© 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. 10-GEMMASCOPE-RES-131K
    7. 43427
    Prev
    Next
    INDEX
    Explanations

    specifying kind, type, variety, sort

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

    The neuron detects generic classifier nouns (like “kind,” “sort,” or “type”) used to introduce a category (e.g. “the kind of…,” “the sort of…,” “the type of…”).

    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
     and
    -2.23
     about
    -2.06
    boneca
    -1.73
     different
    -1.67
    先は
    -1.56
     Dienste
    -1.55
    事項
    -1.52
     as
    -1.52
     طف
    -1.46
     Unterschied
    -1.45
    POSITIVE LOGITS
    ття
    1.48
    vise
    1.48
    вать
    1.45
     artig
    1.41
    he
    1.41
    ters
    1.40
     rigide
    1.40
     firent
    1.39
     laiss
    1.39
    te
    1.39
    Activations Density 0.034%

    No Known Activations