© 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. 24233
    Prev
    Next
    INDEX
    Explanations

    two kinds of

    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
    @student
    -0.10
    çīĸ
    -0.09
    åIJĦè¡ĮåIJĦ
    -0.09
    åıĪ称
    -0.09
    åĩĦ
    -0.09
    ä¸ĢæŃ¥æŃ¥
    -0.09
    ä¸ĢéĿ¢
    -0.09
    anter
    -0.08
    eneg
    -0.08
    åĴĮ个人
    -0.08
    POSITIVE LOGITS
     main
    0.12
    ä¸Ģæĺ¯
    0.11
    ä»ĩ
    0.10
    两大
    0.10
    :
    0.10
    éĤ£æŶ
    0.09
    ä¸Ģ个æĺ¯
    0.09
    åĨį说
    0.09
    éĤ£å°±æĺ¯
    0.08
    主è¦ģ
    0.08
    Activations Density 0.238%

    No Known Activations