© 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. 10832
    Prev
    Next
    INDEX
    Explanations

    vs. opponent names

    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
    æĹģ
    -0.09
     Maid
    -0.09
    åĤį
    -0.08
    èĢģ大
    -0.08
    媪
    -0.08
    第ä¸Ģéĥ¨
    -0.08
    çī¹åĪ«æĺ¯åľ¨
    -0.08
     Mafia
    -0.08
     compart
    -0.08
    ترت
    -0.08
    POSITIVE LOGITS
    æĿ¥èĩª
    0.13
    ä»Ģä¹Īåij¢
    0.12
    åIJĮæł·æĺ¯
    0.12
    æĬĹè¡¡
    0.12
    ç«ŀäºī对æīĭ
    0.12
    оваÑĢ
    0.11
    æĿ¥èĩªäºİ
    0.11
    æĮijæĪĺ
    0.10
     competing
    0.10
    æĮijè¡ħ
    0.10
    Activations Density 0.057%

    No Known Activations