© 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. 35689
    Prev
    Next
    INDEX
    Explanations

    then followed by setting

    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
    é¢ľ
    -0.10
     Ñĩел
    -0.09
    äºĶæľĪ
    -0.09
    æĸ°ä¸ī
    -0.08
    æľīæľŁ
    -0.08
    çļĦçľ¼
    -0.08
    æľīæķĪæľŁ
    -0.08
    ãĤ¦ãĥ³
    -0.08
    shed
    -0.08
    ä¸ĥæľĪ
    -0.08
    POSITIVE LOGITS
    è¿ĽèĢĮ
    0.12
    illion
    0.10
    aira
    0.09
    ãģ¤ãģı
    0.09
     THEN
    0.09
    éĤ£æł·çļĦ
    0.09
    æĺİæĺİ
    0.09
    achu
    0.09
    .habbo
    0.09
    æĬ¤
    0.08
    Activations Density 0.279%

    No Known Activations