© 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. 22-GEMMASCOPE-RES-131K
    7. 130715
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on structural definition tokens in code or markup—e.g. class/namespace/function names and XML tag names used to declare or define entities.

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

    property public attributes

    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
     only
    -0.93
     estrena
    -0.93
     another
    -0.91
     either
    -0.90
    しており
    -0.88
    zsche
    -0.87
    ungsanleitung
    -0.84
    nowned
    -0.84
    nguyen
    -0.82
    kespea
    -0.81
    POSITIVE LOGITS
     everything
    1.31
    包括
    1.27
    各项
    1.18
    everything
    1.15
     EVERYTHING
    1.09
    主な
    1.09
     همه
    1.02
    诸
    1.00
    items
    0.99
     things
    0.95
    Activations Density 0.040%

    No Known Activations