© 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. 27128
    Prev
    Next
    INDEX
    Explanations

    suffix or word part

    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
    esc
    -0.09
    ãĤĤãģ£ãģ¨
    -0.08
    çĩļ
    -0.08
    媽
    -0.08
    ally
    -0.08
    è¾ĥ为
    -0.08
    èij£äºĭä¼ļ
    -0.08
     spit
    -0.07
    çļĦæĥħ
    -0.07
    çļĦ人çī©
    -0.07
    POSITIVE LOGITS
    ç«Łæĺ¯
    0.09
    çĶ¨åĵģ
    0.09
    ä¸įæİĴéĻ¤
    0.09
    äºĭå®ľ
    0.09
    DEV
    0.08
    ä¿Ł
    0.08
    éļ¾é¢ĺ
    0.08
    ä¸Ģ身
    0.08
    eworld
    0.08
    é«ĺåİŁ
    0.08
    Activations Density 1.014%

    No Known Activations