© 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. 2172
    Prev
    Next
    INDEX
    Explanations

    medical, technical, and industrial terms

    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
    iced
    -0.08
    æ®ĸ
    -0.08
     Ult
    -0.08
    رس
    -0.08
    itim
    -0.08
    dish
    -0.07
    åĮ¡
    -0.07
    ccount
    -0.07
    çĶŃ
    -0.07
     ÑĢеб
    -0.07
    POSITIVE LOGITS
    ä¸įåĪ©äºİ
    0.10
    åıĪ称
    0.09
    æĪĸå¤ļ
    0.09
    模å¼ı
    0.09
    åĪ©äºİ
    0.09
    iller
    0.09
     Lans
    0.09
    ä¸įéĢĤåIJĪ
    0.09
     зан
    0.09
    è¿Ļç§įæĸ¹å¼ı
    0.08
    Activations Density 0.870%

    No Known Activations