© 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. 56228
    Prev
    Next
    INDEX
    Explanations

    at

    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
    æĸ°ä¸Ģ代
    -0.10
    arih
    -0.10
    第ä¸ī代
    -0.09
    大海
    -0.09
    èĹı
    -0.09
    ution
    -0.08
    rok
    -0.08
    odate
    -0.08
    åIJµ
    -0.08
    iei
    -0.08
    POSITIVE LOGITS
    潸
    0.09
    ç»Ŀä½³
    0.09
    çħĮ
    0.09
    æĻı
    0.08
    æĢİ
    0.08
    æıī
    0.08
    è¿Ļä¸ī个
    0.08
    кÑĢÑĥÑĤ
    0.08
    çļĦä¸ĢåĪĩ
    0.08
    马åħĭæĢĿ
    0.08
    Activations Density 0.066%

    No Known Activations