© 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. 18406
    Prev
    Next
    INDEX
    Explanations

    kept/keep getting

    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
    管çIJĨæĿ¡ä¾ĭ
    -0.10
    æĽ´å¤ļ
    -0.09
    ç¦ģæŃ¢
    -0.09
     beginnings
    -0.09
    èµĦæºIJéħįç½®
    -0.09
    æľĢç»Ī
    -0.08
    geries
    -0.08
    å®Įåħ¨åı¯ä»¥
    -0.08
    éĢļè¯Ŀ
    -0.08
    ä¸ĢæŶ
    -0.08
    POSITIVE LOGITS
    æľīç½ijåıĭ
    0.10
    ä¸įæĸŃåľ°
    0.10
    ä¸įæĸŃçļĦ
    0.10
    ä¸įåģľçļĦ
    0.10
    ä¼łæĿ¥
    0.10
    ä¸įåģľåľ°
    0.10
    æĥ³çĿĢ
    0.09
    å¾Ģä¸ĭ
    0.09
    æľīä¸Ģ次
    0.09
    åıįå¤į
    0.09
    Activations Density 0.014%

    No Known Activations