© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 15517
    Prev
    Next
    INDEX
    Explanations

    materials and their applications

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

    The neuron activates strongly on material names in product/specification contexts (e.g., steel, rubber, leather, glass, wood).

    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
    用
    0.78
    嗷
    0.73
    HttpPut
    0.66
     käyttää
    0.66
     Sử
    0.65
     Using
    0.64
    льний
    0.64
    是用
    0.64
    최
    0.64
     використову
    0.63
    POSITIVE LOGITS
     doors
    0.87
     walls
    0.76
     outlines
    0.76
     moldings
    0.75
     стены
    0.75
     facades
    0.75
    ഷ്ട്ര
    0.74
     alloy
    0.74
     panels
    0.72
     surfaces
    0.70
    Activations Density 0.168%

    No Known Activations