© 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. 4365
    Prev
    Next
    INDEX
    Explanations

    reforms, revised, bill, Obama, weaker

    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
    ocket
    -0.09
    ä¿ı
    -0.09
    æĦŁåĨĴ
    -0.09
    é£İåij³
    -0.09
    ä¿ĿæĮģçĿĢ
    -0.08
    ден
    -0.08
    æºĸ
    -0.08
    åĴķ
    -0.08
    åĩĨ
    -0.08
    æľįä»İ
    -0.08
    POSITIVE LOGITS
    æķĻè®Ń
    0.12
    æķ´æĶ¹æݪæĸ½
    0.10
    å®ĮåĸĦ
    0.10
    åĩºåı°
    0.10
    æѤåIJİ
    0.09
    åıįæĢĿ
    0.09
    æѤ类
    0.09
    å½ĵåį³
    0.09
    çijŁ
    0.09
    åĬłå¼º
    0.09
    Activations Density 0.115%

    No Known Activations