© 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. 55376
    Prev
    Next
    INDEX
    Explanations

    created by or with

    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
    OrCreate
    -0.11
    人家
    -0.11
    arez
    -0.11
    oret
    -0.10
    edly
    -0.10
    aki
    -0.10
    atically
    -0.09
    weed
    -0.09
    ging
    -0.09
    ãģĹãģ¾ãģĨ
    -0.09
    POSITIVE LOGITS
    OrUpdate
    0.18
    สรร
    0.16
     havoc
    0.15
    åĴĮåıijå±ķ
    0.10
    ../../../
    0.10
    ÑĩеÑģÑĤво
    0.10
     קשר
    0.10
    ReadStream
    0.10
    äºĨä¸įå°ij
    0.09
    ä¸Ģ个æĸ°çļĦ
    0.09
    Activations Density 0.104%

    No Known Activations