© 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. 57058
    Prev
    Next
    INDEX
    Explanations

    book titles and descriptions

    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
    /or
    -0.09
    æĸĩ书
    -0.09
    人群ä¸Ń
    -0.09
    åī¥
    -0.09
    ège
    -0.09
    uding
    -0.08
    овиÑĩ
    -0.08
    cy
    -0.08
    ollar
    -0.08
    çķĭ
    -0.08
    POSITIVE LOGITS
    shelf
    0.16
    keeping
    0.15
    ends
    0.14
    lets
    0.12
    ç±į
    0.12
    elling
    0.11
    ellers
    0.11
    леÑĤ
    0.11
    æīī
    0.11
    marks
    0.10
    Activations Density 0.103%

    No Known Activations