© 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. 57898
    Prev
    Next
    INDEX
    Explanations

    multi-byte characters

    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
    åIJĮæĹ¶ä¹Łæĺ¯
    -0.11
    æĿ¥è¿ĻéĩĮ
    -0.11
    éĻ¤å¤ĸ
    -0.09
    å°±æĺ¯ä¸ºäºĨ
    -0.09
    åĬłåĪĨ
    -0.08
    ucc
    -0.08
    ért
    -0.08
    å²Į
    -0.08
    å¼ķ
    -0.08
    寧
    -0.08
    POSITIVE LOGITS
    è¿ĩç¨ĭä¸Ń
    0.13
    çļĦè¿ĩç¨ĭ
    0.12
    çļĦè¿ĩç¨ĭä¸Ń
    0.11
     нев
    0.10
    éģİç¨ĭä¸Ń
    0.10
    åħ¨è¿ĩç¨ĭ
    0.10
    ä¸Ģåī¯
    0.09
    çļĦæĸ¹å¼ı
    0.09
    ãģ¨ãģĦãģĨãĤĤãģ®
    0.09
    çĦ¶åIJİåĨį
    0.09
    Activations Density 0.351%

    No Known Activations