© 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. 31557
    Prev
    Next
    INDEX
    Explanations

    2 followed by numbers, letters, or GO/kids/news/sister

    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
    Statics
    -0.09
    ucci
    -0.09
    hort
    -0.09
    udd
    -0.09
    åĪĴ
    -0.09
    ILLE
    -0.08
    izia
    -0.08
    ä¸¥æł¼èIJ½å®ŀ
    -0.08
    _mC
    -0.08
     Morm
    -0.08
    POSITIVE LOGITS
    gether
    0.15
    Tier
    0.09
    åĽ½æľī
    0.09
    pond
    0.09
    /from
    0.09
    æݧèĤ¡éĽĨåĽ¢
    0.08
    åľ°ä¸Ń
    0.08
    ombie
    0.08
    æĦıä¹īä¸ĬçļĦ
    0.08
    asted
    0.08
    Activations Density 0.073%

    No Known Activations