© 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. 34-GEMMASCOPE-RES-131K
    7. 110874
    Prev
    Next
    INDEX
    Explanations

    Chinese provinces and other provinces

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

    The neuron fires on toponyms—that is, geographic place names (especially provinces, regions, and cities).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     lógico
    -0.95
    nations
    -0.94
    VARIANT
    -0.93
    电子邮件
    -0.91
     the
    -0.88
     adapted
    -0.86
     fornire
    -0.85
     your
    -0.85
    若い
    -0.84
     mediana
    -0.84
    POSITIVE LOGITS
     province
    2.91
     Province
    2.72
    Province
    1.80
    province
    1.77
    省
    1.73
     state
    1.45
     provincial
    1.42
     Governor
    1.41
    州
    1.40
     State
    1.38
    Activations Density 0.072%

    No Known Activations