© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 2191
    Prev
    Next
    INDEX
    Explanations

    SQL constraints

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

    The neuron fires on words used in technical schema or data-definition contexts (e.g. default, define, name, type, rules).

    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
    expédition
    0.65
    asparagus
    0.64
    snorkel
    0.63
     مقام
    0.62
    睡
    0.62
    ◘
    0.61
    nega
    0.59
    ぢ
    0.59
     линию
    0.57
    झ्
    0.57
    POSITIVE LOGITS
     CONSTRAINT
    1.27
     unique
    1.18
     UNIQUE
    1.15
     Unique
    1.13
     constraint
    1.09
     FOREIGN
    1.09
    CONSTRAINT
    1.09
     PRIMARY
    1.04
     Constraint
    1.04
     foreign
    1.02
    Activations Density 0.019%

    No Known Activations