© 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. 57700
    Prev
    Next
    INDEX
    Explanations

    it's the

    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
    uld
    -0.10
    aux
    -0.10
    家ä¼Ļ
    -0.09
    çªģçł´åı£
    -0.09
    ä¸įåłª
    -0.09
    ä¸ĢåĪ»
    -0.09
    urs
    -0.09
    ä¸Ģæĺ¯
    -0.09
    åıĹçĽĬ
    -0.09
    ä¸įåĨįæĺ¯
    -0.08
    POSITIVE LOGITS
    PathParam
    0.09
    ocratic
    0.09
    Grün
    0.08
    åĩłå¼ł
    0.08
    åħ¸åŀĭçļĦ
    0.08
    第äºĮç§į
    0.08
    (mContext
    0.08
    emann
    0.08
     Wallpaper
    0.08
    追éĹ®
    0.08
    Activations Density 0.030%

    No Known Activations