© 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. 24267
    Prev
    Next
    INDEX
    Explanations

    qualities that make good

    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
    ãģ®ãģĵãģ¨
    -0.10
    çĻ½è¡£
    -0.09
    ä¿Ŀåį«
    -0.09
    骧
    -0.09
    åĩŃçĿĢ
    -0.09
    quo
    -0.09
    éĺ²åį«
    -0.09
    æĪİ
    -0.09
    ency
    -0.09
    è¯ķ
    -0.09
    POSITIVE LOGITS
    好çļĦ
    0.13
     successful
    0.13
     good
    0.12
    ä¸Ģ个好çļĦ
    0.12
    èī¯å¥½
    0.12
    æĪIJåĬŁ
    0.12
    _success
    0.11
    æĪIJ为ä¸Ģ个
    0.11
    successful
    0.11
    Successful
    0.11
    Activations Density 0.303%

    No Known Activations