© 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. 66746
    Prev
    Next
    INDEX
    Explanations

    US politicians' state affiliations

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

    The neuron detects partisan affiliation labels with district or state codes (e.g., “R-398,” “R-360,” “AZ 22.5”).

    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
     Actually
    -1.91
     Even
    -1.88
     any
    -1.86
     two
    -1.85
    !
    -1.84
     with
    -1.80
     numerous
    -1.77
     innumerable
    -1.77
     даже
    -1.76
     coveted
    -1.74
    POSITIVE LOGITS
     gewi
    1.71
    镦
    1.69
     CHRISTMAS
    1.69
    螄
    1.66
     ā
    1.66
     kara
    1.65
    kinder
    1.64
    jeux
    1.63
     Cár
    1.63
     COMPLETE
    1.62
    Activations Density 0.008%

    No Known Activations