© 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. 48787
    Prev
    Next
    INDEX
    Explanations

    don't get me wrong

    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
    åĮĢ
    -0.09
    ude
    -0.09
    lage
    -0.08
    带é¢Ĩ
    -0.08
     sue
    -0.08
    ellation
    -0.08
    åĽŀåĪ°å®¶
    -0.08
    åĽ¾çº¸
    -0.08
     uninsured
    -0.08
    å¤ļç§įå½¢å¼ı
    -0.08
    POSITIVE LOGITS
    å¾Ĺ太
    0.11
    è¦ģ说
    0.10
    æī°
    0.10
    太å¤ļçļĦ
    0.09
    æµİ
    0.09
    çĵ¶
    0.09
    æĺ¯æĢİæł·
    0.09
    è¿ĻæĿ¡
    0.09
    heck
    0.09
    ä¹Łä¸įè¦ģ
    0.09
    Activations Density 0.005%

    No Known Activations