© 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. 63096
    Prev
    Next
    INDEX
    Explanations

    but, however, 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
    åİŁæłĩé¢ĺ
    -0.09
    表éĿ¢ä¸Ĭ
    -0.09
     Bris
    -0.09
    ameron
    -0.09
     dok
    -0.08
    Kir
    -0.08
     Та
    -0.08
    鸢
    -0.08
     gradu
    -0.08
    çļĦ人çĶŁ
    -0.08
    POSITIVE LOGITS
    ä¸į该
    0.13
    ä¸įåºĶ该
    0.13
    ä¸įåºĶ
    0.10
    ä¸įèĩ³äºİ
    0.10
    飧
    0.09
    å°±åºĶ该
    0.09
    ä¹ŁåºĶ该
    0.09
    èĭŁ
    0.09
    çIJĨ
    0.09
    å°Ĩèĩªå·±çļĦ
    0.08
    Activations Density 0.055%

    No Known Activations