© 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. 7965
    Prev
    Next
    INDEX
    Explanations

    head over to

    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
    ä¸Ģäºĭ
    -0.09
    PIC
    -0.09
    çĻ»åľº
    -0.08
    ä¸ī大
    -0.08
    两大
    -0.08
    è·Łä¸įä¸Ĭ
    -0.08
     reb
    -0.08
     коÑĤоÑĢого
    -0.08
     trailing
    -0.08
    zon
    -0.08
    POSITIVE LOGITS
    è¿ĩæĿ¥
    0.11
    éĹ®éĹ®
    0.11
    éłĺ
    0.10
    é¢Ĩ
    0.10
    æĻĴ
    0.10
    æľīä¸Ģä½į
    0.10
    æīĵæŃ»
    0.09
    åĢĴæĺ¯
    0.09
    éĹ¨æ§Ľ
    0.09
    åİ»çľĭçľĭ
    0.09
    Activations Density 0.006%

    No Known Activations