© 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. 33828
    Prev
    Next
    INDEX
    Explanations

    that's right

    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¸Ģéĥ¨
    -0.10
    iaux
    -0.10
    orest
    -0.09
    ledon
    -0.09
    ä¸ĢåĿĹ
    -0.09
    èĪĤ
    -0.09
    _stub
    -0.09
    âĻ¬
    -0.09
    åľ¨æł¡
    -0.09
    åĿļåĽº
    -0.09
    POSITIVE LOGITS
    è¿Ļåı¥è¯Ŀ
    0.11
    åIJİèĢħ
    0.11
     latter
    0.10
    åı¥è¯Ŀ
    0.08
    æļ®
    0.08
    æľĢåIJİä¸Ģ个
    0.08
    èıģ
    0.08
    è¿Ļè¯Ŀ
    0.08
    åºŃ
    0.08
     sentence
    0.08
    Activations Density 0.066%

    No Known Activations