© 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. 63310
    Prev
    Next
    INDEX
    Explanations

    opening sequence

    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
     пÑĢек
    -0.10
    æľŁåĨħ
    -0.10
     Lump
    -0.09
    åIJĽ
    -0.09
    主ä¸ļ
    -0.09
    æľĢçµĤ
    -0.09
    andid
    -0.09
     Mellon
    -0.09
    (ed
    -0.09
    å¡«
    -0.09
    POSITIVE LOGITS
    å¼ĢéŨ
    0.10
    导åħ¥
    0.10
    è·³è·ĥ
    0.09
    å¼Ģ头
    0.09
    ç§ī
    0.09
    borough
    0.09
    witch
    0.08
    说èĩªå·±
    0.08
    å¾ĹçŁ¥
    0.08
    DECLARE
    0.08
    Activations Density 0.077%

    No Known Activations