© 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. 57459
    Prev
    Next
    INDEX
    Explanations

    provided, available, or are

    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
    èł¡
    -0.10
    é»»
    -0.09
    éŃĶ
    -0.09
    lep
    -0.09
    太é«ĺ
    -0.09
    èĢģ乡
    -0.09
    .SizeType
    -0.09
    èĩªçIJĨ
    -0.09
    ç¥ī
    -0.09
    çķ¢ç«Ł
    -0.09
    POSITIVE LOGITS
    æĹ¢æĺ¯
    0.09
     arrayOf
    0.08
    eri
    0.08
    ros
    0.08
    hn
    0.08
     cou
    0.08
    ย
    0.08
    -nav
    0.08
    erts
    0.08
    ertz
    0.07
    Activations Density 0.325%

    No Known Activations