© 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. 5688
    Prev
    Next
    INDEX
    Explanations

    headings and titles

    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
    æĸ°åĬł
    -0.08
    اد
    -0.07
    tern
    -0.07
    æĸ¹éĿ¢
    -0.07
     Easter
    -0.07
    æĬĵ
    -0.07
     Dept
    -0.07
     EOF
    -0.07
    è¾Ľ
    -0.07
     Attribution
    -0.07
    POSITIVE LOGITS
    contents
    0.10
    å¹»
    0.10
    -append
    0.09
    æľ¬ä¹¦
    0.09
    áĢı
    0.09
    çĺ¾
    0.09
     pageTitle
    0.09
    页
    0.09
    èį©
    0.09
     Contents
    0.09
    Activations Density 0.516%

    No Known Activations