© 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. 26396
    Prev
    Next
    INDEX
    Explanations

    contact list

    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
     slick
    -0.09
    /conf
    -0.09
     stiff
    -0.09
    æŁľåı°
    -0.09
    æ´¥
    -0.09
    åıijè¨Ģ人
    -0.09
     GlobalKey
    -0.08
    ĵåIJį
    -0.08
    stro
    -0.08
    Wiki
    -0.08
    POSITIVE LOGITS
    åIJįåįķ
    0.10
    contacts
    0.10
     contacts
    0.10
     list
    0.10
    å¿łå®ŀ
    0.10
    _contacts
    0.09
    åıijåĬ¨
    0.09
    åŃĺéĩı
    0.09
    失æݧ
    0.09
    .contacts
    0.09
    Activations Density 0.071%

    No Known Activations