© 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. 63959
    Prev
    Next
    INDEX
    Explanations

    information storage and use

    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
    éĤ£å°±æĺ¯
    -0.09
    nas
    -0.09
    ningar
    -0.08
    天
    -0.08
     unmatched
    -0.08
    产ä¸ļåįĩ级
    -0.08
    aar
    -0.08
    tracker
    -0.07
    åĢĻ
    -0.07
    åŀ¢
    -0.07
    POSITIVE LOGITS
    EDA
    0.10
    çĨŁç»ĥ
    0.10
    æķĻèģĮå·¥
    0.10
    è¾Ľ
    0.09
    åŃĺåĤ¨
    0.09
    Occurs
    0.09
    Ñıн
    0.09
    ä¹Łä¸įä¼ļ
    0.09
    åħļåĴĮåĽ½å®¶
    0.09
    æķ£åıijçĿĢ
    0.09
    Activations Density 0.023%

    No Known Activations