© 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. 8661
    Prev
    Next
    INDEX
    Explanations

    molecular components

    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
    çļ´
    -0.10
     Alison
    -0.09
    åĺ²ç¬ij
    -0.09
    顶å±Ĥ
    -0.09
    exter
    -0.09
    å¼ĢéĺĶ
    -0.09
    luet
    -0.09
    ä¸ĩæĪ·
    -0.09
    åĻ±
    -0.08
    æ°¤
    -0.08
    POSITIVE LOGITS
    çļĦä¸Ģéĥ¨åĪĨ
    0.08
    æİ¥åĪ°
    0.08
    æľ
    0.08
    浯
    0.08
    éĺIJéĩĬ
    0.07
     copies
    0.07
    å¹¿æ³Ľ
    0.07
    оÑħÑĢан
    0.07
    ä¸ĵèģĮ
    0.07
     prote
    0.07
    Activations Density 1.304%

    No Known Activations