© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37709
    Prev
    Next
    INDEX
    Explanations

    musician, singer, or performer

    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
    强硬
    -0.10
    Ïĭ
    -0.10
    uard
    -0.09
    idth
    -0.09
    IDEO
    -0.09
     illeg
    -0.09
    主æĦı
    -0.09
    ddit
    -0.09
    عاÙħ
    -0.09
    leg
    -0.08
    POSITIVE LOGITS
    缴éĢļ车
    0.09
    æ»´
    0.09
    æį§
    0.09
     bergen
    0.08
    æĥłæ°ij
    0.08
    çĤ¹äºĨ
    0.08
    湿润
    0.08
    治好
    0.08
    æ¼Ķç»İ
    0.08
    辨
    0.08
    Activations Density 0.144%

    No Known Activations