© 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. 47589
    Prev
    Next
    INDEX
    Explanations

    hence why

    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
    åıįæŃ£
    -0.11
    ÄĻ
    -0.10
    ä¸Ńæľī
    -0.10
    Äħc
    -0.10
    åĵĩ
    -0.10
    é«®
    -0.09
    éĵĥ
    -0.09
    æĦıå¤ĸ
    -0.08
    zÄĻ
    -0.08
    è®®äºĭ
    -0.08
    POSITIVE LOGITS
    forth
    0.31
     forth
    0.12
    forward
    0.11
    zza
    0.10
    çĦ¶æĺ¯
    0.09
    eci
    0.09
    ãģĵãģ¨ãģĮãģ§ãģįãĤĭ
    0.09
    ê»ı
    0.09
    ek
    0.09
    iday
    0.09
    Activations Density 0.006%

    No Known Activations