© 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. 23202
    Prev
    Next
    INDEX
    Explanations

    announcements and features

    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
    æľīä¿¡å¿ĥ
    -0.11
    被称为
    -0.11
    ä¹Łä¸įä¾ĭå¤ĸ
    -0.11
    享åıĹåĪ°
    -0.11
    åIJįåı«
    -0.10
    æĺ¯è°ģ
    -0.10
    åħ´èĩ´
    -0.10
    è°Ĵ
    -0.10
    çħ¦
    -0.10
    è¿Ļ座
    -0.09
    POSITIVE LOGITS
    æĹ¨åľ¨
    0.09
    è¾£
    0.09
    nox
    0.09
    鼷
    0.08
    诸
    0.08
     cracked
    0.08
     fancy
    0.08
    éĻ¢
    0.08
     lay
    0.08
    é«ĺæł¡
    0.08
    Activations Density 0.256%

    No Known Activations