© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6544
    Prev
    Next
    INDEX
    Explanations

    beautiful peaks

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

    This neuron fires strongly on words expressing aesthetic praise or beauty (e.g., “beautiful,” “beauty,” “landscape,” “stylish”).

    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_24_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
     Hälfte
    1.22
     情報
    1.16
    𝙈
    1.15
    𝐥
    1.07
    𝘢
    1.04
    ProtectedRoute
    1.02
    ﺓ
    1.01
    극
    1.00
    𝗟
    0.99
     Л
    0.99
    POSITIVE LOGITS
    งาม
    1.41
    สดง
    1.41
     aesthetic
    1.30
     aesthetically
    1.16
     livery
    1.16
     spectacles
    1.14
     cosmetic
    1.11
     spectacle
    1.09
     beaut
    1.08
     manicure
    1.08
    Activations Density 0.822%

    No Known Activations