© 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. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 17790
    Prev
    Next
    INDEX
    Explanations

    sat, sit, stood

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/qwen-3.5-saes/qwen-3.5-4b
    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
    ablo
    -0.06
    azzi
    -0.06
    ipro
    -0.05
    uthor
    -0.05
    çªĿ
    -0.05
    ä½Ĩæĺ¯è¿Ļ个
    -0.05
    onet
    -0.05
    alm
    -0.05
    voli
    -0.05
    adou
    -0.05
    POSITIVE LOGITS
    ÙĩÙħ
    0.07
    åĽ¢çļĦ
    0.06
    éĹ´
    0.06
    åĽ¢
    0.06
    å·²æľī
    0.06
    æĸ¹çļĦ
    0.06
     ÑĢабоÑĤаем
    0.06
     ÙĩÙħÛĮÙĨ
    0.06
    coat
    0.06
    edral
    0.05
    Activations Density 0.001%

    No Known Activations