© 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. 1782
    Prev
    Next
    INDEX
    Explanations

    complex models, how to, adults, studies of, high-end, edge clubs, patients, more than, threatening invasive

    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
    èķŀ
    -0.08
    nad
    -0.08
     Ñģек
    -0.08
     zwar
    -0.08
    æĪĸèĢħ说
    -0.08
    wnd
    -0.07
     promptly
    -0.07
    ãģ¨ãģ¨ãĤĤãģ«
    -0.07
    iles
    -0.07
    çī¹åĪ«æĺ¯
    -0.07
    POSITIVE LOGITS
    åĨįåĪ°
    0.22
    为æŃ¢
    0.16
    çĶļèĩ³æĺ¯
    0.15
     downright
    0.13
     ÙĪØŃتÙī
    0.12
    çĶļèĩ³
    0.12
    ä¹ĥèĩ³
    0.11
    çļĦåľ°æŃ¥
    0.11
     Ñģамого
    0.10
     ones
    0.10
    Activations Density 0.108%

    No Known Activations