© 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. 75370
    Prev
    Next
    INDEX
    Explanations

    class, Job, service, cells

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

    programming language-related terms and code elements in technical contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on programming-specific identifiers and keywords—especially class, function, API, and variable names in code.

    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
     الفيديو
    -1.47
     and
    -1.33
     صفحة
    -1.24
     البحث
    -1.20
     mijn
    -1.20
     powodu
    -1.19
    那些
    -1.16
    timg
    -1.12
     jednej
    -1.10
     horloge
    -1.10
    POSITIVE LOGITS
     geogra
    1.29
     relish
    1.25
     بخوانید
    1.24
     capilar
    1.23
     undermine
    1.23
    ついでに
    1.23
     juntas
    1.22
     mesta
    1.20
     lagar
    1.20
     there
    1.19
    Activations Density 0.050%

    No Known Activations