© 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. 58271
    Prev
    Next
    INDEX
    Explanations

    highly recommended or desirable

    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
    è¶³å¤ŁçļĦ
    -0.11
    æĽ´å¤ļçļĦæĺ¯
    -0.11
    å¤ļå°ij
    -0.10
    å¾Ī好
    -0.10
    æĽ´å¥½çļĦ
    -0.10
    å¾Ī大
    -0.10
    .medium
    -0.09
    èĥ¥
    -0.09
    æĹłä¸į
    -0.09
    æĽ´å¿«
    -0.09
    POSITIVE LOGITS
     noqa
    0.09
    eco
    0.09
    åĴ±
    0.09
    ernels
    0.08
     Äiju
    0.08
    éĥ¨åī¯
    0.08
    hood
    0.08
    ernel
    0.08
    åįłæį®
    0.08
     abbrev
    0.08
    Activations Density 0.292%

    No Known Activations