© 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. 960
    Prev
    Next
    INDEX
    Explanations

    is a part of

    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
    ãģĤãĤĭç¨ĭ度
    -0.16
    ãģĤãĤĭ
    -0.14
    ä¸Ģæĺ¯
    -0.12
    åľ¨äº¬
    -0.12
    ä¸ĢäºĽ
    -0.11
    ä¸Ģ大æī¹
    -0.11
    ãģĵãģ¡ãĤī
    -0.11
    ä»ĬæĹ¥ãģ¯
    -0.11
    åIJĮæł·æĺ¯
    -0.11
     formerly
    -0.11
    POSITIVE LOGITS
    /w
    0.22
    /is
    0.21
    otope
    0.20
     ë¬¼ë¡ł
    0.15
    rael
    0.15
    nt
    0.15
    olation
    0.14
    olate
    0.14
    omorphic
    0.13
    ometric
    0.13
    Activations Density 2.047%

    No Known Activations