© 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. 51720
    Prev
    Next
    INDEX
    Explanations

    created or built

    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
    æĹłæīĢ
    -0.09
    ç©·
    -0.09
     hust
    -0.09
    æĸĩåĮĸåĴĮ
    -0.09
     haircut
    -0.09
    ampa
    -0.08
     Lesb
    -0.08
    彧
    -0.08
    ere
    -0.08
    oug
    -0.08
    POSITIVE LOGITS
    .opensource
    0.10
    imat
    0.09
    诱åıij
    0.09
    å®ŀè´¨æĢ§
    0.09
    æīĢéľĢè¦ģçļĦ
    0.09
     SAS
    0.08
    ä»ĺ
    0.08
    æľ¬æĿ¥
    0.08
    rase
    0.08
    RAR
    0.08
    Activations Density 0.479%

    No Known Activations