© 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. 578
    Prev
    Next
    INDEX
    Explanations

    official, new, norm, complete, legal, decision

    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
    纵横
    -0.10
     Idea
    -0.09
    没æ³ķ
    -0.09
    åĽ½å¤ĸ
    -0.08
    æĪĺåĽ½
    -0.08
    çĪIJ
    -0.08
    æĿµ
    -0.08
    нима
    -0.08
    èĩªè´Ł
    -0.08
    .Plugin
    -0.08
    POSITIVE LOGITS
     REP
    0.10
    é¼ĵåĬ±
    0.09
    天çĦ¶
    0.09
    æľ¬ç«Ļ
    0.09
    uner
    0.08
    elen
    0.08
    å¥ĭè¿Ľ
    0.08
    ele
    0.08
     Airways
    0.08
    æĢĿèĢĥ
    0.08
    Activations Density 0.018%

    No Known Activations