© 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. 2779
    Prev
    Next
    INDEX
    Explanations

    starting new sentences

    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
    è¿Ļ项
    -0.11
    çĦ¶å¾Į
    -0.11
    ä½Ĩæĺ¯ä»ĸ
    -0.10
    ä»İèĢĮ使
    -0.10
    æĺ¯ä¸Ģ项
    -0.10
    è¿Ļ对
    -0.10
    è¿Ļåľº
    -0.10
    å°±è¿Ļæł·
    -0.10
    ãģĿãģĹãģ¦
    -0.10
    è¿Ļä¹Ł
    -0.10
    POSITIVE LOGITS
    ä¸Ģæĸ¹éĿ¢
    0.14
    ä¸Ģæĺ¯
    0.13
    ä¸Ģæĸ¹éĿ¢æĺ¯
    0.13
    åħĪæĺ¯
    0.10
    :
    0.09
    ä¾ĭå¦Ĥ
    0.09
    é¦ĸåħĪæĺ¯
    0.08
    æ¯Ķå¦Ĥ
    0.08
    esp
    0.08
    ä¾ĭãģĪãģ°
    0.08
    Activations Density 0.215%

    No Known Activations