© 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. 1996
    Prev
    Next
    INDEX
    Explanations

    promises, accused, avoids, response, contact

    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¿ĩå¾Ĺ
    -0.10
    anon
    -0.10
    çľĭçĤ¹
    -0.10
    åij¼åIJ¸
    -0.10
    ient
    -0.10
    æµģåĩº
    -0.09
    ä¸ĵ线
    -0.09
    鼷æĸ¯
    -0.09
    èİ·å¾ĹæĽ´å¤ļ
    -0.09
    å¿įä¸įä½ı
    -0.09
    POSITIVE LOGITS
    æī¿è¯º
    0.09
    èģĶç³»
    0.09
     promised
    0.09
    书éĿ¢
    0.09
    æ²ŁéĢļ
    0.09
    æĭĴ
    0.09
    åĩłæ¬¡
    0.09
    çļĦæĢģ度
    0.09
     contact
    0.08
    æīĢ说çļĦ
    0.08
    Activations Density 0.047%

    No Known Activations