© 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. 34-GEMMASCOPE-RES-131K
    7. 45365
    Prev
    Next
    INDEX
    Explanations

    ) as

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

    This neuron fires on SQL query tokens—things like function names, column or table identifiers, literals, and punctuation used in SQL statements.

    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
     каждый
    -0.88
     наших
    -0.85
    ternut
    -0.85
     sendok
    -0.82
    を行い
    -0.82
     каждой
    -0.81
    where
    -0.81
     Nuestro
    -0.80
    いら
    -0.80
     móc
    -0.80
    POSITIVE LOGITS
     as
    2.97
     AS
    1.20
    Vendredi
    1.11
    Dimanche
    0.96
    TOMS
    0.96
     sowie
    0.95
    一样
    0.94
     \"
    0.90
    以及
    0.89
    Жизнь
    0.89
    Activations Density 0.032%

    No Known Activations