© 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. 34061
    Prev
    Next
    INDEX
    Explanations

    objects and materials

    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
    è§ģè¯ģäºĨ
    -0.11
    è¿Ļä¸ĢåĪ»
    -0.10
    æĿ¥çľĭçľĭ
    -0.10
    ä»Ģä¹ĪåİŁåĽł
    -0.10
    åį´ä¸į
    -0.09
    å°±æĺ¯ä¸ºäºĨ
    -0.09
    æĹ©æĹ©
    -0.09
    ä¸įä»ħå¦ĤæѤ
    -0.09
    ä½ľä¸ºä¸ĢåIJį
    -0.09
    èĢĮçİ°åľ¨
    -0.09
    POSITIVE LOGITS
    è¦ģ注æĦı
    0.10
    adoo
    0.09
     Ago
    0.09
    ä¸įåı¯ä»¥
    0.09
    失信
    0.09
    ä¸ĢèĪ¬æĺ¯
    0.09
     Canter
    0.09
    INED
    0.09
    æ¡´
    0.09
    ä¸ĢèĪ¬
    0.08
    Activations Density 0.185%

    No Known Activations