© 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. 10-GEMMASCOPE-RES-131K
    7. 110602
    Prev
    Next
    INDEX
    Explanations

    punctuation and closing angle brackets

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

    This neuron detects formatting markup symbols (e.g. asterisks for emphasis or HTML/Markdown tags) in the text.

    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
    ۔
    -1.49
    。
    -1.44
    atever
    -1.38
     tiga
    -1.35
     và
    -1.34
    bB
    -1.34
     然
    -1.29
     Jahrhunderts
    -1.28
     zda
    -1.26
    Fh
    -1.25
    POSITIVE LOGITS
    owskiego
    1.51
     taglio
    1.45
     ihres
    1.44
    hline
    1.43
    ią
    1.42
    IUrlHelper
    1.37
     tweaking
    1.37
     porque
    1.36
     aceste
    1.36
    lgari
    1.34
    Activations Density 0.003%

    No Known Activations