© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52829
    Prev
    Next
    INDEX
    Explanations

    WARRANTIES or representations

    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
    æľīæĦı
    -0.10
    åħ¥åѦ
    -0.09
    ju
    -0.08
    æ±Łæ·®
    -0.08
    enuine
    -0.08
     Appeals
    -0.08
     Loy
    -0.08
    çļĦæľºä¼ļ
    -0.08
     PD
    -0.08
    空
    -0.08
    POSITIVE LOGITS
    æij©
    0.10
    å¾Ĵ
    0.10
    å®ł
    0.09
    DRV
    0.09
    èĤĵ
    0.09
    çŀ½
    0.09
    åıij表äºĨ
    0.09
    å©¿
    0.08
    æĴ¸
    0.08
     ÙĨÙĬÙĪز
    0.08
    Activations Density 0.011%

    No Known Activations