© 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. 130318
    Prev
    Next
    INDEX
    Explanations

    military, sports, and political figures

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

    The neuron primarily activates on words naming roles, positions, or organized groups (e.g. “starter,” “fleet,” “defenseman,” “troops,” “captain”).

    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
     of
    -2.16
     because
    -2.09
     and
    -2.05
     his
    -1.70
     an
    -1.41
     for
    -1.34
     can
    -1.32
     as
    -1.30
     your
    -1.29
     Акаде
    -1.29
    POSITIVE LOGITS
    wendige
    1.34
     ピンク
    1.28
    orious
    1.27
     poign
    1.18
    кою
    1.18
    Ꮕ
    1.15
     excep
    1.15
     posib
    1.13
     reo
    1.13
     produrre
    1.13
    Activations Density 0.061%

    No Known Activations