© 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. 33423
    Prev
    Next
    INDEX
    Explanations

    .org suffix

    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
    ewire
    -0.11
    海滩
    -0.10
    çѵ
    -0.10
    itta
    -0.09
    夷
    -0.09
    /jav
    -0.09
    汶
    -0.09
    afil
    -0.09
    herits
    -0.09
    amen
    -0.09
    POSITIVE LOGITS
    .org
    0.14
    o
    0.14
    ãģĬ
    0.14
    ο
    0.13
    о
    0.12
    on
    0.12
    os
    0.12
    oh
    0.12
    ï½ı
    0.12
    us
    0.12
    Activations Density 0.475%

    No Known Activations