© 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. 1791
    Prev
    Next
    INDEX
    Explanations

    store database information

    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
    æł¹æį®èĩªå·±
    -0.11
    嵯
    -0.11
    ä¼ĹçŃ¹
    -0.10
    æĬĬèĩªå·±
    -0.10
    让èĩªå·±
    -0.10
    TemplateName
    -0.09
    StackSize
    -0.09
    è¸ı
    -0.09
    为èĩªå·±
    -0.09
    éĢĤåIJĪèĩªå·±
    -0.09
    POSITIVE LOGITS
     data
    0.12
    æķ°æį®
    0.10
    æĤ²è§Ĥ
    0.10
    data
    0.10
     model
    0.09
    DB
    0.09
    å¤ĦçIJĨ
    0.09
     persisted
    0.09
    Data
    0.09
    åŃĺåĤ¨
    0.09
    Activations Density 1.565%

    No Known Activations