© 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. 54937
    Prev
    Next
    INDEX
    Explanations

    on the go

    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
    rawl
    -0.09
     обеÑģп
    -0.09
    odie
    -0.09
    nt
    -0.08
     ÑģобÑģÑĤвенно
    -0.08
    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    -0.08
    .cy
    -0.08
    ials
    -0.08
    ç´ļ
    -0.08
    elim
    -0.08
    POSITIVE LOGITS
     Fur
    0.09
    ãģĨãģ¡ãģ«
    0.09
     tá»Ļc
    0.09
    åıĺå¹»
    0.09
    ç¼ī
    0.08
     fur
    0.08
    abb
    0.08
    holding
    0.08
    kits
    0.08
     tô
    0.08
    Activations Density 0.010%

    No Known Activations