© 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. 9552
    Prev
    Next
    INDEX
    Explanations

    extending further

    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
    smith
    -0.10
    oen
    -0.09
    åı¦å¤ĸ
    -0.09
     sod
    -0.09
    ãģ¾ãģ¾
    -0.08
    åŃĽ
    -0.08
    åĭĥåĭĥ
    -0.08
     vej
    -0.08
    黥
    -0.07
    usal
    -0.07
    POSITIVE LOGITS
     [~,
    0.09
    åĵ¥ä»¬
    0.09
    andbox
    0.09
    SSIP
    0.09
    çļĦæŶéĹ´éĩĮ
    0.09
    تا
    0.08
    ìĦł
    0.08
     mum
    0.08
    çļĦæĹ¥åŃIJéĩĮ
    0.08
    æijŀ
    0.08
    Activations Density 0.010%

    No Known Activations