© 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. 2038
    Prev
    Next
    INDEX
    Explanations

    French politicians and names

    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
     ÑģебÑı
    -0.10
     пÑĢедоÑģÑĤавлÑı
    -0.09
    dings
    -0.09
    夹
    -0.08
    gings
    -0.08
     Nay
    -0.08
    åĮ
    -0.08
    oldt
    -0.08
    涵
    -0.08
    ilst
    -0.08
    POSITIVE LOGITS
    çijĹ
    0.09
    éĨĴäºĨ
    0.09
    éħIJ
    0.09
    åĢĶ
    0.09
    ignite
    0.08
    ãĤĤãģ¡ãĤįãĤĵ
    0.08
    æľīæĿ¡ä»¶çļĦ
    0.08
    æ´²
    0.08
    éĨħ
    0.08
     edged
    0.08
    Activations Density 0.003%

    No Known Activations