© 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. 22-GEMMASCOPE-RES-131K
    7. 128490
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on closing curly‐brace tokens ("}") in code blocks.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    code delimiters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     to
    -3.00
     on
    -2.44
     in
    -2.27
     one
    -2.08
     with
    -1.93
     as
    -1.86
     from
    -1.82
     by
    -1.70
     at
    -1.66
    ,
    -1.63
    POSITIVE LOGITS
    Sonntag
    1.60
     taką
    1.58
    Zitat
    1.55
    Groet
    1.54
     hés
    1.49
    푉
    1.46
     guapo
    1.42
    colgante
    1.42
     précip
    1.42
    disfraz
    1.42
    Activations Density 0.013%

    No Known Activations