© 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. 29630
    Prev
    Next
    INDEX
    Explanations

    prohibit from

    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
    æĹłçĸij
    -0.10
    ä¸įåĨį
    -0.10
    éļıæŶ
    -0.09
    åıĭæĥħ
    -0.09
    å¼
    -0.08
    交åıĭ
    -0.08
    åįĥæĸ¹çĻ¾è®¡
    -0.08
    èĪ¸
    -0.08
     legit
    -0.08
    èĢĮä¸įæĺ¯
    -0.08
    POSITIVE LOGITS
    æľīä»»ä½ķ
    0.18
    ä»»ä½ķå½¢å¼ı
    0.16
    ä»»ä½ķ
    0.16
    éĻ¤éĿŀ
    0.16
    ä»»ä½ķ人
    0.14
    ä»»ä½ķä¸Ģ个
    0.12
    ä»»ä½ķæŶåĢĻ
    0.11
     anybody
    0.11
    ä¸Ŀ毫
    0.11
    unless
    0.10
    Activations Density 0.045%

    No Known Activations