© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7782
    Prev
    Next
    INDEX
    Explanations

    sql keywords and function names

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

    This neuron activates on SQL reserved words and built-in function names (e.g. SELECT, ORDER BY, CASE, DECODE, CONCAT) in uppercase.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    +"
    0.63
    ";
    0.62
    RET
    0.58
    --;
    0.57
    SQLException
    0.57
    );
    0.56
    wC
    0.55
    _
    0.55
    )$:
    0.55
    __
    0.54
    POSITIVE LOGITS
     cast
    0.81
     DISTINCT
    0.80
     concat
    0.80
     distinct
    0.77
     sum
    0.73
     distinctive
    0.73
    TIMESTAMPDIFF
    0.69
    concat
    0.69
     kast
    0.69
     CAST
    0.68
    Activations Density 0.009%

    No Known Activations