© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 26935
    Prev
    Next
    INDEX
    Explanations

    offensive line

    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
    ivery
    -0.12
    ìĦ¤
    -0.10
    æĦŁ
    -0.09
    æĢ§å¼º
    -0.09
    uard
    -0.09
    ungs
    -0.09
    .fn
    -0.09
    æģį
    -0.09
    bis
    -0.09
    æľ¬äºº
    -0.09
    POSITIVE LOGITS
    ehr
    0.09
    è¡®
    0.09
    æĸ¹éĿ¢
    0.09
     tr
    0.08
     me
    0.08
     department
    0.08
    æĸ¹éĿ¢çļĦ
    0.08
    åĬĽéĩı
    0.08
    ä¸ĸçķĮ第ä¸Ģ
    0.08
     tun
    0.07
    Activations Density 0.012%

    No Known Activations