© 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. 8350
    Prev
    Next
    INDEX
    Explanations

    or and

    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
    深深çļĦ
    -0.10
    å°±æĦıåij³çĿĢ
    -0.09
    erp
    -0.09
    éĩįéĩį
    -0.09
    æŀŃ
    -0.09
    çľĩ
    -0.09
    çīµæĮĤ
    -0.09
    åıĪ称
    -0.09
    个åĪ«
    -0.09
    é¼ł
    -0.08
    POSITIVE LOGITS
     ones
    0.21
     Ones
    0.12
    ones
    0.12
    çļĦéĤ£ç§į
    0.12
     ours
    0.11
    ãģªãĤĤãģ®
    0.10
    ä¹Łå¥½
    0.09
     yours
    0.09
     denen
    0.09
    ãĤĤãģ®ãģ¯
    0.09
    Activations Density 0.176%

    No Known Activations