© 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. 32811
    Prev
    Next
    INDEX
    Explanations

    without consent or permission

    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
    åĢĻ
    -0.10
    ä¸įè¦ģ太
    -0.10
    (SIG
    -0.09
    IPS
    -0.09
    ä½ĵè´¨
    -0.09
    çĹ¹
    -0.09
    大åĬĽæĶ¯æĮģ
    -0.09
    ÑĩиÑģлен
    -0.08
    éĢĢ
    -0.08
    çľī
    -0.08
    POSITIVE LOGITS
    æľªç»ı
    0.17
    æľªç»ıæİĪæĿĥ
    0.12
    ç§ģèĩª
    0.11
    ä¸įç»ı
    0.11
    æĵħèĩª
    0.11
     unauthorized
    0.10
    without
    0.10
    ç§ģ
    0.10
    Unauthorized
    0.09
    ä¸įåIJĮæĦı
    0.09
    Activations Density 0.236%

    No Known Activations