© 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. 55945
    Prev
    Next
    INDEX
    Explanations

    framework components and core libraries

    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
    åĴĮæľįåĬ¡
    -0.08
    举æŃ¢
    -0.08
     wy
    -0.08
    åIJĦ个æĸ¹éĿ¢
    -0.08
    Season
    -0.08
    åĴĮåľ°åĮº
    -0.07
    éĢļçŁ¥ä¹¦
    -0.07
    ENUM
    -0.07
    /*------------------------------------------------
    -0.07
     Deluxe
    -0.07
    POSITIVE LOGITS
    è°ħ解
    0.10
    辩è¯ģ
    0.10
    å¦ĤæĦı
    0.09
    ìį¨
    0.09
    æŃ¤é¡¹
    0.09
    èĹķ
    0.09
    çī©è´¨
    0.09
    纯æ´ģ
    0.09
     полно
    0.09
     phá»iji
    0.08
    Activations Density 1.658%

    No Known Activations