© 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. 29979
    Prev
    Next
    INDEX
    Explanations

    from experts and insiders

    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
     ëĤ
    -0.09
    ään
    -0.08
    (dom
    -0.08
     вним
    -0.08
    arez
    -0.08
    orp
    -0.08
    è®®
    -0.08
    æĦı
    -0.08
    åĨĴ
    -0.08
     ChÃŃnh
    -0.08
    POSITIVE LOGITS
    åIJ¬åIJ¬
    0.11
    è§Ĩè§Ĵ
    0.11
    æİ¥åıĹéĩĩ访
    0.10
    åıĹ访èĢħ
    0.10
    è§ĴèIJ½
    0.10
    æIJŃ
    0.09
    âĬĸ
    0.09
    åıĹ访
    0.09
    æĪª
    0.09
    é²ľæ´»
    0.09
    Activations Density 0.094%

    No Known Activations