© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 38182
    Prev
    Next
    INDEX
    Explanations

    for example

    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
    ãģ§ãģĻãģĮ
    -0.09
    -Smith
    -0.08
    rant
    -0.08
    “
    -0.08
    æ¸ħéĨĴ
    -0.08
     Elephant
    -0.08
    æľīåĩłä¸ª
    -0.07
    ке
    -0.07
    é¼ł
    -0.07
    å¯Ŀ
    -0.07
    POSITIVE LOGITS
    çħ§æł·
    0.10
    家ä¼Ļ
    0.09
    ä»ĸä¹Ł
    0.09
     há»ģ
    0.09
    ä»ĸä¼ļ
    0.09
    ä»ĸèĩªå·±
    0.08
    ä»ĸ们ä¼ļ
    0.08
    å®ŀäºĭ
    0.08
    ä¹Łå°±æĺ¯
    0.08
    éĤ£æł·çļĦ
    0.08
    Activations Density 6.706%

    No Known Activations