© 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. 54528
    Prev
    Next
    INDEX
    Explanations

    subject matter or category

    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
    èĥ¥
    -0.10
    ands
    -0.09
    转身
    -0.09
    /gl
    -0.09
    IDS
    -0.09
    øj
    -0.09
    åİ¢
    -0.09
    rades
    -0.09
    顺德
    -0.09
    ikel
    -0.08
    POSITIVE LOGITS
    åĴĮæĬĢæľ¯
    0.10
     Pell
    0.08
    .this
    0.08
    ä¸Ģéĥ¨åĪĨ
    0.08
    æľīéĹ®é¢ĺ
    0.08
    beit
    0.08
    俨
    0.08
     Toys
    0.08
    обÑħодим
    0.07
     rushes
    0.07
    Activations Density 0.524%

    No Known Activations