© 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. 51849
    Prev
    Next
    INDEX
    Explanations

    winners and wins

    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
     capped
    -0.09
    éĢIJ
    -0.08
    erce
    -0.08
    éĢĤåIJĪèĩªå·±
    -0.08
    åĪĨ辨
    -0.07
     MAN
    -0.07
     itself
    -0.07
    å¦ĵ
    -0.07
    serve
    -0.07
    èģĶ
    -0.07
    POSITIVE LOGITS
    éļIJ身
    0.11
    缸åıį
    0.11
    bih
    0.09
    åĬ©æīĭ
    0.09
    çijľ
    0.09
    éĥ¢
    0.09
    ä¸įå±ij
    0.09
    天涯
    0.09
    æĮij
    0.09
    æݬ
    0.09
    Activations Density 0.021%

    No Known Activations