© 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. 40346
    Prev
    Next
    INDEX
    Explanations

    longer, more, new, increase, widely

    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
    eenth
    -0.11
    /../
    -0.10
    ableObject
    -0.09
    OfWork
    -0.09
    é¾Ħ
    -0.09
    è°¥
    -0.09
    ãĤĩãģĨ
    -0.09
    lessly
    -0.09
    éĩĩ访æŶ
    -0.09
    usan
    -0.09
    POSITIVE LOGITS
    (er
    0.08
    çºŃ
    0.07
    locale
    0.07
    ض
    0.07
    è°Ľ
    0.07
    /tty
    0.07
    -Sah
    0.06
    ä¸įä½Ĩ
    0.06
     pragma
    0.06
    \'
    0.06
    Activations Density 33.723%

    No Known Activations