© 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. 30327
    Prev
    Next
    INDEX
    Explanations

    msdn.microsoft.com

    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
    _disp
    -0.09
    è¡·
    -0.09
    zeÅĽ
    -0.08
    ÑĪи
    -0.08
    uen
    -0.08
     disp
    -0.08
    为主é¢ĺçļĦ
    -0.08
    çĵ¯
    -0.08
    è¹Ń
    -0.08
    为主é¢ĺ
    -0.08
    POSITIVE LOGITS
    .microsoft
    0.11
    burgh
    0.10
    ä¸Ģ代
    0.10
    otron
    0.09
    ç¿ģ
    0.09
    ÑIJ
    0.09
    æĪIJé¾Ļ
    0.09
    indows
    0.09
    CUS
    0.09
    bing
    0.09
    Activations Density 0.004%

    No Known Activations