© 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. 5324
    Prev
    Next
    INDEX
    Explanations

    will be done

    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
    eland
    -0.10
    åĽĽå¤§
    -0.09
    åij½
    -0.09
    æľīæĿ¡ä»¶çļĦ
    -0.08
    wn
    -0.08
    缼
    -0.08
    ëĦ¤
    -0.08
    asa
    -0.08
    æĦŁè§īåĪ°
    -0.08
    åĽłåľ°
    -0.08
    POSITIVE LOGITS
    缴æݥ影åĵį
    0.11
    æĺ¯ä½ł
    0.09
    ign
    0.09
    æĹłå½¢
    0.09
    ä¼ļ被
    0.09
    éħ¬
    0.08
    çļĦæľºä¼ļ
    0.08
    ä¿Ŀ管
    0.08
    èĵĸ
    0.08
    马ä¸Ĭå°±
    0.08
    Activations Density 0.193%

    No Known Activations