© 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. 34560
    Prev
    Next
    INDEX
    Explanations

    are permitted

    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
     itself
    -0.11
    omers
    -0.10
    esc
    -0.10
    ses
    -0.09
    ran
    -0.09
    sets
    -0.09
    nore
    -0.09
    pery
    -0.09
    alien
    -0.09
    eltas
    -0.09
    POSITIVE LOGITS
     meisten
    0.12
     themselves
    0.12
    nt
    0.11
    ady
    0.11
    hip
    0.11
    ngine
    0.11
     Ñģами
    0.10
    iminal
    0.10
    /is
    0.10
    ibold
    0.10
    Activations Density 0.492%

    No Known Activations