© 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. 62146
    Prev
    Next
    INDEX
    Explanations

    broadcasting segments

    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
     Recon
    -0.11
    æ¶Ĥ
    -0.10
    å¼ĢæºIJ
    -0.09
    PropertyParams
    -0.09
    дол
    -0.09
    -org
    -0.09
    /org
    -0.09
    å®Ĺ
    -0.09
    ervers
    -0.09
    对èĩªå·±
    -0.09
    POSITIVE LOGITS
    ç´łæĿIJ
    0.12
    é¢Ħæ¡Ī
    0.10
    段
    0.10
     CRE
    0.09
    ADVERTISEMENT
    0.09
    Segment
    0.08
    è°ı
    0.08
    åIJĮæľŁ
    0.08
    arend
    0.08
    ä¸ĵ家ç»Ħ
    0.08
    Activations Density 0.165%

    No Known Activations