© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 70829
    Prev
    Next
    INDEX
    Explanations

    package.json or POM files

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

    This neuron activates on metadata field names and key tokens in project or package manifest files (e.g. package.json and Maven POM elements like “package,” “json,” “project,” “modelVersion,” “version,” “artifactId”).

    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
     你
    -1.29
     ();
    -1.29
     Rè
    -1.27
     mentales
    -1.26
    там
    -1.26
    despite
    -1.25
     pernic
    -1.24
     Katalog
    -1.23
    naments
    -1.21
     doctrina
    -1.21
    POSITIVE LOGITS
     about
    1.41
     around
    1.34
     uniquely
    1.28
     good
    1.27
     new
    1.27
     exquisite
    1.20
     greatly
    1.20
     обеспечения
    1.19
     eight
    1.18
    olja
    1.17
    Activations Density 0.110%

    No Known Activations