© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9753
    Prev
    Next
    INDEX
    Explanations

    of

    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
    tre
    -0.10
    ÑĨенÑĤÑĢ
    -0.10
    åºĶä»ĺ
    -0.09
    ä½Ļé¢Ŀ
    -0.09
    anol
    -0.09
    家åĸ»æĪ·
    -0.09
    wcs
    -0.09
     ка
    -0.08
    åıĹä¸įäºĨ
    -0.08
    èĻİ
    -0.08
    POSITIVE LOGITS
    éĤ£äºº
    0.09
    pery
    0.08
    çļĦ主é¢ĺ
    0.08
    èĢķèĢĺ
    0.08
    ness
    0.08
    æĿ¥å®¾
    0.08
    ä¸İæŶ
    0.08
    æ¾¹
    0.08
    æĺ¥é£İ
    0.08
    ä¹ĭä¸Ĭ
    0.08
    Activations Density 0.326%

    No Known Activations