© 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. 124661
    Prev
    Next
    INDEX
    Explanations

    code snippets and markdown

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

    The neuron is chiefly triggered by section headings or metadata/title labels (e.g. HTML/Markdown/JSON keys and titles) 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_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
     ederim
    -1.30
     THEM
    -1.29
    </h2>
    -1.25
     całego
    -1.23
    </h1>
    -1.23
    Fk
    -1.16
    inafter
    -1.13
    JZ
    -1.13
     vált
    -1.12
     kapcsol
    -1.10
    POSITIVE LOGITS
     is
    1.36
     there
    1.28
     golfe
    1.25
     because
    1.24
     unless
    1.20
     we
    1.19
     fianco
    1.17
     one
    1.16
     Lorsque
    1.15
    NDICE
    1.15
    Activations Density 0.186%

    No Known Activations