© 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. 10885
    Prev
    Next
    INDEX
    Explanations

    documentation and code formatting

    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
    èĥ¼
    -0.10
    surf
    -0.09
     Laden
    -0.09
    ¤í
    -0.09
    ibbon
    -0.09
    æĿ¿æĿIJ
    -0.08
    åĿł
    -0.08
    urous
    -0.08
    ifice
    -0.08
    (png
    -0.08
    POSITIVE LOGITS
    ç¥ĸ
    0.09
    ä¸ĢçĽ´æ²¡æľī
    0.09
    ä¸ī家
    0.09
    饮
    0.08
    ivals
    0.08
    éĽį
    0.08
    loo
    0.08
    çIJ°
    0.08
    Opts
    0.08
    èĪĨ
    0.08
    Activations Density 0.343%

    No Known Activations