© 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. 62869
    Prev
    Next
    INDEX
    Explanations

    type, form, stage, level

    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
    æł·åŃIJ
    -0.10
    带æĿ¥æĽ´å¤ļ
    -0.09
     Pers
    -0.09
    带æĿ¥çļĦ
    -0.09
    ä¾
    -0.08
    身ä¸Ĭ
    -0.08
    çļĦæĢģ度
    -0.08
    ä¹ĭ举
    -0.08
    姿
    -0.08
    ARB
    -0.08
    POSITIVE LOGITS
    à¸Ĥà¸Ńà¸ĩà¸ģาร
    0.10
    ä¸Ģ个æĸ°çļĦ
    0.09
    ombat
    0.09
    oga
    0.09
    Ware
    0.09
    ä¸Ģå¤Ħ
    0.08
    缦
    0.08
    Stars
    0.08
    ume
    0.08
    ä¼ģä¸ļæıIJä¾Ľ
    0.08
    Activations Density 0.237%

    No Known Activations