© 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. 20442
    Prev
    Next
    INDEX
    Explanations

    Slavic words with 'je' or 'e'

    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
     herself
    -0.11
    heimer
    -0.09
    utas
    -0.09
    undry
    -0.09
    æīĭä¸Ń
    -0.09
    rox
    -0.09
    posed
    -0.09
    她æĺ¯
    -0.09
    æľ¬äºº
    -0.09
    pen
    -0.09
    POSITIVE LOGITS
     koje
    0.10
     коÑĤоÑĢое
    0.10
     zosta
    0.09
    èĥĮåıĽ
    0.09
    ÑĪки
    0.09
    ä¸īæĸ¹
    0.09
     gÅĤówne
    0.09
    ä¸ĢåįĬ
    0.09
    ÑĪло
    0.08
    ition
    0.08
    Activations Density 0.186%

    No Known Activations