© 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. 25830
    Prev
    Next
    INDEX
    Explanations

    gay, queer, and drag culture

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

    The neuron activates on personal references to people—proper names and gendered pronouns (e.g. “Patti,” “she,” “her,” “wife”).

    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
     مواليد
    -1.02
     gezet
    -0.87
    ')
    -0.85
     sorption
    -0.84
     system
    -0.83
     increases
    -0.82
    firstname
    -0.81
     sowas
    -0.81
     lomb
    -0.81
    erweise
    -0.81
    POSITIVE LOGITS
     gay
    1.77
     homosexual
    1.54
     queer
    1.45
     gays
    1.43
     lesbian
    1.42
     гей
    1.41
     LGBTQ
    1.33
    🏳
    1.31
     transgender
    1.31
     LGBT
    1.27
    Activations Density 0.064%

    No Known Activations