© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 32471
    Prev
    Next
    INDEX
    Explanations

    punctuation or sentence start

    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
    解æĶ¾æĢĿæĥ³
    -0.13
    æĪĸå¤ļæĪĸå°ij
    -0.09
    津贴
    -0.09
    æĥħåĨµæĿ¥çľĭ
    -0.09
    ä¸ĭä¸Ģ代
    -0.09
    åĽłä¸ºæĪij们
    -0.09
    çļĦåħ³æ³¨
    -0.09
    å¾Īå¤ļæľĭåıĭ
    -0.09
    åĬ©éķ¿
    -0.09
     sek
    -0.08
    POSITIVE LOGITS
    èĢĮä¸Ķ
    0.09
    ãģĹãģĭãĤĤ
    0.08
    並ä¸Ķ
    0.08
    ä¸Ķ
    0.08
     Crane
    0.08
    æį
    0.08
    енÑĮ
    0.08
    é©»
    0.08
     echt
    0.08
    åIJ¬èµ·æĿ¥
    0.08
    Activations Density 0.186%

    No Known Activations