© 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. 22-GEMMASCOPE-RES-131K
    7. 10992
    Prev
    Next
    INDEX
    Explanations

    document section identifiers

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

    The neuron detects section or subsection identifiers (e.g. “sec” tokens and their numeric IDs) in structured documents.

    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
     from
    -1.21
    тивен
    -1.07
    %;
    -1.07
    anız
    -1.05
    azas
    -1.05
     verschiedenen
    -1.03
    retario
    -1.03
    佉
    -1.01
     процес
    -1.01
    τογραφ
    -0.99
    POSITIVE LOGITS
    したのは
    1.12
    _
    1.12
     polskiej
    1.12
    淪
    1.08
    .
    1.06
    聳
    1.05
    くらいで
    1.04
    っかけ
    1.04
    0
    1.04
    inhos
    0.99
    Activations Density 0.042%

    No Known Activations