© 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. 48595
    Prev
    Next
    INDEX
    Explanations

    spokesperson or communications officer

    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
    æĹłç¼ĺ
    -0.10
    é£İä¿Ĺ
    -0.10
    æ®ĺ
    -0.09
    æľĪæľ«
    -0.09
    .Enc
    -0.09
    åħ¬å¯ĵ
    -0.09
    [OF
    -0.09
    wake
    -0.09
    æ®ĭ
    -0.09
    éĻįä½İæĪIJæľ¬
    -0.09
    POSITIVE LOGITS
    å®£ä¼ł
    0.16
    åªĴä½ĵ
    0.11
    (PR
    0.11
    èĪĨæĥħ
    0.11
     PR
    0.11
    ä¼łæĴŃ
    0.11
    å®£ä¼łæ´»åĬ¨
    0.11
    åħ¬åħ³
    0.11
    å®£ä¼łæķĻèĤ²
    0.11
     press
    0.11
    Activations Density 0.141%

    No Known Activations