© 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. 31960
    Prev
    Next
    INDEX
    Explanations

    defect or defects

    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
    ãģ£ãģįãĤĬ
    -0.11
    mit
    -0.10
    æ¦Ķ
    -0.09
    %@",
    -0.09
    ä¾
    -0.09
    hol
    -0.09
    ÙĬات
    -0.09
    ãģŁãģı
    -0.09
    rift
    -0.08
    ä¸ĢåĽŀäºĭ
    -0.08
    POSITIVE LOGITS
    ively
    0.16
    ives
    0.15
    iveness
    0.15
    swith
    0.13
    ivity
    0.12
    累累
    0.12
    ive
    0.11
    uate
    0.11
    ually
    0.11
    raction
    0.10
    Activations Density 0.047%

    No Known Activations