© 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. 35443
    Prev
    Next
    INDEX
    Explanations

    color transformation and change

    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
    è¿Ł
    -0.09
    åįĬ天
    -0.08
    /source
    -0.08
    ÙĪÙĨÙĬ
    -0.08
     Ferr
    -0.08
     Buffered
    -0.08
    oscope
    -0.08
     unc
    -0.08
    ×Ļ×ļ
    -0.08
    ande
    -0.08
    POSITIVE LOGITS
    -addons
    0.10
    çĴĢ
    0.08
    ultan
    0.08
    çħĮ
    0.08
    å¸Ŀ
    0.08
    mpi
    0.08
    çĤİ
    0.08
    â¾Ĭ
    0.08
    æľīåĩłä¸ª
    0.08
    并ä¸įæĦıåij³
    0.08
    Activations Density 0.002%

    No Known Activations