© 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. 21195
    Prev
    Next
    INDEX
    Explanations

    placement

    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
    亲
    -0.09
    rig
    -0.09
    çĶŁæ¶¯
    -0.09
    亲èĩª
    -0.09
    rena
    -0.09
    éĨį
    -0.08
    åĿij
    -0.08
    é©°
    -0.08
    åķľ
    -0.08
    æĿµ
    -0.08
    POSITIVE LOGITS
    åģļä¸Ģ个
    0.09
    Overlap
    0.09
    енд
    0.09
     Tape
    0.09
    æķ´å½¢
    0.08
     кÑĢай
    0.08
    ¼
    0.08
     tape
    0.08
    icle
    0.08
    tape
    0.08
    Activations Density 0.174%

    No Known Activations