© 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. 26831
    Prev
    Next
    INDEX
    Explanations

    questions asking how, what, or why

    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
    adia
    -0.10
    ä¸įä¸Ĭ
    -0.09
    ä¸ĢæĬĬ
    -0.09
    éĥ¨åĪĨ
    -0.09
    åħ¸èĮĥ
    -0.09
    åĽ°
    -0.09
    ä½łæĥ³
    -0.08
    pie
    -0.08
     Morrow
    -0.08
    éĥ¨åĪĨåĨħ容
    -0.08
    POSITIVE LOGITS
    è¿Ļæł·çļĦè¯Ŀ
    0.10
    好åIJĹ
    0.09
    åı¯æĥľ
    0.08
    èµIJ
    0.08
    è¿Ļä¹Īåģļ
    0.08
    è¿Ļæĺ¯
    0.08
    ERN
    0.08
    说çļĦæĺ¯
    0.08
    æĪĸåħ¶ä»ĸ
    0.08
    è¿Ļæł·æīįèĥ½
    0.08
    Activations Density 0.237%

    No Known Activations