© 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. 102173
    Prev
    Next
    INDEX
    Explanations

    Hampshire, Southampton, Portsmouth, Virginia

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

    The neuron fires on proper nouns denoting geographic places (towns, regions, districts, protected areas, 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_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
     isEnabled
    -1.41
     jelas
    -1.40
     is
    -1.38
    师傅
    -1.37
     turísticos
    -1.35
     metálica
    -1.34
     físicos
    -1.34
    
    -1.29
    "--
    -1.25
     estudian
    -1.25
    POSITIVE LOGITS
    ၞ
    1.77
    1
    1.77
    떳
    1.59
    getUrl
    1.57
    0
    1.48
    elements
    1.41
    t
    1.39
    setUrl
    1.38
    Strengths
    1.38
    Ο
    1.38
    Activations Density 0.428%

    No Known Activations