© 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. 47689
    Prev
    Next
    INDEX
    Explanations

    perf or populate

    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
    è¨Ģè¡Į
    -0.10
    æį·
    -0.10
    建åĨĽ
    -0.10
    管å±Ģ
    -0.09
    FTA
    -0.09
    åĩºæĿ¥
    -0.09
    åľ¨å¤ĸ
    -0.09
    å¼±çĤ¹
    -0.09
    tsy
    -0.08
     Inst
    -0.08
    POSITIVE LOGITS
    åħħ足
    0.14
    æ»ĭ润
    0.13
     provision
    0.12
    åħħ足çļĦ
    0.11
    ä¾ĽåºĶ
    0.11
    åħħå®ŀ
    0.11
    åħħ
    0.10
    åħħè£ķ
    0.10
    ERE
    0.10
    åħħåĪĨ
    0.10
    Activations Density 0.174%

    No Known Activations