© 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. 5029
    Prev
    Next
    INDEX
    Explanations

    stop and realize

    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
    ç£Ķ
    -0.10
    ä½Ĩè¿Ļ
    -0.09
    åģıåģı
    -0.09
    èĹıçĿĢ
    -0.09
    bsp
    -0.09
    æ¯ķç«Ł
    -0.09
    éĵĨ
    -0.08
    åĬłæĮģ
    -0.08
    ä¸Ĭåİ»
    -0.08
    é¥Ńèıľ
    -0.08
    POSITIVE LOGITS
    æĴĩ
    0.09
    顾
    0.09
    é»ĺ
    0.09
    æħ¢
    0.09
     slowed
    0.09
    è°Ľ
    0.09
    大åѦæ¯ķä¸ļ
    0.09
    è¿ĻäºĽå¹´
    0.08
    ÑģÑı
    0.08
    çľĭä¸Ģçľĭ
    0.08
    Activations Density 0.005%

    No Known Activations