© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 14947
    Prev
    Next
    INDEX
    Explanations

    sentence concluding punctuation

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

    This neuron detects section or article headings—i.e. the prominent first word (often emphasized or in all-caps) that titles a new piece or segment.

    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
     '
    -1.48
    ᴘ
    -1.40
     but
    -1.37
     remplace
    -1.34
    goes
    -1.32
     sœurs
    -1.32
    cellona
    -1.27
    Showing
    -1.26
    ,
    -1.26
    Was
    -1.25
    POSITIVE LOGITS
    ↵↵
    1.44
     speziellen
    1.37
    ためには
    1.35
     Because
    1.34
     Unfortunately
    1.34
    DESCRIPCIÓN
    1.33
     Sometimes
    1.30
     Кубка
    1.27
     which
    1.27
    èlement
    1.27
    Activations Density 0.089%

    No Known Activations