© 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. 9081
    Prev
    Next
    INDEX
    Explanations

    code file paths and names

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

    The neuron fires on distinctive proper‐name tokens—especially library, framework, project, or module names and acronyms.

    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
     którzy
    -1.45
    nă
    -1.35
    temos
    -1.35
    เพราะ
    -1.22
     mūsų
    -1.21
    ケート
    -1.20
    ‪.‬‬
    -1.20
    หลาย
    -1.19
    เรา
    -1.19
    Similarly
    -1.18
    POSITIVE LOGITS
    even
    1.23
     you
    1.22
     любого
    1.17
    breos
    1.16
     events
    1.15
    ственности
    1.14
     Klage
    1.13
    了什么
    1.11
    -
    1.10
    grade
    1.09
    Activations Density 0.139%

    No Known Activations