© 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. 10-GEMMASCOPE-RES-131K
    7. 73405
    Prev
    Next
    INDEX
    Explanations

    China and Beijing

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

    The neuron fires on mentions of China (and related proper names like Beijing) across different languages.

    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
    𓆏
    -1.55
    Seperti
    -1.54
     evaluates
    -1.45
     kämp
    -1.45
    ྄
    -1.45
    
    -1.45
    Ꮚ
    -1.42
     coba
    -1.36
    蕷
    -1.36
     ditem
    -1.34
    POSITIVE LOGITS
    </h5>
    2.06
     now
    1.63
    </h3>
    1.62
     is
    1.55
    」、「
    1.54
     only
    1.48
     China
    1.47
    _
    1.35
    /**
    1.30
     huvud
    1.25
    Activations Density 0.169%

    No Known Activations