© 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. 18891
    Prev
    Next
    INDEX
    Explanations

    fix/bug/issue/test

    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
    åIJĦç§įåIJĦæł·
    -0.10
    èĤ¯å®ļ
    -0.09
    çĻ½é¢Ĩ
    -0.09
    æĪijè¿ĺæĺ¯
    -0.09
    èµ°äºĨ
    -0.09
    ä»Ģä¹Īåij¢
    -0.09
    大éĥ¨åĪĨ人
    -0.09
    åIJĦæł·
    -0.09
    èĤ¯å®ļæĺ¯
    -0.08
    çĶ·æĸ¹
    -0.08
    POSITIVE LOGITS
    overe
    0.09
    è°IJ
    0.08
    æĽ´å¤ļ精彩
    0.08
    utan
    0.08
    æľīä¸įåIJĮçļĦ
    0.08
    æ³īæ°´
    0.08
    éij
    0.07
    缩
    0.07
    æľīæĽ´å¥½çļĦ
    0.07
    	when
    0.07
    Activations Density 0.609%

    No Known Activations