© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 65465
    Prev
    Next
    INDEX
    Explanations

    description: "List"

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

    The neuron is triggered by presentation‐style metadata such as slide or page numbers (and the surrounding punctuation or quotes) in a document.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     consulte
    -1.32
    fordshire
    -1.21
     halte
    -1.20
     rente
    -1.20
     lignes
    -1.17
     ralenti
    -1.16
     interag
    -1.14
     handels
    -1.14
    trying
    -1.13
     tene
    -1.13
    POSITIVE LOGITS
     our
    1.35
     how
    1.27
     what
    1.22
     information
    1.20
     new
    1.18
     What
    1.17
     How
    1.14
     ALL
    1.14
     and
    1.13
    ,"
    1.11
    Activations Density 0.003%

    No Known Activations