© 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. 63111
    Prev
    Next
    INDEX
    Explanations

    catalysis

    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
    åĿIJçĿĢ
    -0.11
    强èĢħ
    -0.09
    éªģ
    -0.09
    é¨İ
    -0.09
    éªij
    -0.09
    ç§ģèIJ¥
    -0.08
    è´´åIJ§
    -0.08
    éĿĸ
    -0.08
     Lyons
    -0.08
    è®®
    -0.08
    POSITIVE LOGITS
    éĢ¢
    0.10
     ведÑĮ
    0.10
     substrate
    0.09
     stores
    0.09
    ä¸ĵåijĺ
    0.08
    ä¸ŃåĽ½çī¹èī²
    0.08
    åĤ¨èĥ½
    0.08
    æıĨ
    0.08
     substr
    0.08
    adoras
    0.08
    Activations Density 1.087%

    No Known Activations