© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9482
    Prev
    Next
    INDEX
    Explanations

    version of

    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
    åħ¶æīĢ
    -0.11
    hra
    -0.11
     notion
    -0.10
    缸åħ³
    -0.10
    ãģªãĤĤãģ®
    -0.10
    缸åħ³ä¿¡æģ¯
    -0.10
    大家ä¸Ģèµ·
    -0.09
    scape
    -0.09
    è¿Ļæł·çļĦè¯Ŀ
    -0.09
    .fun
    -0.09
    POSITIVE LOGITS
     ×ķש׾
    0.15
     sorts
    0.10
     ilk
    0.10
     ÙĪعÙĨ
    0.10
     utmost
    0.09
    御
    0.08
     пÑĢоиÑģ
    0.08
    åľ¨æĪij
    0.08
    à¹ģละà¸Ħวาม
    0.08
     persuasion
    0.08
    Activations Density 3.022%

    No Known Activations