© 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. 19094
    Prev
    Next
    INDEX
    Explanations

    attractive offerattractive investorsattractive ideaattractive approachattractive alternative

    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
    æľ¬çĿĢ
    -0.09
    ä¸ĢåıĮ
    -0.09
    é£İæł¼
    -0.09
    ä¸ĢåĪĻ
    -0.08
    ä¸į代表
    -0.08
    è·Łä»ĸ
    -0.08
     dens
    -0.08
    ÙĪص
    -0.08
    æĶ¾å¼Ģ
    -0.08
    æĪijåıªæĺ¯
    -0.08
    POSITIVE LOGITS
    åIJ¸å¼ķåĬĽ
    0.26
    åIJ¸å¼ķ
    0.21
    诱æĥij
    0.21
    åIJ¸å¼ķäºĨ
    0.18
    åIJ¸å¼ķæĽ´å¤ļ
    0.15
     attractive
    0.15
    诱
    0.15
    诱人
    0.14
     allure
    0.14
     appeal
    0.14
    Activations Density 0.045%

    No Known Activations