© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 16749
    Prev
    Next
    INDEX
    Explanations

    for you, for me, for yourself

    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
    Negative Logits
    æݪ
    -0.10
     Jack
    -0.09
    æŁIJç§į
    -0.09
    èĪ·
    -0.09
    æĭ³
    -0.08
    forcer
    -0.08
    Jack
    -0.08
    é¢ĦæĸĻ
    -0.08
    对人ä½ĵ
    -0.08
     exc
    -0.08
    POSITIVE LOGITS
    客æĪ·æıIJä¾Ľ
    0.12
    人æ°ijæľįåĬ¡
    0.10
    åIJİèĢħ
    0.09
    èµ°å¾Ĺ
    0.09
    ä¸įåħĭ
    0.09
     reasons
    0.09
    endir
    0.09
     занÑıÑĤ
    0.09
    ä¸ĸ人
    0.09
     Thought
    0.09
    Activations Density 0.189%

    No Known Activations