© 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. 59255
    Prev
    Next
    INDEX
    Explanations

    additional information regarding

    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
     dục
    -0.10
    oux
    -0.09
    Į¨
    -0.09
    æľĢç¾İçļĦ
    -0.09
    axe
    -0.09
    _ASC
    -0.09
    éļ°
    -0.09
    à¸Ĺรà¸ĩ
    -0.08
    .TODO
    -0.08
    .flip
    -0.08
    POSITIVE LOGITS
    é¢Ŀå¤ĸ
    0.15
    (extra
    0.12
    éĻĦåĬł
    0.12
     than
    0.12
     Extra
    0.12
     EXTRA
    0.11
    Extra
    0.10
     extra
    0.10
     Bonus
    0.10
     Additional
    0.10
    Activations Density 0.101%

    No Known Activations