© 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. 60750
    Prev
    Next
    INDEX
    Explanations

    twin, double, counterpart

    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
    ä¸Ģé¡¿
    -0.11
    èģĶåIJĪ
    -0.10
    ä¸įä¸ĭ
    -0.10
    çݩ家们
    -0.09
    讳
    -0.09
    éĻĦå±ŀ
    -0.09
    é¢ĨåħĪ
    -0.09
    èijĨ
    -0.09
    åıĪä¸Ģ次
    -0.09
    æ¯Ĵ
    -0.09
    POSITIVE LOGITS
     counterparts
    0.10
    èººåľ¨åºĬä¸Ĭ
    0.10
    åIJĮçŃī
    0.10
    åı¦ä¸ĢåįĬ
    0.09
     twin
    0.09
     counterpart
    0.09
    éķľåŃIJ
    0.09
    ubber
    0.09
    åı¦ä¸Ģ个
    0.09
    åŃĺåľ¨çļĦ
    0.09
    Activations Density 0.076%

    No Known Activations