© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 108451
    Prev
    Next
    INDEX
    Explanations

    Lua code

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

    The neuron is looking for programming‐language syntax markers—tokens like braces, semicolons, keywords (namespace, using, typeof), etc., that signal code structure.

    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
    nemies
    -0.85
    Maße
    -0.82
    
    -0.77
    Committee
    -0.77
    CONT
    -0.76
     {:
    -0.75
    queryset
    -0.73
    loggedin
    -0.73
    Queryable
    -0.73
     ترك
    -0.73
    POSITIVE LOGITS
     Belong
    0.85
     Lua
    0.85
    Lua
    0.74
     Aufgrund
    0.72
     tổng
    0.71
     hotfix
    0.70
     earthquakes
    0.69
    lua
    0.69
    ത്ര
    0.69
     Bag
    0.68
    Activations Density 0.008%

    No Known Activations