© 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. 10-GEMMASCOPE-RES-131K
    7. 85487
    Prev
    Next
    INDEX
    Explanations

    "@" followed by domain names/identifiers

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

    This neuron fires on tokens beginning with “@”, i.e. scoped package or module names and similar “@…” identifiers in code or JSON.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    ি
    -2.80
    
    -2.52
    żli
    -2.45
    咈
    -2.44
    釹
    -2.34
    
    -2.28
     perſon
    -2.28
     باسلام
    -2.23
    
    -2.23
    of
    -2.20
    POSITIVE LOGITS
    e
    2.92
    y
    2.86
    o
    2.58
    t
    2.52
    ?
    2.42
    d
    2.34
    —
    2.23
     According
    2.08
    -
    2.08
    {
    2.06
    Activations Density 0.001%

    No Known Activations