© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52140
    Prev
    Next
    INDEX
    Explanations

    win or lose

    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
    ients
    -0.09
    abr
    -0.09
    æīĵå¼ĢäºĨ
    -0.09
    çī¹åĪ«æĺ¯åľ¨
    -0.09
    ç¥ŀèī²
    -0.08
    çľĭäºĨçľĭ
    -0.08
    姬
    -0.08
    èļĵ
    -0.08
    弩
    -0.08
    æĦŁè§ī
    -0.08
    POSITIVE LOGITS
    ä¹Łå¥½
    0.23
    ä¸İåIJ¦
    0.21
    ä¸į论
    0.12
    è¿ĺæĺ¯
    0.11
    ëĵł
    0.11
    ä¹Łä¸įä¾ĭå¤ĸ
    0.10
     lẫn
    0.10
    ä¸į管
    0.10
    ä¸į论æĺ¯
    0.10
    æĬij
    0.10
    Activations Density 0.101%

    No Known Activations