© 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. 26179
    Prev
    Next
    INDEX
    Explanations

    for testing, educational, research, or internal use

    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
    ses
    -0.09
    ete
    -0.09
    gress
    -0.09
    对åºĶçļĦ
    -0.09
     Rena
    -0.09
    ç´§çĽ¯
    -0.09
    achten
    -0.09
    æIJ¦
    -0.09
    éĴĪ对
    -0.09
    lion
    -0.08
    POSITIVE LOGITS
     purposes
    0.21
    缮çļĦ
    0.13
     reasons
    0.11
     sake
    0.11
    çĶ¨éĢĶ
    0.11
    ä¸ºçĽ®çļĦ
    0.09
    广大群ä¼Ĺ
    0.09
     Moderator
    0.09
    çļĦè·¯ä¸Ĭ
    0.09
    è¿ĻæĶ¯
    0.09
    Activations Density 0.142%

    No Known Activations