© 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. 71106
    Prev
    Next
    INDEX
    Explanations

    code delimiters and structure

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

    The neuron detects programming‐language syntax elements (parentheses, brackets, braces, and keywords) that mark code structures.

    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
     in
    -1.55
     four
    -1.41
    ;
    -1.40
     five
    -1.39
     six
    -1.38
    iriman
    -1.38
     before
    -1.36
     Initially
    -1.35
    曜日
    -1.34
    ㄹ
    -1.31
    POSITIVE LOGITS
    の場合は
    1.63
    ⏑
    1.60
    atação
    1.56
     nieuwe
    1.51
    
    1.48
     réve
    1.48
    )$.
    1.47
    遄
    1.47
    
    1.45
     déliv
    1.44
    Activations Density 0.012%

    No Known Activations