© 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. 4292
    Prev
    Next
    INDEX
    Explanations

    tai chi and place names

    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
    nels
    -0.09
    à¹īว
    -0.08
    çħ
    -0.08
    们
    -0.08
    ail
    -0.08
     intervening
    -0.08
    nip
    -0.08
    ulin
    -0.08
    åĮĹ京æŶéĹ´
    -0.08
    hare
    -0.08
    POSITIVE LOGITS
    isser
    0.11
    -American
    0.10
    eners
    0.10
    añ
    0.09
    frame
    0.09
    ois
    0.09
    éĶ¢
    0.09
    ieran
    0.09
    enance
    0.09
    ìļĶ
    0.09
    Activations Density 0.198%

    No Known Activations