© 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. 32591
    Prev
    Next
    INDEX
    Explanations

    Imperial or emperor titles

    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
    abouts
    -0.12
     trÃŃ
    -0.10
     ditch
    -0.09
    manship
    -0.09
    olia
    -0.09
     ÑħозÑıйÑģÑĤв
    -0.09
    book
    -0.09
    橱
    -0.08
    aday
    -0.08
    finder
    -0.08
    POSITIVE LOGITS
    éĻĽä¸ĭ
    0.10
    污
    0.10
    ä¸Ģä¸ĸ
    0.09
    perial
    0.09
    istic
    0.09
    éĴ
    0.09
    izing
    0.09
    åĤ¬çĶŁ
    0.09
    éĿĵ
    0.09
    å«ļ
    0.09
    Activations Density 0.013%

    No Known Activations