© 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. 63931
    Prev
    Next
    INDEX
    Explanations

    from the

    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
    IRST
    -0.10
    åĸĪ
    -0.09
    minster
    -0.09
    Enumer
    -0.09
    hift
    -0.09
    æ¹į
    -0.09
    zell
    -0.09
     táºŃn
    -0.09
    åįĩ级æĶ¹éĢł
    -0.09
    ä¸įçĶĺ
    -0.09
    POSITIVE LOGITS
     list
    0.11
    ä¼Ĺå¤ļ
    0.10
    åĪĹ表
    0.10
    ä¼Ĺ
    0.10
    人群ä¸Ń
    0.10
     options
    0.10
    ä¼Ĺ人
    0.09
     among
    0.09
    è¿Ĥ
    0.09
    çľ¾
    0.09
    Activations Density 0.217%

    No Known Activations