© 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. 18966
    Prev
    Next
    INDEX
    Explanations

    questions faced

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

    This neuron fires most strongly on the very first prominent content word of a passage—i.e. the main heading or section-opening keyword.

    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
     that
    -1.59
     of
    -1.54
     miljö
    -1.39
     такого
    -1.35
     león
    -1.30
     suffisamment
    -1.30
     nying
    -1.28
     放題
    -1.27
    timmung
    -1.27
     ブログ
    -1.26
    POSITIVE LOGITS
     and
    1.47
    <blockquote>
    1.29
    ticularly
    1.27
    させて頂
    1.25
    させていただきました
    1.24
    [
    1.23
     herida
    1.21
    нивер
    1.19
     [
    1.18
    osuke
    1.17
    Activations Density 0.088%

    No Known Activations