© 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. 16767
    Prev
    Next
    INDEX
    Explanations

    asking questions

    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
    饯
    -0.11
    åĴ¬
    -0.10
    uria
    -0.09
    åijĬ
    -0.09
    沮丧
    -0.09
    æ©¡
    -0.09
    æĥ©
    -0.09
    ä¹Łä¸įæķ¢
    -0.09
    èµĻ
    -0.09
    é¢Ľ
    -0.09
    POSITIVE LOGITS
    éĹ®
    0.17
    æıIJéĹ®
    0.14
    åķı
    0.14
     вопÑĢоÑģ
    0.14
    éĹ®éģĵ
    0.14
     há»ıi
    0.14
     questions
    0.13
     question
    0.13
     asking
    0.12
    询éĹ®
    0.12
    Activations Density 0.079%

    No Known Activations