© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 39403
    Prev
    Next
    INDEX
    Explanations

    rural surroundings or areas

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

    The neuron fires on words naming types of land use or terrain (e.g. “residential,” “commercial,” “rural,” “fields,” “wooded,” “industrial”).

    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
     it
    -1.48
     in
    -1.41
     then
    -1.38
     such
    -1.37
     будь
    -1.34
     nonché
    -1.32
    $
    -1.28
     for
    -1.27
    us
    -1.25
     given
    -1.20
    POSITIVE LOGITS
     mojej
    1.48
     Soit
    1.41
    maktadır
    1.41
     Họ
    1.38
     bestens
    1.38
     Entrega
    1.38
    
    1.37
     bascul
    1.36
    在我的
    1.35
     pêche
    1.34
    Activations Density 0.066%

    No Known Activations