© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19785
    Prev
    Next
    INDEX
    Explanations

    situation where

    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
    äºĨèĩªå·±çļĦ
    -0.10
    å¤ļä½ĻçļĦ
    -0.09
    æ°Ķ象
    -0.09
    æĦ¤æĢĴ
    -0.09
    çľĭ好
    -0.08
    çľ¼åīįçļĦ
    -0.08
    æĺ¯ä¸Ģ项
    -0.08
    /fixtures
    -0.08
    æ¯ķç«Łæĺ¯
    -0.08
    æĺ¯æľĢ好çļĦ
    -0.08
    POSITIVE LOGITS
     where
    0.16
    æŶåĢĻ
    0.16
    æĹ¶æľŁ
    0.15
    where
    0.14
     où
    0.13
    æĥħåĨµä¸ĭ
    0.13
     ØŃÙĬØ«
    0.12
     donde
    0.12
    ãģ«ãģĬãģĦãģ¦
    0.12
     hvor
    0.12
    Activations Density 0.249%

    No Known Activations