© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19844
    Prev
    Next
    INDEX
    Explanations

    Copyright notices

    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
    讲座
    -0.10
    åŀŃ
    -0.10
    æĬ¢
    -0.10
    ä¸Ģéĺµ
    -0.09
    èį«
    -0.09
    è¿Ļç¯ĩæĸĩç«ł
    -0.09
    墩
    -0.09
    ä¹Łæĺ¯å¦ĤæѤ
    -0.09
    åIJij社ä¼ļ
    -0.08
    èĿĩ
    -0.08
    POSITIVE LOGITS
    åIJĦæĸ¹éĿ¢
    0.10
    arius
    0.09
    åIJĦ个æĸ¹éĿ¢
    0.08
    à¹ģà¸ľà¸Ļ
    0.08
    CONTACT
    0.08
    è¿Ķ
    0.08
    åIJĦæĸ¹
    0.08
    зÑĥ
    0.08
    å¾ģæľį
    0.08
    侬
    0.08
    Activations Density 0.013%

    No Known Activations