© 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. 11310
    Prev
    Next
    INDEX
    Explanations

    Last names and suffixes like -stein, -owitz

    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
     Baptist
    -0.09
     recap
    -0.09
    èĺĸ
    -0.09
    åŀ«
    -0.09
     pudd
    -0.09
    çľģéĴ±
    -0.08
    rega
    -0.08
    èĮ«çĦ¶
    -0.08
    çĬ¯
    -0.08
    åĽŀ头
    -0.08
    POSITIVE LOGITS
    çĬ¹
    0.15
     Jewish
    0.14
     Jews
    0.12
     jewish
    0.12
    owitz
    0.11
    çĮ¶
    0.11
    以èī²åĪĹ
    0.10
     kosher
    0.10
     Hebrew
    0.10
     Katz
    0.10
    Activations Density 0.180%

    No Known Activations