© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 38589
    Prev
    Next
    INDEX
    Explanations

    political speeches and debates

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

    The neuron detects formal parliamentary or meeting-minutes procedural actions and agenda items (e.g. “tabled,” “raised,” “Motion of Thanks,” “report,” “in the chair,” etc.).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     is
    -2.02
    f
    -1.81
    in
    -1.75
     there
    -1.69
     of
    -1.66
    www
    -1.66
    self
    -1.56
    mo
    -1.46
    three
    -1.46
    te
    -1.45
    POSITIVE LOGITS
     który
    1.40
    ก่อน
    1.37
     ",";
    1.33
    kaos
    1.30
    assum
    1.27
    ;
    1.27
    hadap
    1.25
    吳
    1.24
    我們
    1.24
    Groet
    1.23
    Activations Density 0.004%

    No Known Activations