© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 110581
    Prev
    Next
    INDEX
    Explanations

    html head tags

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

    HTML opening tags and metadata elements in document headers.

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

    The neuron detects HTML document boilerplate, especially tags and attributes in the `<head>` section (e.g. `<html lang="…">`, `<meta charset=…>`, `<title>`, `<script>`).

    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
     by
    -1.61
     from
    -1.32
     give
    -1.27
     they
    -1.21
    A
    -1.19
    ۴
    -1.18
     allow
    -1.17
     “
    -1.16
     naprawdę
    -1.16
    faden
    -1.15
    POSITIVE LOGITS
    <bos>
    1.37
    ➩
    1.35
    cided
    1.33
    essary
    1.30
    Ⓡ
    1.23
     zei
    1.22
    ProjectName
    1.21
    ⇴
    1.21
    khov
    1.21
    utado
    1.21
    Activations Density 0.099%

    No Known Activations