© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38335
    Prev
    Next
    INDEX
    Explanations

    programming keywords and constructs

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

    The neuron detects the main topic terms in a question’s title—i.e. the core technical keywords naming the specific programming concept or problem.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     eventuell
    -1.30
    𝗺
    -1.27
     Carthag
    -1.27
     Appetit
    -1.25
     Ekonomi
    -1.25
     uniques
    -1.23
    恚
    -1.23
    Zgod
    -1.21
    ihr
    -1.20
    Configura
    -1.18
    POSITIVE LOGITS
     like
    1.51
     what
    1.45
     can
    1.40
    1.38
     filós
    1.32
     called
    1.31
     now
    1.24
    ↵↵
    1.23
     how
    1.23
     different
    1.22
    Activations Density 0.406%
    No Known Activations