© 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. 49798
    Prev
    Next
    INDEX
    Explanations

    both sexes/genders/kinds

    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
    æĽ´å¤ļ
    -0.09
    urat
    -0.09
    ê¹IJ
    -0.08
     otherwise
    -0.08
    åħ¨èº«
    -0.08
     mult
    -0.08
    å¤ļæł·æĢ§
    -0.08
    ãĤģ
    -0.08
    溷
    -0.08
    ложений
    -0.07
    POSITIVE LOGITS
     sexes
    0.16
    /all
    0.14
    หมà¸Ķ
    0.14
     genders
    0.12
    ç±»åŀĭçļĦ
    0.11
    æĸĩåĮĸåĴĮ
    0.11
     kinds
    0.11
     halves
    0.11
    -sex
    0.10
     sides
    0.10
    Activations Density 0.087%

    No Known Activations