© 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. 16117
    Prev
    Next
    INDEX
    Explanations

    STEM and LGBTQ+

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

    The neuron primarily detects uppercase-letter acronyms for sexual‐orientation and gender‐identity groups (e.g. LGBT, GLBT, LGBTQ, STEM, etc.).

    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
    After
    -1.41
     or
    -1.39
     Biaya
    -1.38
    uniqlo
    -1.35
     sentada
    -1.33
    して下さい
    -1.32
     hidratante
    -1.31
     asesino
    -1.30
     easy
    -1.30
     after
    -1.30
    POSITIVE LOGITS
     エン
    1.30
    developed
    1.27
    without
    1.23
     intermin
    1.20
    yeah
    1.20
     condemns
    1.20
    ViewChild
    1.19
    che
    1.18
    me
    1.17
    </strong>
    1.16
    Activations Density 0.020%

    No Known Activations