© 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. 48693
    Prev
    Next
    INDEX
    Explanations

    elevated levels of

    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
    awks
    -0.09
    æľīéĴ±
    -0.08
    çļĦä¹łæĥ¯
    -0.08
    æİĤ
    -0.08
    oucher
    -0.08
    åIJ©
    -0.08
    ç´ł
    -0.08
    åĨ¤
    -0.08
    agra
    -0.08
    éĿ¢
    -0.08
    POSITIVE LOGITS
     levels
    0.19
    levels
    0.17
     amounts
    0.15
    Levels
    0.14
     Levels
    0.14
    (levels
    0.14
     ÑĥÑĢов
    0.13
     nive
    0.12
    æ°´åĩĨ
    0.12
     mức
    0.12
    Activations Density 0.210%

    No Known Activations