© 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. 34085
    Prev
    Next
    INDEX
    Explanations

    convict criminals felon

    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
    æĸ°ä¸ī
    -0.10
    éģĩ
    -0.10
    åºĻ
    -0.09
    åĻİ
    -0.09
    éħ¬
    -0.09
    å»Ł
    -0.09
    инг
    -0.09
    æ´Ħ
    -0.08
    ä¿Ŀåģ¥
    -0.08
     caffe
    -0.08
    POSITIVE LOGITS
    çĬ¯ç½ª
    0.11
    æ¶Ĥ
    0.10
    åĽ¢ä¼Ļ
    0.10
    åĪĨåŃIJ
    0.10
    è¿Ŀæ³ķçĬ¯ç½ª
    0.10
    ä½ľæ¡Ī
    0.09
    æĸĹ
    0.09
    罪
    0.09
     criminals
    0.09
    AtPath
    0.09
    Activations Density 0.072%

    No Known Activations