© 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. 50157
    Prev
    Next
    INDEX
    Explanations

    idiom, patent document, head injury

    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
     ÑģобÑģÑĤвенно
    -0.09
     xuyên
    -0.09
    èĤŁ
    -0.09
    ood
    -0.08
    åĴĮ个人
    -0.08
    åĬ¼
    -0.08
    ç½ij绾ç§ijæĬĢ
    -0.08
    å¼ĢæĿ¥
    -0.08
    ร
    -0.08
    /var
    -0.07
    POSITIVE LOGITS
    å¤ļä½Ļ
    0.11
    仨
    0.11
     COPYRIGHT
    0.09
    advertisement
    0.09
    æľī两ç§į
    0.09
    оÑĦ
    0.09
    ä¹ĭåŁİ
    0.08
    linger
    0.08
    íĭĢ
    0.08
    èĦļä¸ĭ
    0.08
    Activations Density 1.986%

    No Known Activations