© 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. 21079
    Prev
    Next
    INDEX
    Explanations

    regions in Asia and Europe

    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
    é«ĵ
    -0.09
    æ·³
    -0.08
    è¯Ŀ说
    -0.08
    å²IJ
    -0.08
    éĢĨè¡Į
    -0.08
    ä¸¤åľ°
    -0.08
    è§ģè¯Ĩ
    -0.08
    dì
    -0.08
    æ£Ģçĸ«
    -0.08
    allis
    -0.08
    POSITIVE LOGITS
    西äºļ
    0.14
    äºļæ´²
    0.13
    æĭīä¸ģ
    0.13
     Asia
    0.12
    举åįĹäºļ
    0.12
     Asian
    0.11
     asia
    0.11
    欧洲
    0.11
     Latin
    0.11
    Asian
    0.11
    Activations Density 0.203%

    No Known Activations