© 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. 5472
    Prev
    Next
    INDEX
    Explanations

    C++ code and libraries

    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
    åİ¥
    -0.10
     Hayes
    -0.08
    稿
    -0.08
    woo
    -0.08
    ãĤ´
    -0.08
    igor
    -0.08
    GOR
    -0.07
    ä¿Ł
    -0.07
    ulo
    -0.07
    wig
    -0.07
    POSITIVE LOGITS
    .cpp
    0.16
    (std
    0.14
     std
    0.14
    cpp
    0.14
    ,std
    0.13
     boost
    0.13
    <std
    0.13
    /std
    0.13
    .hpp
    0.12
    /cpp
    0.12
    Activations Density 1.708%

    No Known Activations