© 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. 20559
    Prev
    Next
    INDEX
    Explanations

    d- prefix or d capitalized 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
    own
    -0.12
    degree
    -0.11
    ump
    -0.11
    OfDay
    -0.11
    aked
    -0.10
    æ»´æ»´
    -0.10
    è°ł
    -0.10
    ද
    -0.10
    ¦
    -0.10
    дел
    -0.10
    POSITIVE LOGITS
    veloper
    0.13
    .parseDouble
    0.13
    stroy
    0.12
    artment
    0.12
    ughters
    0.12
    ãģĬãģı
    0.12
    çĶŁäº§æĢ»åĢ¼
    0.12
    ìĦ¯
    0.11
    iating
    0.11
    ials
    0.11
    Activations Density 2.665%

    No Known Activations