© 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. 70869
    Prev
    Next
    INDEX
    Explanations

    code and structured text punctuation

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

    The neuron is detecting semicolon punctuation (end‐of‐statement markers) in code.

    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
    鶿
    -3.59
    琲
    -3.11
    躥
    -3.03
    㩒
    -2.83
    
    -2.75
    嗵
    -2.75
    ⁛
    -2.70
    辚
    -2.69
    𖣔
    -2.69
    襌
    -2.69
    POSITIVE LOGITS
    e
    3.30
    ).
    3.28
     for
    2.88
     if
    2.84
    [
    2.77
     as
    2.70
     because
    2.67
    ."
    2.66
    ";
    2.66
    ';
    2.56
    Activations Density 0.008%

    No Known Activations