© 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. 624
    Prev
    Next
    INDEX
    Explanations

    section headers and lists

    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
    Negative Logits
    æĦıåij³çĿĢ
    -0.10
    æĸ¹å¼ıè¿Ľè¡Į
    -0.10
    æ·±åħ¥äººå¿ĥ
    -0.10
    ä¸į让
    -0.10
    æ²Į
    -0.09
    çŃ¹å»º
    -0.09
    ÑģÑĤо
    -0.09
    åĮĸè¿Ľç¨ĭ
    -0.09
    责令
    -0.08
    ARING
    -0.08
    POSITIVE LOGITS
     bic
    0.08
    ä¸į说
    0.08
    inker
    0.08
    lops
    0.08
    رÙģ
    0.08
    bine
    0.08
    ponsors
    0.07
    å¦Ĥæŀľè¯´
    0.07
    ardin
    0.07
    à®ļ
    0.07
    Activations Density 0.285%

    No Known Activations