© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 35826
    Prev
    Next
    INDEX
    Explanations

    nerd or dork

    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
    æĥħåĨµæĿ¥çľĭ
    -0.10
    大纲
    -0.09
    orate
    -0.09
    åĿļå®ļ
    -0.09
    æĥħåĨµ
    -0.09
    大家éĥ½åľ¨
    -0.08
    让æĤ¨
    -0.08
    -generation
    -0.08
    è¹Ī
    -0.08
    ç¥ŀå·ŀ
    -0.08
    POSITIVE LOGITS
    é¹Ĭ
    0.10
    ishly
    0.09
    ivative
    0.09
    çķĮçļĦ
    0.09
    iras
    0.09
    ighest
    0.09
    è¿ĻçĤ¹
    0.09
    ific
    0.09
     Eisen
    0.09
    -corner
    0.09
    Activations Density 0.006%

    No Known Activations