© 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. 10-GEMMASCOPE-RES-131K
    7. 12640
    Prev
    Next
    INDEX
    Explanations

    code syntax

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

    This neuron detects programming‐code tokens (e.g. variable names, function calls/definitions, loops, and other code syntax) within the text.

    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
     a
    -1.42
     リング
    -1.29
    eres
    -1.28
     Η
    -1.26
     indeed
    -1.23
     تصریح
    -1.20
     Käufer
    -1.18
    ské
    -1.14
     NOT
    -1.12
    pasang
    -1.12
    POSITIVE LOGITS
    állítás
    1.58
    넋
    1.55
     cuyas
    1.49
    Since
    1.46
    懨
    1.44
    Soorten
    1.43
     whose
    1.42
     gebou
    1.41
     Tenemos
    1.40
     teniendo
    1.38
    Activations Density 0.087%

    No Known Activations