© 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. 63130
    Prev
    Next
    INDEX
    Explanations

    demo generator

    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
    çķĶ
    -0.10
    orus
    -0.09
    ness
    -0.08
     Blvd
    -0.08
    大éĩı
    -0.08
    onio
    -0.08
    ych
    -0.08
    <?=
    -0.08
    Quiet
    -0.08
    姿
    -0.08
    POSITIVE LOGITS
    stration
    0.16
    strate
    0.16
    ì¡´
    0.10
    ivre
    0.09
    /demo
    0.09
    elves
    0.09
    æĵŀ
    0.09
    pest
    0.09
    gee
    0.09
    åħ«åŃĹ
    0.09
    Activations Density 0.017%

    No Known Activations