© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51775
    Prev
    Next
    INDEX
    Explanations

    've, have, has

    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
    èµĦæºIJéħįç½®
    -0.10
    åĨľåī¯
    -0.09
    éªĬ
    -0.09
    åįĬå°ıæŶ
    -0.09
    èIJ½å¹ķ
    -0.08
    æĺĻ
    -0.08
    çīĤ
    -0.08
     découvert
    -0.08
    转åħ¥
    -0.08
    é¾Ļæ¹ĸ
    -0.08
    POSITIVE LOGITS
    ä¸Ģ缴æĺ¯
    0.23
    ä¸ĢçĽ´ä»¥æĿ¥
    0.21
    ä¸Ģ缴éĥ½æĺ¯
    0.21
    ä¸Ģ缴
    0.20
    ä¸Ģ缴éĥ½
    0.20
    éķ¿æľŁä»¥æĿ¥
    0.17
    ä¸ĢçĽ´åľ¨
    0.17
    ä¸ĢåIJij
    0.15
    ä¸Ģè´¯
    0.15
    æĪijä¸Ģ缴
    0.15
    Activations Density 0.200%

    No Known Activations