© 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. 27898
    Prev
    Next
    INDEX
    Explanations

    anime and Asian demographics

    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
    ellar
    -0.10
     ÑħоÑĤÑı
    -0.09
    çķ´
    -0.09
    æį¢å±Ĭ
    -0.09
     foreseeable
    -0.09
    ÑĤеÑĢ
    -0.09
    sWith
    -0.08
    ç¼İ
    -0.08
    ulet
    -0.08
    ×ķת
    -0.08
    POSITIVE LOGITS
    agra
    0.09
    idel
    0.09
    ĭ
    0.09
    ität
    0.09
    pery
    0.09
    yo
    0.09
    -Pacific
    0.09
    ocking
    0.09
    amp
    0.09
    åĪ°äºĨ
    0.08
    Activations Density 0.014%

    No Known Activations