© 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. 47959
    Prev
    Next
    INDEX
    Explanations

    software names and programming concepts

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

    The neuron flags technical or domain-specific tokens—things like command-line tools, acronyms, file-path segments, API names, or scientific shorthand.

    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
     for
    -1.59
    avaient
    -1.39
     their
    -1.31
    CANADA
    -1.27
     each
    -1.27
    hawaii
    -1.23
     Melhor
    -1.19
    一定的
    -1.18
    thinkable
    -1.16
    了些
    -1.16
    POSITIVE LOGITS
     externas
    1.54
     gratuitas
    1.52
     will
    1.52
     је
    1.45
     exclusivas
    1.38
    plevel
    1.38
    ā
    1.35
     komende
    1.29
    attare
    1.27
     sẽ
    1.23
    Activations Density 0.153%

    No Known Activations