© 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. 19262
    Prev
    Next
    INDEX
    Explanations

    document metadata or headers

    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
    详
    -0.11
    ç¹ģçIJIJ
    -0.11
    æĦıä¹īä¸ĬçļĦ
    -0.10
    ä¸ĵ访
    -0.09
    æĹłç§ģ
    -0.09
    詳ãģĹãģı
    -0.09
    å½±éŁ³
    -0.09
    èľĪ
    -0.09
    èѬ
    -0.09
    ippi
    -0.09
    POSITIVE LOGITS
    ä¸Ĭæľī
    0.09
    åħ³äºİ
    0.09
     Gauss
    0.09
    gang
    0.08
    babel
    0.08
    WAR
    0.07
     munch
    0.07
    æĪijè¦ģ
    0.07
     about
    0.07
     https
    0.07
    Activations Density 0.592%

    No Known Activations