© 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. 130852
    Prev
    Next
    INDEX
    Explanations

    This neuron detects standalone numeric tokens that represent indentation or bullet‐list spacing values in structured document markup.

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

    special characters or formatting

    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
    明日は
    -2.31
    さぁ
    -2.31
    の商品
    -2.31
    ごめん
    -2.27
    久し
    -2.23
    あぁ
    -2.19
    えっ
    -2.11
    もうすぐ
    -2.06
    マジで
    -2.06
     refroid
    -2.03
    POSITIVE LOGITS
    </i>
    1.92
    al
    1.88
     have
    1.84
     Also
    1.80
    h
    1.79
     What
    1.77
     In
    1.74
    が
    1.68
     portrayed
    1.66
     are
    1.62
    Activations Density 0.003%

    No Known Activations