© 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. 49886
    Prev
    Next
    INDEX
    Explanations

    real-world assistance and experience

    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
    为广大
    -0.10
    éĥ¨åī¯
    -0.09
    ç»ĻåĪ«äºº
    -0.09
    è¿Ļæł·åģļ
    -0.09
    éļ¾æ°ij
    -0.09
    à¹Ģà¸Ńา
    -0.09
    schools
    -0.09
    该åħ¬åı¸
    -0.09
    èIJ¥ä¸ļæī§çħ§
    -0.09
    ÙģÙĪض
    -0.08
    POSITIVE LOGITS
    éĩįè§Ĩ
    0.09
    åĽ¾ä¹¦é¦Ĩ
    0.09
    ä¸¥æł¼
    0.08
    æ¯ıä¸Ģ天
    0.08
    ä¸¥æł¼çļĦ
    0.08
    丰å¯Įå¤ļ彩
    0.08
    æ°Ķæ°Ľ
    0.08
     attain
    0.08
     attent
    0.08
    eat
    0.08
    Activations Density 0.055%

    No Known Activations