© 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
    2. Qwen3-32B
    3. 32-RESID-BATCHTOPK-65K
    4. 8457
    Prev
    Next
    INDEX
    Explanations

    articles and possessives precede nouns

    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
    浴室
    -0.10
    æµ£
    -0.09
    å¦Ĵ
    -0.09
    åĨħå¿ĥ
    -0.09
    空æ°Ķä¸Ń
    -0.09
    ä¸Ńç§ĭèĬĤ
    -0.09
    etty
    -0.09
    orro
    -0.08
    çݯå¢ĥä¸Ń
    -0.08
    (pe
    -0.08
    POSITIVE LOGITS
     draft
    0.09
     implements
    0.09
    æĶ¾åľ¨
    0.09
    è¯ģæĺİ
    0.09
     drafts
    0.08
    èµ°å¾Ĺ
    0.08
    è¯Ŀè¯ŃæĿĥ
    0.08
    éĴ¥åĮĻ
    0.08
    ä¸ĢåłĨ
    0.08
    yen
    0.08
    Activations Density 0.100%

    No Known Activations