© 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. 54004
    Prev
    Next
    INDEX
    Explanations

    bays and harbors

    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
    ä¸Ŀ
    -0.10
    çµ²
    -0.09
    ç«ĭ项
    -0.08
    ÑıÑĤ
    -0.08
    rollers
    -0.08
    uz
    -0.08
    æĿ¥è¿ĻéĩĮ
    -0.08
    æĿ¥åĪ°
    -0.08
     hy
    -0.08
    irie
    -0.08
    POSITIVE LOGITS
    深度èŀįåIJĪ
    0.10
    æ°ijèIJ¥ç»ıæµİ
    0.09
    wap
    0.09
    iphy
    0.09
    æµħ
    0.09
    åĸĩåıŃ
    0.09
    æ¹¾
    0.09
    ä¸Ģåı£
    0.09
    åĬŀ好
    0.09
    arde
    0.09
    Activations Density 0.053%

    No Known Activations