© 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. 10-GEMMASCOPE-RES-131K
    7. 13371
    Prev
    Next
    INDEX
    Explanations

    UI elements and actions

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

    The neuron fires on element names inside angle‐bracket tags (e.g. XML/HTML/SVG tags like `<Canvas>`, `<Path>`, `<Stroke>`).

    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
     to
    -1.44
     \\
    -1.36
    ށ
    -1.26
    vamo
    -1.21
    それとも
    -1.19
    ljeno
    -1.19
    すすめ
    -1.16
     новую
    -1.12
     telah
    -1.11
    unable
    -1.11
    POSITIVE LOGITS
     debe
    1.35
     tuve
    1.34
    ๐
    1.23
     deben
    1.20
     \
    1.20
    &&(
    1.17
     forder
    1.17
     vectores
    1.16
     sugi
    1.14
     panicked
    1.14
    Activations Density 0.171%

    No Known Activations