© 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. 7986
    Prev
    Next
    INDEX
    Explanations

    bright/vibrant colors

    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
    å·²ä¹ħçļĦ
    -0.10
    ä¸ĵåijĺ
    -0.09
    ideo
    -0.09
    à¤Ŀ
    -0.09
    abler
    -0.09
     Ø¢ÙĦ
    -0.09
    нÑĤ
    -0.09
    _cats
    -0.09
    èĶº
    -0.09
    ä¸ĵçıŃ
    -0.09
    POSITIVE LOGITS
    èī³
    0.11
    é²ľèī³
    0.11
     bright
    0.10
     vibrant
    0.10
    èĢĢçľ¼
    0.09
     vibr
    0.09
     brightly
    0.09
     colors
    0.08
     colorful
    0.08
     Polly
    0.08
    Activations Density 0.168%

    No Known Activations