© 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. 63637
    Prev
    Next
    INDEX
    Explanations

    do more/less/a lot/a 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
    åIJĦ个
    -0.09
    æ®ĭ
    -0.08
    æĽ´å¤ļä¿¡æģ¯
    -0.08
    .vol
    -0.08
    寥
    -0.08
    èįı
    -0.07
    éķ¿å¯¿
    -0.07
    篦
    -0.07
    æѤåĪ»
    -0.07
    cher
    -0.07
    POSITIVE LOGITS
    ä¸į代表
    0.10
    ä¹ĭ人
    0.09
    èĬŁ
    0.09
    ç»Īç»ĵ
    0.09
    çļĦä¸ĢéĿ¢
    0.09
    æĻºéĢł
    0.09
    Bru
    0.09
     Jah
    0.09
    客家
    0.09
    Builders
    0.09
    Activations Density 0.161%

    No Known Activations