© 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. 10728
    Prev
    Next
    INDEX
    Explanations

    deception and misdirection

    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
    éļIJå½¢
    -0.11
    æķ´é¡¿
    -0.11
    CommandEvent
    -0.10
    å¡į
    -0.10
    æ£Ģ
    -0.10
    icates
    -0.10
     бÑĭ
    -0.10
    æ²ı
    -0.09
    åĿIJçĿĢ
    -0.09
    erea
    -0.09
    POSITIVE LOGITS
    è¿·æĥij
    0.11
    诱
    0.11
     deco
    0.11
    (dummy
    0.10
    åĪĨæķ£
    0.10
    åģĩ
    0.10
    å²Ķ
    0.10
    æĹłåħ³
    0.09
    åıĭ好
    0.09
    误导
    0.09
    Activations Density 0.089%

    No Known Activations