© 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. 29444
    Prev
    Next
    INDEX
    Explanations

    winning

    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
     Dest
    -0.10
    ponge
    -0.10
     DEST
    -0.09
    Dest
    -0.09
     Destiny
    -0.09
     cum
    -0.09
    _lb
    -0.09
    æĮ½åĽŀ
    -0.08
    绵
    -0.08
    _capabilities
    -0.08
    POSITIVE LOGITS
     вал
    0.09
    é«ĺä½İ
    0.09
    åŁİå¸Ĥ管çIJĨ
    0.09
    çĥŃ线
    0.09
    åĬłä»¥
    0.09
    ë¹Ļ
    0.08
    åħ±è¯Ĩ
    0.08
    æĪIJç«ĭ
    0.08
    subs
    0.08
    æĪIJç«ĭäºĨ
    0.08
    Activations Density 0.043%

    No Known Activations