© 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. 4576
    Prev
    Next
    INDEX
    Explanations

    set up, set of

    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
    èĥ¡åIJĮ
    -0.09
    忽çĦ¶
    -0.09
    æŃĮ声
    -0.09
    è£Ĥ
    -0.09
    è¿ĺæĺ¯å¾Ī
    -0.09
    ãģ£ãģ
    -0.09
    è¶³å¤ŁçļĦ
    -0.09
    æıIJåīį
    -0.09
    é©Ń
    -0.09
    iling
    -0.08
    POSITIVE LOGITS
    oes
    0.09
    æŀķ
    0.09
    éĹ·
    0.08
    нÑĤ
    0.08
    èİĺ
    0.08
    à¹īà¸Ńย
    0.08
    rub
    0.08
    è¿Ļ个èģĮä¸ļ
    0.08
    enberg
    0.08
    elah
    0.08
    Activations Density 0.031%

    No Known Activations