© 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. 16832
    Prev
    Next
    INDEX
    Explanations

    shortly after

    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
    åľī
    -0.10
    è¡®
    -0.10
    å¾Īæĥ³
    -0.09
    æĹ¶ä»£ä¸ŃåĽ½
    -0.09
    ulled
    -0.08
    éĿŀ常éĢĤåIJĪ
    -0.08
    佬
    -0.08
    erne
    -0.08
    éĿŀ常åĸľæ¬¢
    -0.08
     downside
    -0.08
    POSITIVE LOGITS
    ç½¢äºĨ
    0.10
     thôi
    0.10
    æĬ¬èµ·
    0.09
    æİ¥åĪ°
    0.09
    emente
    0.09
    è¸ıåħ¥
    0.09
    便åı¯
    0.09
    æĪĸå¤ļ
    0.09
    sheets
    0.09
    æĪIJç«ĭ以æĿ¥
    0.08
    Activations Density 0.017%

    No Known Activations