© 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. 12095
    Prev
    Next
    INDEX
    Explanations

    as children, Saturday afternoons, less harmful, US soil

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

    The neuron consistently lights up on common function words—especially high‐frequency prepositions and particles such as “in,” “to,” “as,” “on,” 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_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
     monies
    -1.02
    他已经
    -0.93
     allerede
    -0.93
    /
    -0.93
     प्रोड
    -0.93
    而不
    -0.92
     males
    -0.91
    然的
    -0.90
     maiores
    -0.90
    کنند
    -0.89
    POSITIVE LOGITS
     their
    1.19
     dizaines
    1.15
     emporter
    1.09
     künf
    1.09
     ‌
    1.09
    พวก
    1.08
     équipes
    1.05
    江苏
    1.04
     échanges
    1.03
    ẤT
    1.02
    Activations Density 0.126%

    No Known Activations