© 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. 18811
    Prev
    Next
    INDEX
    Explanations

    killer spy, killer whales, killer stalking

    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
    een
    -0.10
    wise
    -0.10
    åIJ¹
    -0.10
    ÙĨÙħ
    -0.09
    alk
    -0.09
    æ¹¾
    -0.08
    è¿Ļè¾¹
    -0.08
    èħĨ
    -0.08
    ooks
    -0.08
    ãĤ¥
    -0.08
    POSITIVE LOGITS
    éĶı
    0.13
     Whip
    0.12
    lien
    0.11
    欲
    0.10
    æĹłæ¯Ķ
    0.10
    оÑģÑĮ
    0.10
    çļĦä¿¡å¿ĥ
    0.09
    天使
    0.09
    mime
    0.09
    æ¦ľ
    0.09
    Activations Density 0.009%

    No Known Activations