© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 40026
    Prev
    Next
    INDEX
    Explanations

    author:

    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
     вн
    -0.09
    hal
    -0.09
    uilder
    -0.09
    èĩ³æĸ¼
    -0.09
     hal
    -0.08
    _hal
    -0.08
    alendar
    -0.08
    çĤ¼
    -0.08
    æݨ广åºĶçĶ¨
    -0.08
    å±¥
    -0.08
    POSITIVE LOGITS
     пÑĢаво
    0.09
    izers
    0.09
    شار
    0.09
    åľ°éĿ¢
    0.09
    (s
    0.08
    微信åı·
    0.08
    ophon
    0.08
    eturn
    0.08
     Moran
    0.08
    entine
    0.08
    Activations Density 0.040%

    No Known Activations