© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8969
    Prev
    Next
    INDEX
    Explanations

    ux

    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
    è¡Įäºĭ
    -0.10
    é½ij
    -0.10
    äºĨä¸Ģåı£
    -0.10
    äºĨä¸Ģæī¹
    -0.10
    æĿĮ
    -0.10
    åĸı
    -0.09
    äºĨä¸Ģç³»åĪĹ
    -0.09
    ä¸įäºĨ
    -0.09
    æľ¬ç½ijç«Ļ
    -0.09
    æ·Ŀ
    -0.09
    POSITIVE LOGITS
    ะ
    0.09
    olders
    0.07
    onis
    0.07
    æīĢåľ¨çļĦ
    0.07
     ЧеÑĢ
    0.07
    ilated
    0.07
    oyal
    0.07
    abit
    0.07
    опеÑĢ
    0.07
    imum
    0.07
    Activations Density 0.013%

    No Known Activations