© 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. 18913
    Prev
    Next
    INDEX
    Explanations

    learn the

    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
    Pri
    -0.08
    è°ģçŁ¥éģĵ
    -0.08
    Membership
    -0.08
    éĻ¢å£«
    -0.08
    Gov
    -0.08
    ãģĻãģ¹
    -0.08
    æĭĬ
    -0.08
     collaborations
    -0.08
    ÑĤиÑĢ
    -0.07
    erus
    -0.07
    POSITIVE LOGITS
    å¤ĸ交éĥ¨
    0.12
    åĽŀåĽ½
    0.11
    æĬ¤çħ§
    0.11
    çļĦä¸ĸçķĮ
    0.11
    æĸĩåĮĸ交æµģ
    0.11
    ä¸¤åĽ½
    0.10
    åIJĮèĥŀ
    0.10
    æ´¾éģ£
    0.10
     Lonely
    0.10
    å½ĵåľ°
    0.09
    Activations Density 0.031%

    No Known Activations