© 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. 9161
    Prev
    Next
    INDEX
    Explanations

    code syntax and structure

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

    This neuron detects the start of code blocks by firing strongly on block‐opening braces “{” and their following indentation (series of spaces/tabs).

    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
    ༵
    -1.70
     bebek
    -1.66
     parfü
    -1.65
    Soundtrack
    -1.65
     sitten
    -1.64
    좆
    -1.62
     parfüm
    -1.61
    suche
    -1.61
    itieren
    -1.59
     炖
    -1.59
    POSITIVE LOGITS
    sp
    1.73
    n
    1.72
    b
    1.68
    www
    1.64
    In
    1.62
    p
    1.58
    ??
    1.56
    mo
    1.55
    u
    1.55
    f
    1.53
    Activations Density 0.016%

    No Known Activations