© 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. 49997
    Prev
    Next
    INDEX
    Explanations

    takeaway lesson

    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
    èĭŁ
    -0.11
    箪
    -0.09
    ç«Ļ
    -0.09
     Garner
    -0.09
    é©·
    -0.08
    为äºĨ让
    -0.08
    积
    -0.08
    åIJĵ
    -0.08
    ä¸ĭéĿ¢æĺ¯å°ı
    -0.08
    åħ¨ç¨ĭ
    -0.08
    POSITIVE LOGITS
     takeaway
    0.13
    æĶ¶èİ·
    0.11
     lesson
    0.11
    æķĻè®Ń
    0.10
    ä»İä¸Ń
    0.10
    èģĶç³»æĸ¹å¼ı
    0.09
    带ç»Ļ
    0.09
     lessons
    0.09
    lessons
    0.09
    åį°è±¡
    0.09
    Activations Density 0.052%

    No Known Activations