© 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. 3584
    Prev
    Next
    INDEX
    Explanations

    per se

    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
    åĭī
    -0.08
    shire
    -0.08
    dir
    -0.08
    åľ¨æĪijåĽ½
    -0.08
     cach
    -0.08
    ordinate
    -0.08
    elligence
    -0.08
    ãĤ¿ãĥ«
    -0.08
    apot
    -0.08
    ghan
    -0.08
    POSITIVE LOGITS
    èĢĮæĺ¯
    0.12
    æĦıä¹īä¸ĬçļĦ
    0.12
    æľ¬èº«çļĦ
    0.11
    æĽĿ
    0.10
    大家éĥ½åľ¨
    0.09
    ä¸Ńå°ı
    0.09
    æľ¬èº«
    0.09
    åIJĮ缣
    0.09
    Allen
    0.09
    Misc
    0.09
    Activations Density 0.053%

    No Known Activations