© 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. 30886
    Prev
    Next
    INDEX
    Explanations

    heat and cooling towers

    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
    ç¼Ŀ
    -0.10
    åij¼åºĶ
    -0.09
    让人
    -0.08
    ibir
    -0.08
    æĶ¶åıĹ
    -0.08
    æĶ¾æīĭ
    -0.08
    æĿIJ
    -0.08
    åĪ³
    -0.08
    设ç«ĭäºĨ
    -0.08
    åĪ¶
    -0.08
    POSITIVE LOGITS
    å¡Ķ
    0.10
    åĨ·åį´
    0.10
    çĥŃ
    0.10
     thermal
    0.09
    æķ£çĥŃ
    0.09
    é¢Ľ
    0.09
    éªĨ
    0.09
     cooled
    0.09
     cooling
    0.09
    thermal
    0.09
    Activations Density 0.074%

    No Known Activations