© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 63898
    Prev
    Next
    INDEX
    Explanations

    back to

    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
    Negative Logits
    æĹģ
    -0.10
    èĭŁ
    -0.09
    itters
    -0.09
    ä¸Ĭä¸ĭ
    -0.09
    ãģ¨ãģ£ãģ¦
    -0.09
    æĹģè¾¹çļĦ
    -0.09
    æ¼ĵ
    -0.09
    àªĪ
    -0.08
    ailed
    -0.08
    èĩ
    -0.08
    POSITIVE LOGITS
    åĽŀæĿ¥
    0.15
    纥
    0.13
    (Return
    0.13
    åĽŀåĪ°
    0.13
    åĽŀä¾Ĩ
    0.12
    Ñĩив
    0.12
    åĽŀåİ»
    0.11
    éĩįè¿Ķ
    0.11
    带åĽŀ
    0.11
    åĽŀæĿ¥äºĨ
    0.11
    Activations Density 0.280%

    No Known Activations