© 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. 34-GEMMASCOPE-RES-131K
    7. 4037
    Prev
    Next
    INDEX
    Explanations

    html title and script tags

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

    This neuron activates on structural metadata markers and section headers (e.g. tokens like “title,” “description,” “xaml,” “usage,” and templating tags such as “<%”), i.e. words that label or begin document sections.

    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
     knew
    -0.73
    ঃ
    -0.72
    境的
    -0.71
     suggesting
    -0.70
    するか
    -0.68
     Tipper
    -0.68
     despite
    -0.68
     whom
    -0.67
    ровано
    -0.67
     cancelled
    -0.66
    POSITIVE LOGITS
    remember
    0.85
    رض
    0.77
     trám
    0.76
    尧
    0.75
     platter
    0.73
    toCharArray
    0.73
    getStart
    0.73
     ดอก
    0.73
    čki
    0.72
    umbria
    0.71
    Activations Density 0.001%

    No Known Activations