© 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. 9437
    Prev
    Next
    INDEX
    Explanations

    varied, diverse, variable

    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
    iams
    -0.11
    绿水
    -0.11
    æĹłå£°
    -0.10
     Said
    -0.10
    åħ´æĹº
    -0.09
     Ùħض
    -0.09
    æĹłäºº
    -0.09
     Steph
    -0.09
    éĩįè¦ģæĦıä¹ī
    -0.09
    æį¢åıĸ
    -0.09
    POSITIVE LOGITS
    iveness
    0.11
    æ´İ
    0.08
    åı£åij³
    0.08
    è§ĴèIJ½
    0.08
    èĮĥåĽ´
    0.08
    æĪª
    0.08
     diverse
    0.08
    迳
    0.08
    ç«Ł
    0.08
    跨度
    0.08
    Activations Density 0.136%

    No Known Activations