© 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. 31754
    Prev
    Next
    INDEX
    Explanations

    everything from

    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
    个åĪ«
    -0.11
    è¾ĥå¤ļ
    -0.09
    TAB
    -0.09
    è¾ĥå·®
    -0.08
    ä¹ĭ声
    -0.08
     trunc
    -0.08
    agen
    -0.08
    æĻ·
    -0.08
    ament
    -0.08
    llib
    -0.08
    POSITIVE LOGITS
     ranging
    0.17
    ä»İå°ı
    0.13
    æĹ¢æľī
    0.13
    èĩ³
    0.12
     ranged
    0.12
    ä¸Ģ缴åĪ°
    0.12
    æĹłè®ºæĺ¯
    0.12
    ë¶ĢíĦ°
    0.11
    ä¸į管æĺ¯
    0.11
    åĮħæĭ¬
    0.11
    Activations Density 0.269%

    No Known Activations