© 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. 27116
    Prev
    Next
    INDEX
    Explanations

    responsibility, participation, efficiency, satisfaction, practice, control, care, equity, metabolism

    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
    osti
    -0.09
    دÙĨÙĬ
    -0.09
    æĸ°çļĦ
    -0.08
    ä¸īä½įä¸Ģä½ĵ
    -0.08
     tendency
    -0.08
    æĢĿç»´æĸ¹å¼ı
    -0.08
     way
    -0.08
    ç»ıè¿ĩ
    -0.08
    ä¾Ľæ±Ĥ
    -0.08
    éģĩåĪ°äºĨ
    -0.08
    POSITIVE LOGITS
    建设åĴĮ
    0.14
    管çIJĨåĴĮ
    0.13
    çIJĨ解åĴĮ
    0.11
     success
    0.09
     recall
    0.09
    .documentation
    0.09
     overs
    0.09
    aje
    0.08
     action
    0.08
     re
    0.08
    Activations Density 0.488%

    No Known Activations