© 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. 34-GEMMASCOPE-RES-131K
    7. 57652
    Prev
    Next
    INDEX
    Explanations

    gender presentation and identity

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

    The neuron activates on words and phrases that signal gender identity or roles—especially feminine references (e.g. female, feminine, sissy, pregnant, cross-dresser, drag, guy vs. girl).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    errors
    -0.95
     individually
    -0.82
     sorriso
    -0.80
     détach
    -0.77
     citizenship
    -0.76
    mble
    -0.75
     Personalis
    -0.75
    楣
    -0.74
     갈
    -0.74
    ּוֹ
    -0.73
    POSITIVE LOGITS
     feminine
    2.44
     female
    1.92
     gender
    1.72
    female
    1.69
     girly
    1.66
    femin
    1.66
     women
    1.63
     femininity
    1.55
     femin
    1.53
     females
    1.52
    Activations Density 0.044%

    No Known Activations