© 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. 47622
    Prev
    Next
    INDEX
    Explanations

    progress and positive states

    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
    ç¿ĺ
    -0.09
    LEAN
    -0.09
     Lean
    -0.08
    æĥij
    -0.08
    erd
    -0.08
    ler
    -0.08
    Lean
    -0.08
    æĿĥ
    -0.08
     Geme
    -0.08
    åĸ»
    -0.08
    POSITIVE LOGITS
    好åĿı
    0.14
    æĢ»ä½ĵ
    0.12
    åģ¶
    0.12
    è´¨éĩıåĴĮ
    0.12
    èī¯å¥½
    0.12
    èī¯å¥½çļĦ
    0.12
    ä¸Ģçīĩ
    0.11
    产åĵģè´¨éĩı
    0.11
    好转
    0.11
    æľįåĬ¡è´¨éĩı
    0.11
    Activations Density 0.134%

    No Known Activations