© 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. 10-GEMMASCOPE-RES-131K
    7. 41893
    Prev
    Next
    INDEX
    Explanations

    capital city

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

    This neuron activates on mentions of “capital city,” specifically spotting when a location is described as a capital.

    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
    ᯾
    -2.48
    朏
    -2.45
    鲮
    -2.45
    ︲
    -2.44
    踌
    -2.39
    蛲
    -2.39
    ౭
    -2.38
    咴
    -2.34
    
    -2.28
     reklama
    -2.25
    POSITIVE LOGITS
    m
    3.20
     This
    3.16
    c
    3.13
    y
    3.06
    o
    3.06
    r
    2.88
    k
    2.84
    n
    2.67
    t
    2.66
    e
    2.59
    Activations Density 0.088%

    No Known Activations