© 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. 20899
    Prev
    Next
    INDEX
    Explanations

    of the

    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
    é¾ĻéŨ
    -0.10
    åĭī
    -0.09
    æ¯ı人
    -0.09
    éļĶå£ģ
    -0.09
    åĨĴ
    -0.09
    æĦıä¹īä¸ĬçļĦ
    -0.09
    à¹Īà¸Ļ
    -0.09
    æĺİæĹ¥
    -0.09
    åįģåĪĨéĴŁ
    -0.09
    çĭIJ
    -0.09
    POSITIVE LOGITS
    饺åŃIJ
    0.08
    jon
    0.08
    contre
    0.08
    丼
    0.08
    itten
    0.08
    é«ĺåĵģè´¨
    0.08
    loit
    0.08
    鸢
    0.08
    oter
    0.08
    breadcrumb
    0.08
    Activations Density 0.119%

    No Known Activations