© 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. 2481
    Prev
    Next
    INDEX
    Explanations

    unexpected outcomes and defying odds

    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¸Ģå¦ĤæĹ¢å¾Ģ
    -0.11
    èIJ½ãģ¡çĿĢ
    -0.10
    ä¸ĩçī©
    -0.09
    çľ¼ç¥ŀ
    -0.09
    æľªæĿ¥åıijå±ķ
    -0.09
     Ops
    -0.09
    缺å¸Ń
    -0.08
    èĩ»
    -0.08
    å·¥ä½ľä¼ļè®®
    -0.08
    _future
    -0.08
    POSITIVE LOGITS
    éĢĨè¢Ń
    0.20
    å¥ĩ迹
    0.14
     unexpected
    0.14
    unexpected
    0.13
    Unexpected
    0.12
    æīĵçł´
    0.12
    éĢĨ转
    0.12
    æīĵçł´äºĨ
    0.12
    æĦıå¤ĸ
    0.12
    é¢łè¦Ĩ
    0.12
    Activations Density 0.081%

    No Known Activations