© 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. 32250
    Prev
    Next
    INDEX
    Explanations

    more than just

    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
    ä¸į说
    -0.11
    æĪijåıªæĺ¯
    -0.11
    ä¸įå±ŀäºİ
    -0.10
    é¢ł
    -0.09
    jaw
    -0.09
    ä¸įéĢĤåIJĪ
    -0.09
    æĪijå°±
    -0.08
    ä¸įåºĶ该
    -0.08
    åıªçŁ¥éģĵ
    -0.08
     Ñģли
    -0.08
    POSITIVE LOGITS
    æ¶īåıĬåĪ°
    0.12
    åħ¼
    0.12
    åħ¨æĸ¹ä½į
    0.10
    棹
    0.10
    åħ¼åħ·
    0.09
    æĽ´æ·±
    0.09
    åĨħ涵
    0.09
    å®ŀè´¨
    0.09
    è¿ĺåĮħæĭ¬
    0.09
     ÑģобÑģÑĤвенно
    0.09
    Activations Density 0.058%

    No Known Activations