© 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. 11968
    Prev
    Next
    INDEX
    Explanations

    easier than, easier to

    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
    为ä¹ĭ
    -0.11
    æ¼Ķ
    -0.09
    oda
    -0.09
    åĢľ
    -0.08
    ä¸Ńå¿ĥåŁİåĮº
    -0.08
     ê·¸ëłĩ
    -0.08
    å¦ĸæĢª
    -0.08
     LATIN
    -0.08
    人çĶŁçļĦ
    -0.08
    esso
    -0.08
    POSITIVE LOGITS
     dÃłng
    0.09
    andes
    0.09
    åĵį
    0.09
     ÃŃch
    0.08
    åĴĮåıijå±ķ
    0.08
    èĩĤ
    0.08
     nonexistent
    0.08
    -than
    0.08
    onz
    0.08
    udder
    0.08
    Activations Density 0.015%

    No Known Activations