© 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. 58251
    Prev
    Next
    INDEX
    Explanations

    foreign words and place names

    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
    inary
    -0.10
    æķ´ä¸ª
    -0.09
    iken
    -0.09
    岸边
    -0.09
    è£Ķ
    -0.09
    æķ´ä¸ªäºº
    -0.09
    ansom
    -0.08
    lion
    -0.08
     зÑĢениÑı
    -0.08
    å¼Ģå¹ķå¼ı
    -0.08
    POSITIVE LOGITS
    çłĶ究人åijĺ
    0.10
    åĴª
    0.09
    åıijè¨Ģ人
    0.09
    é£İå°ļ
    0.09
    å¾Ĥ
    0.09
    楣
    0.09
    INGS
    0.08
    place
    0.08
    MediaType
    0.08
    mates
    0.08
    Activations Density 0.006%

    No Known Activations