© 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. 7227
    Prev
    Next
    INDEX
    Explanations

    grassley, burns, sale, attorney, box, chant, facility, ten

    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
    èİ«
    -0.10
    ä¹į
    -0.09
     Ferdinand
    -0.09
    ãģ¾ãģļ
    -0.08
    ä¼ĹæīĢ
    -0.08
    é¦ĸåħĪæĺ¯
    -0.08
    åģĥ
    -0.08
    ffa
    -0.08
    ä¾
    -0.08
    etto
    -0.08
    POSITIVE LOGITS
    åIJİèĢħ
    0.12
    å°ļ
    0.11
    ä¹ŁåĽłæѤ
    0.11
    è¿Ļä¹Łæĺ¯
    0.10
    è°Ľ
    0.10
    å®ĥ
    0.10
    让å®ĥ
    0.09
    å®ĥ们
    0.09
    èĩ³ä»Ĭ
    0.09
    æ°§æ°Ķ
    0.09
    Activations Density 0.067%

    No Known Activations