© 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. 20560
    Prev
    Next
    INDEX
    Explanations

    official looking

    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
    .persistent
    -0.11
    çĽĬ
    -0.10
    åİ»è¿ĩ
    -0.10
    ç²Ĺ
    -0.10
     ÃŃch
    -0.09
    é¤IJ饮
    -0.09
    anon
    -0.09
    èľ¥
    -0.09
    asti
    -0.09
    纳æĸ¯
    -0.09
    POSITIVE LOGITS
    çľĭä¸Ĭåİ»
    0.27
    çľĭèµ·æĿ¥
    0.26
    åIJ¬èµ·æĿ¥
    0.23
    -looking
    0.22
    çľĭèµ·ä¾Ĩ
    0.20
    æł·åŃIJ
    0.20
     sounding
    0.17
    çļĦæł·åŃIJ
    0.16
     looking
    0.16
     looks
    0.15
    Activations Density 0.068%

    No Known Activations