© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 38441
    Prev
    Next
    INDEX
    Explanations

    "say" followed by question/example

    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
     semiclassical
    -0.10
    sch
    -0.09
    åľ¨æĪijçľĭæĿ¥
    -0.09
    ëĿ¼ëıĦ
    -0.09
    validated
    -0.09
    çĶŁäº§æĢ»åĢ¼
    -0.09
    uring
    -0.09
    sw
    -0.08
    ´
    -0.08
    svg
    -0.08
    POSITIVE LOGITS
    ings
    0.11
     Äijua
    0.10
    ä¸ĢéŨ
    0.09
    yyyy
    0.09
    OUN
    0.09
    ields
    0.09
    things
    0.09
    ä¸įåħĭ
    0.09
    éŨåı£
    0.09
    åĽŀäºĭ
    0.09
    Activations Density 0.032%

    No Known Activations