© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21799
    Prev
    Next
    INDEX
    Explanations

    weapon or crime related words

    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
    .outer
    -0.10
    å®ĺåı¸
    -0.09
    æĪijéĥ½
    -0.08
    绦
    -0.08
    æĺ¯æĪijåĽ½
    -0.08
    DBG
    -0.08
    è¿ĻæĿ¡è·¯
    -0.08
    御
    -0.08
    åIJĦç§įåIJĦæł·
    -0.08
     stap
    -0.08
    POSITIVE LOGITS
    书æ³ķ
    0.09
    ÑĢев
    0.09
    æķ£åıij
    0.09
    \Factories
    0.09
    æĹłå£°
    0.08
    ä½ľäºĨ
    0.08
    çŃīå¤ļ个
    0.08
    roken
    0.08
    deaux
    0.08
    ãģ¨ãģĦãģĨãģĵãģ¨ãģ§
    0.08
    Activations Density 0.064%

    No Known Activations