© 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. 6594
    Prev
    Next
    INDEX
    Explanations

    don't ask

    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
    æ·ĸ
    -0.09
    女æĸ¹
    -0.08
    åºĶåıĬæŶ
    -0.08
    å¾®åķĨ
    -0.08
     //~
    -0.08
    æijĺ
    -0.08
    æıIJé«ĺäºĨ
    -0.08
     inter
    -0.08
    梳
    -0.08
     stripper
    -0.08
    POSITIVE LOGITS
    讳
    0.13
    åĽŀéģ¿
    0.13
    鸵
    0.12
    模ç³Ĭ
    0.12
    以åħį
    0.11
    æ·¡åĮĸ
    0.11
    èĢ»
    0.11
    çļĦèī¯å¥½
    0.11
    æĹłçŁ¥
    0.11
    çļĦç¾İ好
    0.11
    Activations Density 0.080%

    No Known Activations