© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9378
    Prev
    Next
    INDEX
    Explanations

    killing people, injuries, or deaths

    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
    éĢ
    -0.10
    eneg
    -0.10
    Ware
    -0.10
    Õ®
    -0.09
    ANEL
    -0.09
    iphers
    -0.09
    æ®ĸ
    -0.08
    openh
    -0.08
    éĴİ
    -0.08
    ril
    -0.08
    POSITIVE LOGITS
    æĺ¨æĹ¥
    0.10
    æĺ¨æĻļ
    0.10
    è¿Ļ两天
    0.10
    人æµģ
    0.09
    æĺ¨å¤©
    0.09
    å±ĢåĬ¿
    0.09
    æĺ¨
    0.09
    æĬ¥éģĵ称
    0.09
    åĩºåĬ¨
    0.09
    èµĻ
    0.09
    Activations Density 0.080%

    No Known Activations