© 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. 53777
    Prev
    Next
    INDEX
    Explanations

    repeating values or elements

    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
    æīij
    -0.11
    æĮĩ导æĦıè§ģ
    -0.10
    æ»´
    -0.10
    çķ¿
    -0.10
    çıŃåŃIJ
    -0.09
    ä¸Ģæĸ¹éĿ¢
    -0.09
    åĨįä¹Ł
    -0.09
    ë¦Ħ
    -0.09
    inha
    -0.09
    ç°ª
    -0.09
    POSITIVE LOGITS
    gis
    0.08
     pres
    0.08
    Serialized
    0.08
     self
    0.08
    usc
    0.08
    engan
    0.08
    顺çĿĢ
    0.08
     flakes
    0.07
     sent
    0.07
    说èĩªå·±
    0.07
    Activations Density 0.001%

    No Known Activations