© 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. 11053
    Prev
    Next
    INDEX
    Explanations

    Steve Beshear's, Trent Richardson, Barack Obama

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

    This neuron detects multi-letter capitalized named entities—especially organization names, band names, sports teams, acronyms, and other proper nouns.

    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
     as
    -1.66
     then
    -1.66
     two
    -1.62
     more
    -1.55
     of
    -1.40
     with
    -1.38
     immediately
    -1.34
     about
    -1.33
     and
    -1.32
     both
    -1.31
    POSITIVE LOGITS
     espé
    1.48
     rense
    1.47
     remé
    1.41
    姈
    1.39
     jezelf
    1.38
    eriam
    1.38
    久し
    1.37
     poudre
    1.35
    嗳
    1.34
    acyjnych
    1.33
    Activations Density 0.925%

    No Known Activations