
    1.38
    zości
    1.35
    *,
    1.34
    ən
    1.34
    ?,
    1.34
     To
    1.34
    Activations Density 0.095%

    No Known Activations