© 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. 63691
    Prev
    Next
    INDEX
    Explanations

    sister site/station

    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
    /devices
    -0.11
    缸åħ³äººåijĺ
    -0.08
    amar
    -0.08
    èĦijæµ·ä¸Ń
    -0.08
    èĦijæµ·
    -0.08
    路桥
    -0.08
    çļĦåIJĮåѦ
    -0.08
    å¸Ĥä¸Ńå¿ĥ
    -0.08
    /Linux
    -0.08
    ä¸ĢéĹ´
    -0.08
    POSITIVE LOGITS
    ä¹ĭä½ľ
    0.12
    /general
    0.10
    nant
    0.09
    iest
    0.09
    èĪ¬çļĦ
    0.09
    lösung
    0.09
    人çī©
    0.09
     Wife
    0.09
    /default
    0.08
    èĤ¥æĸĻ
    0.08
    Activations Density 0.477%

    No Known Activations