© 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. 29407
    Prev
    Next
    INDEX
    Explanations

    faster more frequent

    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
    ä¸įåIJĮç¨ĭ度
    -0.09
    åıĤä¸İèĢħ
    -0.09
    æijĩ
    -0.08
    ä½ķ度
    -0.08
    è·Ľ
    -0.08
    afort
    -0.08
    æİı
    -0.08
    éĵ©
    -0.08
    ä¸įæĸŃå¢ŀ强
    -0.08
    åĬ¨æijĩ
    -0.08
    POSITIVE LOGITS
    æľĢå¿«çļĦ
    0.21
    å¿«éĢŁ
    0.20
    æĽ´å¿«
    0.20
    å¿«
    0.19
    éĢŁåº¦å¿«
    0.19
    è¾ĥå¿«
    0.18
    尽快
    0.18
    å¿«æį·
    0.18
    æľĢå¿«
    0.17
     fast
    0.17
    Activations Density 0.241%

    No Known Activations