© 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. 49967
    Prev
    Next
    INDEX
    Explanations

    design and construction

    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
    ä¸ĢçĤ¹
    -0.09
    çļĦæĦıè§ģ
    -0.09
    monkey
    -0.08
    åIJĮåѦ
    -0.08
    åĪ°æĿ¥
    -0.08
    åħħ
    -0.08
    жи
    -0.08
    æĹ¶ä»£çļĦ
    -0.08
    bos
    -0.08
    æIJŀ
    -0.08
    POSITIVE LOGITS
    足以
    0.14
    使å¾Ĺ
    0.14
    åĨ³å®ļäºĨ
    0.10
    让å®ĥ
    0.10
    æĦıåij³çĿĢ
    0.10
    (coeffs
    0.10
    è®©ä½ł
    0.10
    æĬĬä½ł
    0.09
    使åħ¶
    0.09
    ä¸ĢåĪĩéĥ½
    0.09
    Activations Density 0.187%

    No Known Activations