© 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. 62134
    Prev
    Next
    INDEX
    Explanations

    explicit sexual terms

    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
    shire
    -0.11
    ç¾İæ´²
    -0.10
     Gry
    -0.09
    é¬ĺ
    -0.09
    åı¥
    -0.08
    zed
    -0.08
    zag
    -0.08
    éĥ½å¾Ī
    -0.08
    eer
    -0.08
    æ·ij
    -0.08
    POSITIVE LOGITS
     Harden
    0.09
    ì§ĢëıĦ
    0.08
    éĢı
    0.08
    äºĮæīĭæĪ¿
    0.08
    æ±ľ
    0.08
     Tiles
    0.08
    .synthetic
    0.08
    ittel
    0.08
    entic
    0.08
    Tiles
    0.08
    Activations Density 0.009%

    No Known Activations