© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 14807
    Prev
    Next
    INDEX
    Explanations

    ethnic groups and indigenous peoples

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

    The neuron activates on terms referring to human groups or populations—e.g. ethnicities, indigenous peoples, tribes, and the word “people.”

    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
     time
    -1.46
     costs
    -1.42
     test
    -1.38
    とお
    -1.37
     then
    -1.34
     a
    -1.34
     your
    -1.29
     setup
    -1.27
     оружи
    -1.26
     persönlich
    -1.26
    POSITIVE LOGITS
     they
    1.81
    他们
    1.48
    重置
    1.48
    обзор
    1.48
     themselves
    1.48
     вони
    1.44
    wherein
    1.42
    📜
    1.40
     of
    1.38
    他們
    1.38
    Activations Density 0.071%

    No Known Activations