© 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. 5859
    Prev
    Next
    INDEX
    Explanations

    possessive pronoun followed by owning word

    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
    æģį
    -0.09
    ä¿¡å¿ĥ
    -0.09
    ä¸ĢçĻ¾
    -0.08
    DEST
    -0.08
     Bran
    -0.08
    tent
    -0.08
    à¸Ľà¸¥à¸²à¸¢
    -0.08
    缸åħ³
    -0.08
    ANK
    -0.08
     tentative
    -0.08
    POSITIVE LOGITS
    åľ¨æ¸¸æĪıä¸Ń
    0.10
     наз
    0.09
     sorts
    0.09
    åıĮæīĭ
    0.09
    ndata
    0.09
     course
    0.09
     cours
    0.08
    aÅĤ
    0.08
    ä½İä¸ĭ
    0.08
    åĭī
    0.08
    Activations Density 0.039%

    No Known Activations