© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 31615
    Prev
    Next
    INDEX
    Explanations

    professional or expert roles

    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
    Not in Any Lists

    No Comments

    Negative Logits
    管
    -0.09
    æĿĨ
    -0.09
    å¼Ģæ°´
    -0.09
    御
    -0.08
    ä¸ĸ代
    -0.08
    gem
    -0.08
    Gem
    -0.08
    身
    -0.08
    åĩºåĬ¨
    -0.08
     Comm
    -0.08
    POSITIVE LOGITS
    å¿ĥçIJĨåĴ¨è¯¢
    0.11
    Psy
    0.10
    å¿ĥçIJĨåѦ
    0.10
    å¿ĥçIJĨåģ¥åº·
    0.09
     Dart
    0.09
    æĿ¥è®¿
    0.09
    客æĪ·ç«¯
    0.09
    æĦ¤æĢĴ
    0.09
    忱
    0.09
    çļĦå¿ĥçIJĨ
    0.08
    Activations Density 0.222%

    No Known Activations