© 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. 29966
    Prev
    Next
    INDEX
    Explanations

    race and racial topics

    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
    inery
    -0.10
    rico
    -0.09
    æ´µ
    -0.09
     Habit
    -0.09
    sf
    -0.09
     paraph
    -0.09
    fol
    -0.09
    æĪ·ç±į
    -0.08
    çļĦé£Łçī©
    -0.08
    俱
    -0.08
    POSITIVE LOGITS
    ç§įæĹı
    0.12
     racially
    0.10
     racial
    0.10
    iya
    0.09
    奴éļ¶
    0.09
    çĨĶ
    0.09
     race
    0.08
     Rooney
    0.08
     Intersection
    0.08
    令
    0.08
    Activations Density 0.367%

    No Known Activations