© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9748
    Prev
    Next
    INDEX
    Explanations

    percent ranges

    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
    æĢ§åĴĮ
    -0.12
    æĤ¬å´ĸ
    -0.09
    utsch
    -0.09
    仨
    -0.09
    æĦīå¿«
    -0.09
    æħĪæĤ²
    -0.09
    å·´æĭī
    -0.09
    chw
    -0.09
    京津
    -0.08
    å¿Ŀ
    -0.08
    POSITIVE LOGITS
     depending
    0.14
    depending
    0.13
    ãģ¨ãģĦãģ£ãģŁ
    0.11
    缸ç»ĵåIJĪ
    0.10
    å±ĬæŶ
    0.09
    çŁŃæľŁåĨħ
    0.09
     whichever
    0.09
    æŃ¢
    0.08
    ä¹ĭéĹ´
    0.08
    æĢ»ä¹ĭ
    0.08
    Activations Density 0.263%

    No Known Activations