© 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. 48970
    Prev
    Next
    INDEX
    Explanations

    age followed by a noun

    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
    ä¸Ńå°ıåŀĭ
    -0.12
    ä¸Ńå°ı
    -0.10
     Stanton
    -0.09
    çļĦéģĵçIJĨ
    -0.09
    çħĬ
    -0.09
    åIJĦç§įåIJĦæł·
    -0.09
    çīĪæĿĥ声æĺİ
    -0.09
     doÄŁru
    -0.08
    =com
    -0.08
    åĢ¼å¾Ĺä¸Ģ
    -0.08
    POSITIVE LOGITS
    èĻij
    0.10
    è¸ıåħ¥
    0.09
    è¸ıä¸Ĭ
    0.08
    ssi
    0.08
    ssel
    0.08
    çľī
    0.08
    cul
    0.08
    æĿı
    0.08
    igel
    0.08
    tons
    0.08
    Activations Density 0.018%

    No Known Activations