© 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. 63193
    Prev
    Next
    INDEX
    Explanations

    says

    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
    主æĮģåı¬å¼Ģ
    -0.10
    ittel
    -0.10
    åĴĮå¹³
    -0.10
    ä¸ĢéĿ¢
    -0.09
    Õµ
    -0.09
    ç¦ĭ
    -0.09
     СегоднÑı
    -0.09
    NotificationCenter
    -0.09
    ç»ıèIJ¥æ´»åĬ¨
    -0.09
    PathComponent
    -0.09
    POSITIVE LOGITS
    unte
    0.10
    edly
    0.08
    ä¸Ĭåİ»
    0.08
    åį³ä½¿æĺ¯
    0.08
     spur
    0.08
    hell
    0.08
    yb
    0.08
    å°½
    0.08
    åĩ¡æĺ¯
    0.08
    greens
    0.07
    Activations Density 0.049%

    No Known Activations