© 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. 27981
    Prev
    Next
    INDEX
    Explanations

    false eyelashes, faux fur, lights, volume

    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
    ç³ķ
    -0.10
    èĽĨ
    -0.10
    èĢķèĢĺ
    -0.10
    é¦Ĭ
    -0.10
    èħ°éĥ¨
    -0.09
    èĤĵ
    -0.09
    æľīçĶ¨
    -0.09
    è°ĥåij³
    -0.09
    æľīæĦıä¹ī
    -0.09
     Bai
    -0.09
    POSITIVE LOGITS
     fake
    0.10
    æī¿æİ¥
    0.10
    åĪ«
    0.09
    åįĹæµ·
    0.09
    lashes
    0.09
    æĩĴ
    0.09
    åģĩ
    0.08
    çıł
    0.08
     lashes
    0.08
    (fake
    0.08
    Activations Density 0.133%

    No Known Activations