© 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. 26813
    Prev
    Next
    INDEX
    Explanations

    responsible for, derives, does

    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
    uzzi
    -0.10
    å¦ŀ
    -0.09
    æĪijè¿ĺæĺ¯
    -0.09
    andler
    -0.09
    笪
    -0.08
    ãģĤãĤĮ
    -0.08
    exterity
    -0.08
    ex
    -0.08
    æĪijå¦Ī
    -0.08
    epend
    -0.08
    POSITIVE LOGITS
    æĸ°çļĦ
    0.08
    临
    0.08
    vir
    0.08
    å½ĵåľ°çļĦ
    0.08
     conte
    0.08
    èıł
    0.08
    足
    0.08
    æħµ
    0.08
     Depart
    0.07
    anium
    0.07
    Activations Density 0.111%

    No Known Activations