© 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. 2569
    Prev
    Next
    INDEX
    Explanations

    beloved child, pet, hero, stars

    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
    佬
    -0.10
    aris
    -0.10
    æ¯Ķäºļ
    -0.10
    رÙĬÙģ
    -0.10
     bitten
    -0.09
    äºĨä¸Ģ份
    -0.09
    oka
    -0.09
    bian
    -0.09
    äºĮèĢħ
    -0.09
    åĽŀäºĭ
    -0.09
    POSITIVE LOGITS
    çĸĻ
    0.11
    çļĦæĸ°
    0.09
    çµ®
    0.09
    ä¸Ģè´¯
    0.09
    enso
    0.09
    èĴ¸
    0.09
     tắc
    0.08
    çŀ©çĽ®
    0.08
    ement
    0.08
    çļĦä»·æł¼
    0.08
    Activations Density 0.004%

    No Known Activations