© 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. 15487
    Prev
    Next
    INDEX
    Explanations

    religion, raids, Timeline, helping, direct, electric

    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
     sẵ
    -0.18
    çľĭçĤ¹
    -0.17
    ä¹ŁåĽłæѤ
    -0.17
     Conan
    -0.17
    Reusable
    -0.16
    è°ĥæŁ¥æĺ¾ç¤º
    -0.16
    oreach
    -0.16
    äºĨèµ·æĿ¥
    -0.16
    两èĢħ
    -0.16
    ultiple
    -0.15
    POSITIVE LOGITS
    讫
    0.21
    çļĦ人æĿ¥è¯´
    0.20
    éĤ£ä¹Īå¤ļ
    0.20
    èĢĮ导èĩ´
    0.19
    çŃīæĸ¹éĿ¢çļĦ
    0.19
    åĪ°åºķæĺ¯
    0.19
     supposedly
    0.19
    æł¹æľ¬ä¸į
    0.18
    /how
    0.18
    为主é¢ĺçļĦ
    0.18
    Activations Density 3.142%

    No Known Activations