© 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. 2636
    Prev
    Next
    INDEX
    Explanations

    being that

    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
    åİĨåı²æĸ°
    -0.10
    åıijå±ķçIJĨ念
    -0.10
    ToBounds
    -0.10
    èħĨ
    -0.10
    身边çļĦ
    -0.09
    å¾Īé«ĺ
    -0.09
    æĥ´
    -0.09
    è¿Ļ群
    -0.09
    ÙĪظ
    -0.09
    åįģ足
    -0.09
    POSITIVE LOGITS
     being
    0.16
    being
    0.13
    Being
    0.11
     sendo
    0.10
     Being
    0.10
    ä½¼
    0.09
     siendo
    0.09
     étant
    0.09
    -being
    0.09
    ä¹Łå°±æĺ¯
    0.09
    Activations Density 0.118%

    No Known Activations