© 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. 39620
    Prev
    Next
    INDEX
    Explanations

    chapter titles and fantasy names

    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
    aping
    -0.09
    è°ĥçIJĨ
    -0.08
    表çĻ½
    -0.08
    éģĵ士
    -0.08
     Hercules
    -0.08
    opsy
    -0.08
    å©¢
    -0.08
    é¾Ľ
    -0.08
    èī²ç´ł
    -0.08
    ooting
    -0.08
    POSITIVE LOGITS
    è¸ıåħ¥
    0.11
    åŁİå¸ĤçļĦ
    0.09
    ä¹ĭåŁİ
    0.09
    缴å±ŀ
    0.09
    çŀ³
    0.08
    侬
    0.08
    è¡ĢèĦī
    0.08
    é¼ĵ
    0.08
    .splitext
    0.08
    Nonnull
    0.08
    Activations Density 0.176%

    No Known Activations