© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 99444
    Prev
    Next
    INDEX
    Explanations

    repository, service, controller, factory

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

    This neuron detects Java class- or interface-type names for backend components—specifically those ending in Dao, Mapper, Service, or Repository.

    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
    -1.90
     before
    -1.79
     not
    -1.78
     two
    -1.74
     while
    -1.73
     or
    -1.73
     through
    -1.67
     when
    -1.64
     about
    -1.59
     after
    -1.59
    POSITIVE LOGITS
    ︲
    1.50
    significantly
    1.42
     Tenemos
    1.42
    considering
    1.42
     เรา
    1.41
    țiune
    1.36
     typelib
    1.35
     głów
    1.35
    ఙ
    1.35
     które
    1.34
    Activations Density 0.017%

    No Known Activations