© 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. 53678
    Prev
    Next
    INDEX
    Explanations

    President's/adviser's/senators'/Obama's/King Juan Carlos University's/President Obama’s/Bush's/F.'s former adviser

    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
    ä¸Ńå¿ĥåŁİå¸Ĥ
    -0.10
    ordinator
    -0.10
    arda
    -0.09
    éĹªèĢĢ
    -0.09
    æĢ»çĽij
    -0.09
    çļĦèµĦéĩij
    -0.09
    ä¸ĢèĪ¬äºº
    -0.09
    第ä¸ĢåIJį
    -0.08
    骨头
    -0.08
    çĢ
    -0.08
    POSITIVE LOGITS
    亲èĩª
    0.10
    çĶŁæĹ¥
    0.10
    æī¹ç¤º
    0.09
    身边
    0.09
    çªĺ
    0.09
    oyo
    0.09
    俦
    0.09
    身边çļĦ
    0.08
    à¸Ĺรà¸ĩ
    0.08
    -appointed
    0.08
    Activations Density 0.237%

    No Known Activations