© 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. 22065
    Prev
    Next
    INDEX
    Explanations

    Bristol City

    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
    ÑĭÑĤа
    -0.09
    ise
    -0.09
    uate
    -0.09
    uating
    -0.09
    ãģıãĤĮ
    -0.09
    es
    -0.08
    颤
    -0.08
     dozen
    -0.08
    usted
    -0.08
    çĻ¸
    -0.08
    POSITIVE LOGITS
    shire
    0.10
    ogle
    0.09
    fully
    0.09
    ê¹IJ
    0.09
     rÃłng
    0.09
    عÙĦاÙħ
    0.09
    ìĿį
    0.09
    人çļĦ
    0.09
     Naughty
    0.08
    èįı
    0.08
    Activations Density 0.046%

    No Known Activations