© 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. 55343
    Prev
    Next
    INDEX
    Explanations

    behaves differently

    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
    éĨª
    -0.08
    èŀįåħ¥
    -0.08
    æłı
    -0.08
    è¿©
    -0.08
    UGE
    -0.08
    éĨĴ
    -0.07
    æĦıå¤ĸ
    -0.07
    pta
    -0.07
    TypeEnum
    -0.07
    ãĥ¼ãĥł
    -0.07
    POSITIVE LOGITS
    å¦ĤæѤ
    0.27
    è¿Ļæł·
    0.27
    è¿Ļä¹Īåģļ
    0.26
    è¿Ļæł·åģļ
    0.26
    å°±æĺ¯è¿Ļæł·
    0.25
    æľīè¿Ļæł·çļĦ
    0.23
    ä¹Łæĺ¯å¦ĤæѤ
    0.22
    éĤ£æł·
    0.22
     ÑĤаким
    0.21
    éĢĻ樣
    0.21
    Activations Density 0.198%

    No Known Activations