© 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. 21281
    Prev
    Next
    INDEX
    Explanations

    personal details

    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
    atty
    -0.09
    综åIJĪä½ĵ
    -0.09
    ør
    -0.09
    çļĦéĴ±
    -0.08
    letic
    -0.08
    anca
    -0.08
    vell
    -0.08
    atted
    -0.08
    .argv
    -0.08
    ultipart
    -0.08
    POSITIVE LOGITS
    æĺ¯åIJ¦æľī
    0.11
    èģĶç³»æĸ¹å¼ı
    0.11
    å¿ħè¦ģçļĦ
    0.10
    å§ĵåIJį
    0.10
     location
    0.10
    æīĢå±ŀ
    0.10
    æīĢåľ¨åľ°
    0.10
    缸åħ³ä¿¡æģ¯
    0.10
    ç»ıèIJ¥èĮĥåĽ´
    0.09
     type
    0.09
    Activations Density 0.162%

    No Known Activations