© 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. 33937
    Prev
    Next
    INDEX
    Explanations

    parts of

    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
    æ¡¢
    -0.09
    eron
    -0.09
    ron
    -0.09
    ering
    -0.09
    rum
    -0.09
    Crime
    -0.08
    好åĿı
    -0.08
    åıĺåĮĸ
    -0.08
    æ¶Ī
    -0.08
    orsk
    -0.08
    POSITIVE LOGITS
    éĺŁä¼į
    0.10
    æĿ¥åģļ
    0.09
    身åŃIJ
    0.09
    æĪĸå¤ļ
    0.09
    çļĦ带é¢Ĩ
    0.08
    裤
    0.08
     ven
    0.08
    亲
    0.08
    ä¹ĥèĩ³
    0.08
    èĪ°éĺŁ
    0.08
    Activations Density 0.055%

    No Known Activations