© 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. 30793
    Prev
    Next
    INDEX
    Explanations

    I/we/you/they should

    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
    say
    -0.09
    ä½łä¼ļåıijçİ°
    -0.09
    ä½łä»¥ä¸º
    -0.08
    好ä¸į容æĺĵ
    -0.08
    并ä¸įä¼ļ
    -0.08
    åıĺæĪIJäºĨ
    -0.08
    æĺ¯ä¸įä¼ļ
    -0.08
    راÙĨ
    -0.08
    éĥ½ä¸įä¼ļ
    -0.08
    çĶ¨äºĨ
    -0.08
    POSITIVE LOGITS
    åºĶ该
    0.29
    ä¸įåºĶ该
    0.26
    ä¹ŁåºĶ该
    0.26
    åºĶå½ĵ
    0.25
    éĥ½åºĶ该
    0.25
     should
    0.25
    åºĶåľ¨
    0.24
    å°±åºĶ该
    0.23
    æĩī該
    0.23
    åºĶ
    0.22
    Activations Density 0.125%

    No Known Activations