© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 1024
    Prev
    Next
    INDEX
    Explanations

    attention to detail

    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
    ema
    -0.10
    /render
    -0.09
    ÑĥÑĤ
    -0.09
     vital
    -0.09
     AUTHORS
    -0.08
     Flat
    -0.08
    too
    -0.08
    æĶ¯æĮģåĬĽåº¦
    -0.08
    IVAL
    -0.08
    çľĭä¸įåĩº
    -0.08
    POSITIVE LOGITS
    åĵģè´¨
    0.19
    ç²¾çĽĬæ±Ĥç²¾
    0.19
    è´¨éĩı
    0.18
    è´¨éĩıåĴĮ
    0.17
     quality
    0.16
    åĵģ質
    0.15
    質éĩı
    0.15
    çļĦè´¨éĩı
    0.14
     Quality
    0.14
    quality
    0.14
    Activations Density 0.104%

    No Known Activations