© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130840
    Prev
    Next
    INDEX
    Explanations

    The neuron responds to mentions of the “body” or specific body parts (e.g. “ass,” “butt,” “body,” “aching”) across contexts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    body

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     their
    -0.95
     malady
    -0.95
    ッキリ
    -0.94
     won
    -0.92
     enthalten
    -0.91
     microscopy
    -0.89
     whitish
    -0.86
     damaged
    -0.84
     in
    -0.84
     тыс
    -0.83
    POSITIVE LOGITS
     body
    1.48
    全身
    1.38
     Body
    1.38
     BODY
    1.23
    Body
    1.20
     hura
    1.19
     cinto
    1.17
    身
    1.16
    ׇ
    1.14
     kulak
    1.11
    Activations Density 0.064%

    No Known Activations