© 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. 2530
    Prev
    Next
    INDEX
    Explanations

    Excel spreadsheets

    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
     scn
    -0.09
    ãģ¬
    -0.09
    ties
    -0.09
    gap
    -0.08
    笤
    -0.08
    é£Ł
    -0.08
    ë¥ł
    -0.08
    /proto
    -0.08
    ası
    -0.08
    mares
    -0.08
    POSITIVE LOGITS
    çĶµä»·
    0.10
    iled
    0.09
    带走
    0.09
    æĬ¬å¤´
    0.08
    çķ´
    0.08
    鬯
    0.08
    ä¸Ńæĸĩ
    0.08
     gro
    0.08
    erate
    0.08
     Natalie
    0.08
    Activations Density 0.198%

    No Known Activations