© 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. 5308
    Prev
    Next
    INDEX
    Explanations

    longer, more, extra

    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
    ä¸į足
    -0.09
    俯
    -0.09
    积
    -0.09
    .Release
    -0.09
    _TOO
    -0.08
    :UI
    -0.08
    è§ĤéŁ³
    -0.08
    دار
    -0.08
    ä¸įä¼ļåĨį
    -0.08
     updates
    -0.08
    POSITIVE LOGITS
    é¢Ŀå¤ĸ
    0.14
    å¢ŀåĬłäºĨ
    0.13
    (extra
    0.13
     slower
    0.11
     extra
    0.11
     longer
    0.11
     Extra
    0.11
    extra
    0.11
    _EXTRA
    0.10
    å¢ŀåĬł
    0.10
    Activations Density 0.117%

    No Known Activations