© 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. 2187
    Prev
    Next
    INDEX
    Explanations

    or without

    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
    åĴĮ社ä¼ļ
    -0.10
    ä¸įä»ħåı¯ä»¥
    -0.10
    ä¸įä»ħ
    -0.10
     Suns
    -0.09
    ä¸įä»ħè¦ģ
    -0.09
    åĪ«è¯´
    -0.09
    _plural
    -0.09
    OrCreate
    -0.09
    åĩłçĤ¹
    -0.08
    ä¸įå¾Ĺä¸į
    -0.08
    POSITIVE LOGITS
    çļĦ带é¢Ĩ
    0.09
     depending
    0.09
    æī¿æĭħ责任
    0.09
     prior
    0.09
    ä¸Ģ个好çļĦ
    0.09
    /non
    0.08
    羣å¿ĥ
    0.08
    羣è¯ļ
    0.08
    å½ĵå±Ģ
    0.08
    depending
    0.08
    Activations Density 0.073%

    No Known Activations