© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20707
    Prev
    Next
    INDEX
    Explanations

    Waterloo and place names

    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
    avig
    -0.11
    å®ĿçŁ³
    -0.09
    atura
    -0.09
    马
    -0.09
    STALL
    -0.09
    umer
    -0.08
    ives
    -0.08
    äs
    -0.08
    orage
    -0.08
    elves
    -0.08
    POSITIVE LOGITS
    onia
    0.10
    ensis
    0.09
    åIJ»
    0.09
    Ế
    0.09
    imat
    0.09
    ä¸Ģ天
    0.09
    ä¸Ģæĸ¹
    0.09
    -largest
    0.08
    ãģĦãģı
    0.08
    é¢ĺæĿIJ
    0.08
    Activations Density 0.002%

    No Known Activations