© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 113178
    Prev
    Next
    INDEX
    Explanations

    taking office, president, power

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

    The neuron fires on verbs and prepositional phrases that describe assuming or taking on a role or ownership (e.g. “taking office,” “took over,” “coming to power,” “inherited”).

    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
    崀
    -1.41
    ouncement
    -1.36
    
    -1.35
    粜
    -1.29
     ATTORNEY
    -1.28
    IContainer
    -1.27
    somebody
    -1.27
     Assess
    -1.27
    hline
    -1.25
     splendid
    -1.25
    POSITIVE LOGITS
     new
    1.44
     as
    1.41
     específicas
    1.25
    より
    1.25
     Shortly
    1.24
    There
    1.23
     föränd
    1.23
     aislada
    1.20
    Since
    1.15
     Perhaps
    1.15
    Activations Density 0.052%

    No Known Activations