© 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. 49916
    Prev
    Next
    INDEX
    Explanations

    evidenced by

    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
    lez
    -0.10
    	HX
    -0.09
    /fw
    -0.09
    ãģĦãĤīãģ£ãģĹãĤĥ
    -0.09
    inflate
    -0.09
    楣
    -0.09
    ç¾Ĭæ¯Ľ
    -0.09
     hum
    -0.09
    ibles
    -0.09
    dana
    -0.09
    POSITIVE LOGITS
    ä½ĵçİ°åľ¨
    0.09
    许å¤ļ
    0.09
    วล
    0.09
    aled
    0.09
    2
    0.09
    ä¹İ
    0.08
    dv
    0.08
    èİ«è¿ĩäºİ
    0.08
     many
    0.08
    ä½Ĩä¸įéĻIJäºİ
    0.08
    Activations Density 0.131%

    No Known Activations