© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 3418
    Prev
    Next
    INDEX
    Explanations

    fortress, castle, fortified structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ulti
    -0.11
    笫
    -0.09
     ×ijת×ķר
    -0.09
    èĨĢ
    -0.09
    /generated
    -0.09
    åĬ¨çī©
    -0.08
    bourne
    -0.08
    asar
    -0.08
    æĹħç¨ĭ
    -0.08
    æĺ¥å¤ı
    -0.08
    POSITIVE LOGITS
    åĨħçļĦ
    0.09
     destiny
    0.09
    æĢ§å¼º
    0.08
     bre
    0.08
    utton
    0.08
    æĪIJé¾Ļ
    0.08
     setDefaultCloseOperation
    0.08
     Cuá»Ļc
    0.08
    STYPE
    0.08
     Franken
    0.08
    Activations Density 0.025%

    No Known Activations