© 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. 48840
    Prev
    Next
    INDEX
    Explanations

    I guess

    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
    åĪĩ
    -0.10
    ç»Ŀ对
    -0.09
    ä¼¼ä¹İ
    -0.09
    å¾Ģå¾Ģ
    -0.08
    éĥ¾
    -0.08
    ä¸įå½±åĵį
    -0.08
     ÑĢÑĥк
    -0.08
    åĵģ
    -0.08
    Earn
    -0.08
    aines
    -0.07
    POSITIVE LOGITS
    ä¹Łç®Ĺ
    0.11
    è°Īè°Ī
    0.10
    ç®Ĺæĺ¯
    0.10
    ä¹Łç®Ĺæĺ¯
    0.09
    ä¹Łæ²¡ä»Ģä¹Ī
    0.09
    ngr
    0.09
    æ¼Ĥ
    0.09
    åĿŀ
    0.09
    鹤
    0.09
    ä¿Ĺè¯Ŀ说
    0.09
    Activations Density 0.028%

    No Known Activations