© 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. 10-GEMMASCOPE-RES-131K
    7. 111045
    Prev
    Next
    INDEX
    Explanations

    html tags

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

    HTML tags and markup elements, particularly opening tags like `<script>`, `<meta>`, `<title>`, `<link>`, and `<head>`.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron activates on HTML element tags (e.g. script, meta, title, link, style) appearing inside angle brackets.

    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
    komplet
    -1.23
    Disliked
    -1.18
     internetu
    -1.10
    IGraphics
    -1.09
    怎么回事
    -1.09
    Awe
    -1.09
    čem
    -1.09
    ›
    -1.08
     telefonu
    -1.08
    hláš
    -1.07
    POSITIVE LOGITS
     herbes
    1.63
     such
    1.37
     estadounid
    1.28
    FOR
    1.25
    AND
    1.23
    Such
    1.19
     такую
    1.18
    ⅼ
    1.17
    sch
    1.16
    Here
    1.16
    Activations Density 0.071%

    No Known Activations