© 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. 47751
    Prev
    Next
    INDEX
    Explanations

    bit

    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
    èµ·æĿ¥
    -0.11
    ä¸įäºĨ
    -0.11
    ä¸ĢæŃ¥æŃ¥
    -0.10
    è¿Ļä¸ĢåĪĩ
    -0.10
    è¿Ļä¸ĢåĪ»
    -0.09
    çĴ¨
    -0.09
    ä¸ĢçĤ¹ä¹Łä¸į
    -0.09
    è¿Ļä¸ĢçĤ¹
    -0.08
    manship
    -0.08
    æŀģå°ij
    -0.08
    POSITIVE LOGITS
    /all
    0.11
    ìĶ©
    0.10
    ä¸įå°ıçļĦ
    0.10
    war
    0.09
    ÑĩиÑģлен
    0.09
    åħ¶ä»ĸçļĦ
    0.09
     dozen
    0.08
    RS
    0.08
    ä¸įéĶĻçļĦ
    0.08
    大çļĦ
    0.08
    Activations Density 0.020%

    No Known Activations