© 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. 49826
    Prev
    Next
    INDEX
    Explanations

    articles with titles

    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
    åijĺ
    -0.09
    imus
    -0.09
    æ¢ĥ
    -0.08
    æ¹Ľ
    -0.07
    åĦ¿
    -0.07
    catch
    -0.07
    ä¸Ģä¸ĭåŃIJ
    -0.07
    袤
    -0.07
     fict
    -0.07
    shal
    -0.07
    POSITIVE LOGITS
     Soros
    0.10
    annon
    0.09
    åĬŃ
    0.09
    ãĥĹãĥªãĥ³
    0.09
    vron
    0.09
    год
    0.09
     اÙĦعرب
    0.09
     ÑĢаÑģположен
    0.09
    /from
    0.09
    rottle
    0.09
    Activations Density 0.110%

    No Known Activations