© 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. 26418
    Prev
    Next
    INDEX
    Explanations

    information in parentheses or bold text

    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
    تÙĤ
    -0.10
    ause
    -0.09
    干货
    -0.08
    æĸ°åĵģ
    -0.08
    æŁIJä¸Ģ
    -0.08
    èį¯æĿIJ
    -0.08
    ä½IJ
    -0.08
     paternal
    -0.08
    ä¹IJåĻ¨
    -0.08
    ixe
    -0.08
    POSITIVE LOGITS
    æĭ¬
    0.13
    поÑĢ
    0.09
    è¿Ľè´§
    0.09
    çīµå¼ķ
    0.09
    çŀ¿
    0.09
    åĨĴçĿĢ
    0.08
    овиÑĩ
    0.08
    åĢĮ
    0.08
    çģ¯ç¬¼
    0.08
    æºħ
    0.08
    Activations Density 0.076%

    No Known Activations