© 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. 16856
    Prev
    Next
    INDEX
    Explanations

    announcements and dates

    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
    表述
    -0.10
    ä¸įåłª
    -0.09
    theless
    -0.09
    åĪĨéĴŁåIJİ
    -0.09
    æł¹æľ¬
    -0.09
    ä¸Ģå¤ľ
    -0.09
    ioned
    -0.09
    ç»İ
    -0.08
    idge
    -0.08
    bench
    -0.08
    POSITIVE LOGITS
    ä¸Ģ声
    0.09
    åŃIJå¼Ł
    0.09
    æĩĤå¾Ĺ
    0.09
     hô
    0.09
    å¾Ĵ
    0.08
    ubes
    0.08
    aly
    0.08
    åºĬä½į
    0.08
    lay
    0.08
    loud
    0.07
    Activations Density 0.031%

    No Known Activations