© 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. 1670
    Prev
    Next
    INDEX
    Explanations

    demolish, demography, or democracy

    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
    ivity
    -0.10
    åĽŀæĥ³
    -0.10
    shan
    -0.09
    ropy
    -0.09
    صÙĦ
    -0.08
    åįļæĸĩ
    -0.08
    ivi
    -0.08
     Vick
    -0.08
    Ò¯
    -0.08
    keit
    -0.08
    POSITIVE LOGITS
    stration
    0.13
    strate
    0.12
     Dem
    0.10
    äºĨ好å¤ļ
    0.09
    _CM
    0.09
    kas
    0.09
    äºĨ许å¤ļ
    0.09
    inations
    0.09
    rief
    0.08
    igned
    0.08
    Activations Density 0.019%

    No Known Activations