© 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. 32253
    Prev
    Next
    INDEX
    Explanations

    open to suggestions

    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
    注æĦıåĬĽ
    -0.09
    ä»İä¸ļ
    -0.09
    ä¸Ńå¿ĥåŁİå¸Ĥ
    -0.09
    åįĥæĸ¹çĻ¾è®¡
    -0.08
    ailing
    -0.08
    Binder
    -0.08
    aginator
    -0.08
    ste
    -0.08
    æŀľå®ŀ
    -0.08
    agara
    -0.08
    POSITIVE LOGITS
    æİ¥åıĹ
    0.10
    æĸ°æĬĢæľ¯
    0.10
    åħĪè¿ĽæĬĢæľ¯
    0.10
    fullscreen
    0.10
    åĴĮæĶ¯æĮģ
    0.09
    樵
    0.09
    ä¸Ģçľĭ
    0.09
    éĻĮçĶŁ
    0.09
    æİ¥åıĹäºĨ
    0.09
    缴è¨Ģ
    0.09
    Activations Density 0.014%

    No Known Activations