© 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. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 14761
    Prev
    Next
    INDEX
    Explanations

    Japanese and Chinese characters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-14b-saes/saes_Qwen_Qwen3-14B_batch_top_k/resid_post_layer_20
    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
     ÐIJлекÑģан
    -0.26
    tw
    -0.21
    _SPE
    -0.17
    æĮ«
    -0.17
    Bulletin
    -0.17
    éĢIJæ¸IJ
    -0.17
    裳
    -0.17
    geist
    -0.17
    ãģĵ
    -0.17
    æķĹ
    -0.17
    POSITIVE LOGITS
    é¾Ħ
    0.21
    åħ¨åŁŁ
    0.20
    ä½ıåĿĢ
    0.20
    æŀķ头
    0.19
    æ²¼
    0.18
    åķª
    0.18
     milfs
    0.18
    íĿĶ
    0.17
    pdev
    0.17
    èĬ±èįī
    0.17
    Activations Density 0.016%

    No Known Activations