© 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. 18355
    Prev
    Next
    INDEX
    Explanations

    Hong Kong names or Chinese words

    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
    æĸ¯åŁº
    -0.10
    .uni
    -0.10
    èįīåİŁ
    -0.10
    æ©Ľ
    -0.10
    èĭ«
    -0.10
    æĪIJçĨŁçļĦ
    -0.10
    åĢĴ
    -0.10
    é¦ķ
    -0.09
    çĦ¦
    -0.09
    æł¡åĽŃ
    -0.09
    POSITIVE LOGITS
    粤
    0.09
    港澳
    0.09
    æĹłåĬĽ
    0.09
    ä¸Ģåı£
    0.09
    æ½®
    0.09
     Ñĩа
    0.08
     Hong
    0.08
    æĢ¡
    0.08
    åĬŁå¤«
    0.08
    åŁł
    0.08
    Activations Density 0.067%

    No Known Activations