© 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. 8898
    Prev
    Next
    INDEX
    Explanations

    locations of places

    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
     rub
    -0.09
    rule
    -0.09
    (rule
    -0.09
    æ¯ı次éĥ½
    -0.09
    achten
    -0.08
    åĴ¸
    -0.08
    å°±ä¸ļå²Ĺä½į
    -0.08
     rule
    -0.08
    contest
    -0.08
    èħ¥
    -0.08
    POSITIVE LOGITS
    trak
    0.11
    åĴ½åĸī
    0.10
    享æľī
    0.10
    æĶĺ
    0.09
    æĿĵ
    0.09
    ç¾İ丽çļĦ
    0.09
    éĿŀçī©è´¨
    0.08
    .components
    0.08
    illo
    0.08
    åİĨåı²æĤłä¹ħ
    0.08
    Activations Density 0.110%

    No Known Activations