© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8371
    Prev
    Next
    INDEX
    Explanations

    words or punctuation followed by content

    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
    èĮ±
    -0.09
    ogl
    -0.08
    ase
    -0.07
    åľ¨ç½ij绾
    -0.07
     candidacy
    -0.07
     Coc
    -0.07
    ä¸ĢåĪĩ
    -0.07
    ALLE
    -0.07
    ole
    -0.07
    Ho
    -0.07
    POSITIVE LOGITS
    รม
    0.11
    dı
    0.10
    ä¹Łæĺ¯å¦ĤæѤ
    0.10
     вообÑīе
    0.10
    åĿª
    0.10
    æİĴè¡Į
    0.10
    å°±è¿Ļä¹Ī
    0.10
    è¿Ļè¯Ŀ
    0.09
    .dictionary
    0.09
    ystems
    0.09
    Activations Density 0.269%

    No Known Activations