© 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. 34246
    Prev
    Next
    INDEX
    Explanations

    users, people, customers, person

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

    This neuron fires on words naming the primary actors or entities (people or other key “agents”) under discussion (e.g. caregivers, woman, user, customers, people, objects).

    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
    噚
    -1.20
    🌺
    -1.14
    のお客様
    -1.10
    犹如
    -1.02
     people
    -1.00
     or
    -1.00
    ができます
    -0.97
    犧
    -0.97
    名が
    -0.96
    UALLY
    -0.96
    POSITIVE LOGITS
    their
    1.15
    加減
    1.13
     is
    1.12
     getPassword
    1.06
    during
    1.02
     AppModule
    1.01
     cuando
    1.00
     rootReducer
    1.00
     isInitialized
    0.99
     durante
    0.98
    Activations Density 0.231%

    No Known Activations