© 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. 32241
    Prev
    Next
    INDEX
    Explanations

    should be, must be, accept

    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
    ä¹ĭæĺŁ
    -0.09
    æĥ¶
    -0.09
    å¼Ģè¾Ł
    -0.09
    oz
    -0.09
    åħ¨äº§ä¸ļéĵ¾
    -0.08
    ikon
    -0.08
    ä½Ĩä»ĸ们
    -0.08
    ì§ľ
    -0.08
    ramid
    -0.08
    å¼ĢèĬ±
    -0.08
    POSITIVE LOGITS
    å°½éĩı
    0.10
    éģµå®Ī
    0.10
    æľĢ好æĺ¯
    0.10
    ä¸įè¦ģ太
    0.09
    ç¨Ĺ
    0.09
    æľĢ好
    0.09
    æł¼å¼ı
    0.08
    æĭĴç»Ŀ
    0.08
     Tow
    0.08
    gers
    0.08
    Activations Density 0.086%

    No Known Activations