© 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. 33764
    Prev
    Next
    INDEX
    Explanations

    Pathways, MBHD, Tusculum, Bebo, Langar, Tarique, Secwepemc, Myer, Ingenuityfest, Isospora

    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
    emean
    -0.08
    AllWindows
    -0.08
    ridged
    -0.08
    ÑĮев
    -0.08
    ioned
    -0.08
    ร
    -0.07
    (docs
    -0.07
    zung
    -0.07
    ftware
    -0.07
    dep
    -0.07
    POSITIVE LOGITS
    bies
    0.11
    æĪĸèĢħåħ¶ä»ĸ
    0.11
    /her
    0.10
     à¹Ĩ
    0.10
    å¤ļ彩
    0.10
    å·´å·´
    0.09
    æĪĸåħ¶ä»ĸ
    0.09
    åIJĦç§įåIJĦæł·
    0.09
    stead
    0.09
    atie
    0.09
    Activations Density 0.764%

    No Known Activations