© 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. 2112
    Prev
    Next
    INDEX
    Explanations

    content and material from sites

    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
    ction
    -0.10
    符
    -0.09
    交代
    -0.09
    麾
    -0.08
     Hardy
    -0.08
    èĩ´åij½
    -0.08
    ç»ıåİĨ
    -0.08
    ç«¥
    -0.08
    ctor
    -0.08
    ffer
    -0.08
    POSITIVE LOGITS
    ä¹ĭåŁİ
    0.11
    åĽ¾æĸĩ
    0.10
    æľªç»ıæİĪæĿĥ
    0.10
    æľ¬ç½ijç«Ļ
    0.10
    kur
    0.09
    æłĹ
    0.09
     Nap
    0.09
     Shower
    0.08
    _CAST
    0.08
    æĿ¥èĩªäºİ
    0.08
    Activations Density 0.102%

    No Known Activations