© 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. 33648
    Prev
    Next
    INDEX
    Explanations

    London, MacKay, DR, gastric

    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
    骥
    -0.10
    人åĿĩ
    -0.09
    rena
    -0.09
    ihan
    -0.09
    忱
    -0.09
    arching
    -0.09
    ÑĨиÑĤ
    -0.09
    æĿª
    -0.08
    èĢ
    -0.08
    æķŀ
    -0.08
    POSITIVE LOGITS
    ä»Ģä¹Īåľ°æĸ¹
    0.09
    丢
    0.09
    å¸®ä½ł
    0.09
     moments
    0.09
    æĿ¥èĩª
    0.09
    æĪijä¸Ģ缴
    0.08
    ä½ķãģĭ
    0.08
    åIJįåı«
    0.08
    æĿ¥èĩªäºİ
    0.08
    éļıæŶ
    0.08
    Activations Density 0.135%

    No Known Activations