© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50942
    Prev
    Next
    INDEX
    Explanations

    guilty plea or convicted

    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
    æŃ¤åŁºç¡Ģä¸Ĭ
    -0.09
    hil
    -0.09
    ми
    -0.09
    pu
    -0.09
    åijĺ
    -0.09
    Ñıн
    -0.08
    鸥
    -0.08
    _mut
    -0.08
    .named
    -0.08
    åįİ丽
    -0.08
    POSITIVE LOGITS
    ä¸İåIJ¦
    0.12
    è¿Ļæł·çļĦè¯Ŀ
    0.10
    çģĮ
    0.09
    飬
    0.09
    orse
    0.09
    ess
    0.09
    ä¸Ģäºĭ
    0.09
    esses
    0.08
    -looking
    0.08
    姿æĢģ
    0.08
    Activations Density 0.005%

    No Known Activations