© 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. 22-GEMMASCOPE-RES-131K
    7. 72537
    Prev
    Next
    INDEX
    Explanations

    chat groups and clubs

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

    The neuron detects occurrences of the word “group” (in any form or casing) in code identifiers and text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     allemand
    -1.14
     frança
    -1.13
     alguns
    -1.11
     çocuklar
    -1.10
    cumin
    -1.09
     considérons
    -1.05
     néer
    -1.04
    镗
    -1.02
    有不
    -1.02
    individ
    -1.02
    POSITIVE LOGITS
     group
    1.48
     membership
    1.27
     created
    1.24
     groups
    1.23
     leader
    1.16
     their
    1.16
     memberships
    1.16
     this
    1.10
     formed
    1.07
     aperçu
    1.07
    Activations Density 0.026%

    No Known Activations