© 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. 2266
    Prev
    Next
    INDEX
    Explanations

    believe that

    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
    æĺ¯æĢİä¹Ī
    -0.10
    æĺ¯å¦Ĥä½ķ
    -0.09
    åįĩé«ĺ
    -0.09
     ÑĥÑģили
    -0.09
    æĭ§
    -0.09
    ahren
    -0.09
    为æĬĵ
    -0.08
    æ®ĩ
    -0.08
    odb
    -0.08
    çľĭèµ·æĿ¥
    -0.08
    POSITIVE LOGITS
    åĽºå®ļçļĦ
    0.09
    åĢ¼å¾Ĺä¸Ģ
    0.09
    人éĢī
    0.08
    ãģ´
    0.08
    åħħåĪĨåıijæĮ¥
    0.08
    ä¸ĭä¸Ģ个
    0.08
    éĿŀ常éĢĤåIJĪ
    0.08
    Cascade
    0.08
    è¡©
    0.08
    ä½łåºĶ该
    0.08
    Activations Density 0.024%

    No Known Activations