© 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. 48679
    Prev
    Next
    INDEX
    Explanations

    prevalence of

    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
    /current
    -0.10
    æĪijå¾Ī
    -0.09
    ä¸Ĭæµ·è¯ģåĪ¸
    -0.09
    æľīåı¯èĥ½
    -0.09
    Beer
    -0.09
    æ¯ı次éĥ½
    -0.09
    åķ¤éħĴ
    -0.08
    è¡Ģç³ĸ
    -0.08
    ustria
    -0.08
    atures
    -0.08
    POSITIVE LOGITS
    cape
    0.10
    è¿ŀ
    0.09
    å°±ä¸ļå²Ĺä½į
    0.09
    ison
    0.09
    éľĢè¦ģçĶ¨åĪ°
    0.09
    ä¸įåIJĪæł¼
    0.09
     ÑįÑĤим
    0.09
    æĬ¥éģĵ
    0.08
    éľĢè¦ģçĶ¨
    0.08
     cave
    0.08
    Activations Density 0.112%

    No Known Activations