© 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. 21740
    Prev
    Next
    INDEX
    Explanations

    module following definition or declaration

    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
    ness
    -0.11
    iable
    -0.10
    ely
    -0.10
    ech
    -0.10
    tings
    -0.09
    ened
    -0.09
    -0.09
    oon
    -0.09
    åIJĮæĹ¶ä¹Łæĺ¯
    -0.08
    ening
    -0.08
    POSITIVE LOGITS
    åĨħçļĦ
    0.13
    .exports
    0.12
    ulo
    0.12
     sclerosis
    0.11
    룬
    0.11
    ives
    0.11
    .forRoot
    0.11
    mers
    0.10
    ULO
    0.10
    åĮĸçļĦ
    0.10
    Activations Density 0.050%

    No Known Activations