© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 93328
    Prev
    Next
    INDEX
    Explanations

    regular expressions

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

    The neuron fires on uses of regular‐expression code—especially imports of the Python “re” module and calls like re.search/re.compile.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    Republic
    -0.71
    ⏲
    -0.70
    Rabbit
    -0.70
     mysqli
    -0.70
    )__
    -0.69
    ificance
    -0.69
     فهم
    -0.69
    extAlignment
    -0.69
     parfum
    -0.68
     cerr
    -0.68
    POSITIVE LOGITS
     pattern
    3.44
    regex
    3.34
     regex
    3.31
     Pattern
    3.17
    pattern
    3.13
    Pattern
    3.13
     patterns
    2.94
    Regex
    2.92
     Regex
    2.78
    regexp
    2.73
    Activations Density 0.320%

    No Known Activations