© 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. 53432
    Prev
    Next
    INDEX
    Explanations

    personal pronouns and familial relationships

    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
    åĿĩ使åħ¶
    -0.08
    åĩ¤åĩ°
    -0.08
     Pathfinder
    -0.08
    ypi
    -0.08
    宥
    -0.08
    herits
    -0.08
    éģĵ
    -0.07
    空æ°Ķè´¨éĩı
    -0.07
    باÙĨ
    -0.07
    _associ
    -0.07
    POSITIVE LOGITS
    åĴĮä»ĸçļĦ
    0.21
    æĪijåĴĮ
    0.19
     и
    0.13
    ä¼ģä¸ļåĴĮ
    0.13
     vÃł
    0.13
    à¹ģละ
    0.13
    ê³¼
    0.13
     and
    0.13
    ä¼ļåIJĮ
    0.13
    åħļåĴĮ
    0.13
    Activations Density 0.222%

    No Known Activations