© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 11658
    Prev
    Next
    INDEX
    Explanations

    character types and roles

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

    This neuron detects words naming groups or identities—e.g. ethnicities, religions, or even fantasy races.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Indigenous
    0.66
    受欢迎
    0.59
     journeys
    0.56
    己
    0.55
     Hispanic
    0.54
     людини
    0.54
     Congolese
    0.54
     Jewish
    0.53
    价
    0.53
    外国人
    0.52
    POSITIVE LOGITS
    ország
    0.77
    लैंड
    0.67
    族的
    0.65
     males
    0.63
     infants
    0.63
     Infants
    0.63
    Culture
    0.62
    leder
    0.61
    족
    0.61
    istan
    0.60
    Activations Density 0.237%

    No Known Activations