© 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. 53543
    Prev
    Next
    INDEX
    Explanations

    cake and eat it too

    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
    åIJĥçļĦ
    -0.10
    æĦŁ
    -0.10
    使ä¹ĭ
    -0.09
     Erie
    -0.09
    身价
    -0.09
    æĪij个人
    -0.09
     Brill
    -0.09
    纪念é¦Ĩ
    -0.09
    olec
    -0.08
    éĺ´è°ĭ
    -0.08
    POSITIVE LOGITS
    伦æķ¦
    0.11
    Colour
    0.09
     Colour
    0.09
     hoá
    0.09
    åĪ©çī©æµ¦
    0.09
     ÙģÙī
    0.09
    åŀĨ
    0.09
    æłı
    0.08
    colo
    0.08
     Southampton
    0.08
    Activations Density 1.682%

    No Known Activations