© 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. 31969
    Prev
    Next
    INDEX
    Explanations

    admiration, imagination, support, ire, respect, wallet

    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
    宾客
    -0.09
    èįī
    -0.08
    身价
    -0.08
    为主线
    -0.08
    æ¿Ĵ临
    -0.08
    人家
    -0.08
    ä»ĸ对
    -0.08
    æĪIJé¾Ļ
    -0.08
    .hardware
    -0.08
    éªĺ
    -0.07
    POSITIVE LOGITS
    éĿĴçĿIJ
    0.12
     attention
    0.12
    æķ¬çķı
    0.12
    çľ¼çIJĥ
    0.12
    çļĦåħ³æ³¨
    0.11
    å°Ĭæķ¬
    0.11
    attention
    0.10
    Attention
    0.10
    注æĦıåĬĽ
    0.10
    çŀ©çĽ®
    0.10
    Activations Density 0.048%

    No Known Activations