© 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. 28002
    Prev
    Next
    INDEX
    Explanations

    erotica and lifestyle content involving specific fetishes and dating

    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
    éĻ¢å£«
    -0.10
    èģĴ
    -0.09
    ç¿¡
    -0.09
    å¹ķ
    -0.09
    辩
    -0.09
    äºļè¿IJä¼ļ
    -0.09
    åľ°ä¸Ń
    -0.09
    _FAR
    -0.08
    ulary
    -0.08
    俪
    -0.08
    POSITIVE LOGITS
     BDSM
    0.11
     Doll
    0.09
    责任å¿ĥ
    0.09
     Doc
    0.09
     Deg
    0.08
    çĿ£
    0.08
    æĻļ
    0.08
    强åĮĸ
    0.08
     spanking
    0.08
     openly
    0.08
    Activations Density 0.157%

    No Known Activations