© 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. 107529
    Prev
    Next
    INDEX
    Explanations

    United States Court of Appeals; Olympic events

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

    This neuron fires on prominent capitalized category or entity labels (e.g. “United,” “States,” “Women,” “Olympic,” “Sports”) typically found in headings or Wikipedia category tags.

    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
    头上
    -1.04
     their
    -0.96
     lourdes
    -0.93
     turista
    -0.91
     other
    -0.88
     gümüş
    -0.88
     karbon
    -0.88
     gotas
    -0.88
     factores
    -0.85
     museo
    -0.85
    POSITIVE LOGITS
     todavía
    0.92
    我一直
    0.88
     малыша
    0.84
    ugt
    0.83
    ceptives
    0.82
    因為
    0.80
     Secretaries
    0.80
    aliyun
    0.79
    ônio
    0.79
    كيف
    0.78
    Activations Density 0.011%

    No Known Activations