© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37703
    Prev
    Next
    INDEX
    Explanations

    cited or deemed precedent

    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
    (common
    -0.09
    æįĮ
    -0.09
    åĿļ强
    -0.09
    æ·¡
    -0.09
    å¿ĹæĦ¿
    -0.08
    kip
    -0.08
    伺
    -0.08
    aly
    -0.08
    ingu
    -0.08
     sust
    -0.08
    POSITIVE LOGITS
    æŀ¥
    0.10
    á»ĵ
    0.09
    ä»Ģä¹ĪæĦıæĢĿ
    0.09
    arden
    0.08
    èĵī
    0.08
    å·²æĺ¯
    0.08
    å¿
    0.08
    пад
    0.08
    èĭij
    0.08
    ModifiedDate
    0.08
    Activations Density 0.005%

    No Known Activations