© 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. 19738
    Prev
    Next
    INDEX
    Explanations

    focus or emphasis

    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
    ä¸ĩ人次
    -0.10
    åı¯ä»¥èİ·å¾Ĺ
    -0.09
    shal
    -0.09
    äºĽ
    -0.08
    çļĦæĸ¹å¼ıæĿ¥
    -0.08
    rels
    -0.08
     dáºŃy
    -0.08
    åľ¨åĨħçļĦ
    -0.08
    lex
    -0.08
    .rt
    -0.08
    POSITIVE LOGITS
     focus
    0.31
     emphasis
    0.26
    侧éĩį
    0.25
    èģļçĦ¦
    0.25
    focus
    0.25
    Focus
    0.25
    强è°ĥ
    0.24
     focuses
    0.23
     Focus
    0.23
    注éĩį
    0.23
    Activations Density 0.081%

    No Known Activations