© 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. 37045
    Prev
    Next
    INDEX
    Explanations

    reflexive pronouns doing something

    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
    çľ¼çľĭ
    -0.09
    示æĦı
    -0.08
    éĿĴ
    -0.08
     Helm
    -0.08
    èĩªä»İ
    -0.08
    بت
    -0.08
    è°ģçŁ¥éģĵ
    -0.08
    obra
    -0.08
    andler
    -0.08
    tog
    -0.08
    POSITIVE LOGITS
     needing
    0.10
    为ä¹ĭ
    0.08
    æī¿æĭħ
    0.08
    置身
    0.08
    潸
    0.08
    车ä¸Ĭ
    0.08
    å¤§éĽ¨
    0.08
     preg
    0.08
     bri
    0.08
    ç¿©
    0.08
    Activations Density 0.005%

    No Known Activations