© 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. 34-GEMMASCOPE-RES-131K
    7. 92752
    Prev
    Next
    INDEX
    Explanations

    HTML/XML namespaces

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

    This neuron fires on fragments of URLs or XML namespace URIs (e.g. the “http://www.w3.org/…”, numeric, slash, colon, and dot tokens).

    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
     іде
    -0.81
    akka
    -0.78
     Tages
    -0.77
    manni
    -0.77
    jeh
    -0.74
    拋
    -0.74
     العد
    -0.72
     технологі
    -0.72
    sensitive
    -0.71
    olunt
    -0.71
    POSITIVE LOGITS
     SVG
    1.27
    SVG
    1.06
    svg
    1.02
     svg
    0.99
    Illustrator
    0.92
     récon
    0.88
     kabar
    0.86
     illust
    0.86
     Schema
    0.85
    tigt
    0.84
    Activations Density 0.013%

    No Known Activations