© 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. 86832
    Prev
    Next
    INDEX
    Explanations

    punctuation and numbers in code

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

    The neuron fires on punctuation separators—especially commas and semicolons—used to delimit values (e.g. in code, raw data arrays, or formatted numeric lists).

    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
     trajets
    -1.43
     ランプ
    -1.41
     réessayer
    -1.40
    隐隐
    -1.38
     entret
    -1.38
     OCTOBER
    -1.38
    other
    -1.37
    .
    -1.36
     achats
    -1.36
     visiter
    -1.34
    POSITIVE LOGITS
     to
    2.25
    GRAFÍA
    1.45
     more
    1.42
    Those
    1.40
     However
    1.37
    各式
    1.36
     Numerous
    1.36
     варіан
    1.34
     inconsistencies
    1.30
     various
    1.29
    Activations Density 0.011%

    No Known Activations