© 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. 6556
    Prev
    Next
    INDEX
    Explanations

    main reason / biggest difference

    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
    à¹Ģà¸īà¸ŀาะ
    -0.08
    æķ´ä¸ª
    -0.08
    ppers
    -0.07
    åĺİ
    -0.07
    ary
    -0.07
    èµ°åIJij
    -0.07
    æīį
    -0.07
    vre
    -0.07
    éĥ¤
    -0.07
     resilient
    -0.07
    POSITIVE LOGITS
    å¸ĮæľĽéĢļè¿ĩ
    0.10
    ä¿Ŀè¯ģ
    0.09
    èĽĨ
    0.09
    å°¤åħ¶
    0.09
    åĢ¼å¾Ĺ注æĦı
    0.09
     issu
    0.09
    acer
    0.09
    èĻļ
    0.09
    /copyleft
    0.09
    çľĭçĤ¹
    0.09
    Activations Density 0.095%

    No Known Activations