© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 53231
    Prev
    Next
    INDEX
    Explanations

    throughout time and space

    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
    åı¯æĢľ
    -0.09
    ä¸ĢæĬĬ
    -0.09
    inem
    -0.09
    ä¸ĢæŶ
    -0.08
    åĩĨåħ¥
    -0.08
    æĪijçİ°åľ¨
    -0.08
    /cms
    -0.08
    linger
    -0.08
    burg
    -0.08
    æ¦Ľ
    -0.08
    POSITIVE LOGITS
     throughout
    0.15
    贯穿
    0.14
    Throughout
    0.13
    å§ĭç»Ī
    0.13
     Throughout
    0.12
    ä¿ĿæĮģçĿĢ
    0.11
    åľ¨æķ´ä¸ª
    0.11
    ä¸įæĸŃåľ°
    0.11
    åħħ满äºĨ
    0.10
    ãģļãģ£ãģ¨
    0.10
    Activations Density 0.034%

    No Known Activations