© 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. 39966
    Prev
    Next
    INDEX
    Explanations

    Harry Potter characters and concepts

    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
    让æ¶Īè´¹èĢħ
    -0.12
    寨
    -0.10
    çļĦ女人
    -0.10
    åħ¬çĽĬ
    -0.10
    天åłĤ
    -0.10
    åĵŃäºĨ
    -0.09
    çĬ·
    -0.09
    ä¸ĢåĪĩéĥ½æĺ¯
    -0.09
    [next
    -0.09
    奶ç²ī
    -0.08
    POSITIVE LOGITS
     Rowling
    0.10
     Hogwarts
    0.10
     parchment
    0.10
    åħ¨éĿ¢æİ¨è¿Ľ
    0.09
     fer
    0.09
     Apprentice
    0.08
    çªĥ
    0.08
    umbledore
    0.08
    åĽ¾ä¹¦é¦Ĩ
    0.08
    è¿Ļä¸Ģå¹´
    0.08
    Activations Density 0.070%

    No Known Activations