© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 54485
    Prev
    Next
    INDEX
    Explanations

    format, font, fills, display, property

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

    The neuron is tuned to spot occurrences of the word “style” (and closely related styling tokens) in code or markup.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ּוֹ
    -1.84
    ִּ
    -1.32
     QtCore
    -1.26
     журнали
    -1.24
    IInterface
    -1.23
    queryInterface
    -1.23
    OrWhiteSpace
    -1.21
     désolés
    -1.20
    Жит
    -1.20
     verze
    -1.20
    POSITIVE LOGITS
    ׇ
    1.49
     piace
    1.33
     chré
    1.30
     on
    1.26
     caffè
    1.20
     merce
    1.19
     sedia
    1.17
    年は
    1.16
    कि
    1.15
     Lampung
    1.15
    Activations Density 0.059%

    No Known Activations